Metal Detection Guide en

The Metal Detection Guide
Building an Effective Programme

No part of this manual may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying and recording,
for any purpose without the express written permission of Mettler-Toledo Safeline Limited.
This documentation is furnished with restricted rights.
The information contained in this manual is provided to assist producers in developing and implementing an effective metal detection programme.
Mettler-Toledo Safeline Limited does not warrant the accuracy or applicability of the information contained herein and as such is specifically not responsible
for property damage and/or personal injury, direct or indirect for damages and/or failures which may be caused as a result of using the information provided.
Contents Page
Introduction 2
Introduction to Metal Detection 6
Key Design Features 12
Factors Limiting Sensitivity 18
Systems Design & Applications 26
Metal Detection, X-ray Inspection or Both Technologies? 36
Reasons For a Metal Detection Programme 42
Building an Effective Programme 46
Prevention of Metal Contamination 48
Selecting Control Points 52
Operating Sensitivity 56
Installation and Commissioning 60
Performance Validation, Verification and Monitoring 64
Dealing With Suspect and Rejected Product 76
Data Analysis and Programme Improvement 80
Data, Connectivity and Improving Performance 84
Calculating Total Cost of Ownership for In-Line Metal Detection Equipment 90
The Principles of Due Diligence For Quality Control and Legal Defence 100
Understanding Challenging Applications for Improved Metal Detection 106
Explosion Protection for Metal Detectors 112
© 2016 Mettler-Toledo Safeline Ltd

1
Introduction
Most manufacturers and processors in the food and pharmaceutical industries recognise the need for metal detection
systems and that they are an essential element of any efficient and effective quality-focused regime. There is an ever
greater need for metal detectors in an increasingly competitive market-place driven by a growing range of complex
factors including changing customer requirements, the tightening of industry standards, and the growth of regulatory
and legislative bodies.
However, merely installing metal detectors as stand-alone

equipment will not necessarily guarantee the manufacture
of products that are free of metal contaminants; metal
detector installations need to form part of an effective
overall metal detection programme. This updated guide has
been rewritten to include the latest changes in technology,
industry trends and standards, legislation and regulatory
requirements (SLR’s).
An effective metal detection programme can, of course,
provide protection against product failure and recalls due
to metal contamination. More than that though, an effective
metal detection programme can help to maintain hard-won
supplier certification status, whilst also reducing overall
operating costs.
In the event of a legal claim, an effective metal detection
programme can also support proof that reasonable
precautions and due diligence have been applied in the
manufacturing process.
The requirements for an effective metal detection
programme, as well as the benefits of adopting such a
programme, are summarised in diagram 1a.

2
External Inputs
Customer/Industry Consortia Industry Standards Regulatory Authorities
Retail Consortia e.g. GFSI International Standards e.g. Food Safety Legislation
standards BRC, IFS ISO 22000 USDA
Retailer Codes Industry Standards e.g. HACCP FDA, FSMA, ANVISA
Brand Consumer Codes SQF 1000/2000 Code Certification Bodies
GMP
Effective Metal Detection Programme
Metal Detection System Selection

1. Introduction to Metal Detection
2. Key Design Features
3. Factors Limiting Sensitivity
4. Systems Design & Applications
5. Metal Detection, X-ray Inspection or Both Technologies?

6. Reasons For a Metal Detection Programme
7. Building an Effective Programme
8. Prevention of Metal Contamination
9. Selecting Control Points
10. Operating Sensitivity
11. Installation and Commissioning
12. Performance Validation, Verification and Monitoring
13. Dealing With Suspect and Rejected Product
14. Data Analysis and Programme Improvement
15. Data, Connectivity and Improving Performance
16. Calculating Total Cost of Ownership for In-Line Metal Detection Equipment
17. The Principles of Due Diligence For Quality Control and Legal Defence
18. Understanding Challenging Applications for Improved Metal Detection
19. Explosion Protection for Metal Detectors
Producer, Retailer and Consumer Benefits
Reasonable Precautions Reduced Failure Costs Protection Against Product Maintain

Recall/Claims Certification Status
Due Diligence Retailer Brand Maintain Approved

Confidence Protection Supplier Status
Table 1a: Effective metal detection programme

3
Introduction
This guide provides a comprehensive reference source for

those involved in food safety, and provides an insight into all
aspects of metal detection, from basic principles through to
implementing a comprehensive metal detection programme.
Chapters 1 to 4 provide an overview of how metal detectors
work, together with an explanation of important design
features. These chapters also offer an insight into factors
which potentially limit performance of the equipment. There
is also a description of the integration of metal detectors
with effective rejection systems.
Chapter 5 gives consideration to how both metal detection
and x-ray inspection systems could be used separately
or together to provide the maximum protection against an
increased range of possible contamination materials.
Chapters 6 to 15 demonstrate that simply installing a
metal detection system alone will not necessarily provide
adequate levels of protection against metal contamination.
For proper and thorough protection, a comprehensive metal
detection programme is required – and the key elements of
such a programme are explained here in detail.
Chapter 16 to 19 offer practical guidance including some
thoughts on how to calculate the total cost of in-line metal
detection ownership and a description of how through
better system due diligence food industry standards can
be exceeded. The penultimate chapter considers how to
overcome challenging food applications using the latest
in-line metal detection systems. The final chapter explains
explosion protection.

4
Introduction
Notes

5
1
Introduction to Metal Detection

This chapter provides a broad overview of components and operating principles, so as to communicate an overall
understanding of metal detection systems.
The information in this chapter can then be built upon, based on the information in subsequent chapters - and this
should lead to a clear understanding of metal detection technology, equipment capabilities and performance.
1
Introduction to Metal Detection 1.1 Sources of Metal Contamination
Metal contamination sources are numerous – and even
1.1 Sources of Metal Contamination the most stringent controls cannot prevent the occasional
incident in which small pieces of metal find their way into
1.2 What is a Metal Detection System? products destined for consumer consumption.
1.3 Where Can a Metal Detection System Be Used? Good working practices will minimise the likelihood of metal
1.4 Balanced Coil System contaminants entering the production flow; furthermore,
correct equipment design and appropriate selection will
1.5
Ferrous-in-Foil (FIF) Detection maximise the likelihood of reliably detecting and rejecting
1.6
Detection Modes any metal particles that have found their way into products.
1.7 W
hy Should you Choose the Correct Metal Detector? Contamination normally comes from the following sources:
• Raw Materials
Typical examples include metal tags and lead shot in
meat; wire in wheat; screen wire in powder material;
tractor parts in vegetables; hooks in fish; staples; wire
strapping from material containers.
• Personal Effects
Buttons; pens; jewellery; coins; keys; hair-clips; thumb-
tacks; pins; paper clips, etc.
• Maintenance
Screwdrivers and similar tools; swarf and welding slag
(following repairs); copper wire off-cuts (following
electrical repairs); miscellaneous items resulting from
inefficient clean-up or carelessness; metal shavings from
pipe repair.
• In-plant Processing
The danger of contamination exists every time the
product is handled or passes through a process.
Crushers, mixers, blenders, slicers and transport systems
can all act as sources of metal contamination. Examples
of metal contamination from these sources include
broken screens, metal slivers from milling machines, and
foil from reclaimed products.
Identifying the likely source of contamination is a vitally
important stage in developing a successful overall metal
detection programme.

6
1.2 What is a Metal Detection 4. Automatic Rejection System
An automatic reject device is frequently fitted to the transport
System? system in order to remove any contaminated product from
An industrial metal detection system is a sophisticated the production line. There are many different styles available
piece of equipment used to detect and reject unwanted including ‘air blasts’, ‘push arms’, ‘drop flaps’, etc. The style of
metal contamination. When properly installed and operated, the reject device installed will depend on the type of product
it helps to reduce metal contamination and improve food being inspected (Refer to Chapter 4 for further information).
safety. A typical metal detection system consists of four In addition to the four main parts of a metal detection
main parts: system, other important items may include:
1. Detector Coil or ‘Detector Head’ • A lockable container fixed to the side of the conveyor, to
Most modern metal detectors fall into one of two main collect and hold rejected product
categories, with respect to the ‘Detector Head’, which is that • A full-length cover between detector and reject device
part of the metal detector system that identifies the presence • A fail safe alarm which operates if the metal detector
of metal contamination: develops a fault
• A reject confirmation device, with sensors and timers,
• The first type of metal detector utilises a ‘balanced coil’
to confirm that contaminated product is actually rejected
Detector Head. Detectors of this design are capable
from the line
of detecting all metal contaminant types, including
• A beacon and/or audible alarm that alerts operators to
ferrous, non-ferrous and stainless steels, in fresh and
various other events, such as an automated warning that
frozen products. The products being inspected can be
a detector is due to be tested or that the reject bin is full
either unwrapped or wrapped, and can include products
• Numerous optional fail safe systems to raise level of
wrapped in metallised films.
“Due Diligence”
• The second detector type utilises permanent magnets in
a ‘Ferrous-In-Foil’ Detector Head. These Detector Heads
are capable of detecting ferrous metals and magnetic 1.3 Where Can a Metal Detection
stainless steels only within fresh or frozen products
which are packed in an aluminium foil wrapping. System Be Used?
Whilst it is recognised that other technologies exist, this guide Metal detectors may be used at various stages of a
concentrates mainly on the ‘balanced coil’ detector type − and production process:
(to a much lesser extent) on Ferrous-In-Foil (FIF) technologies. 1. Bulk ‘In-Process’ Inspection
Detector Heads can be manufactured in virtually any size, • Eliminates metal before it can be broken into
in order to suit the product being inspected. They may be smaller pieces
rectangular or round, and may be mounted horizontally, • Protects processing machinery from damage
vertically or on an incline. • Avoids product and packaging waste by subsequently
rejecting a finished higher-value product
Each Detector Head has an opening (known as an
‘aperture’) through which product passes. When a metal Typical examples include bulk inspection of meat blocks prior
contaminant is detected by the detector Detector Head, a to grinding, ingredients for pizza toppings and grain products.
signal is sent to the electronic control system. 2. Finished Product Inspection
2. User Interface/Control Panel • No danger of subsequent contamination
The user interface is the front-end of the electronic control • Ensures compliance with retailer and consumer brand
system, and is often mounted directly on the Detector quality standards
Head. However, the user interface can be mounted remotely A combination of bulk and finished product inspection will
(with connecting cables) if the Detector Head is too small, provide optimum protection.
or if the Detector Head is installed in an inconvenient or
inaccessible location. The most common types of metallic contamination include:
• Ferrous (iron)
3. Transport System • Non-ferrous (brass, copper, aluminium, lead)
The transport system is used to pass the product to be • Various types of stainless steel (magnetic and
inspected through the aperture of the metal detector. The non-magnetic
most common type of transport system is a conveyor.
Of the three types listed above, ferrous metal is generally
Alternatives include:
the easiest to detect − and relatively simple detectors (or
• A plastic chute with the detector mounted on an incline even magnetic separators) can perform this task well.
• A non-metallic pipe, mounted horizontally or vertically.
This type of transport system is commonly used in the Stainless steel alloys are extensively used in the food
inspection of powders and liquids industry, but are often the most difficult to detect, especially
common non-magnetic grades such as 316 and 304.

7
1
Non-ferrous metals, such as brass, copper, aluminium and

lead, usually fall between these two extremes, although
in larger metal detectors operated at higher frequencies,
non-ferrous metal may be harder to find than non-magnetic
stainless steel.
Only metal detectors using an alternating current ‘balanced
coil’ system have the capability to detect small particles of
Figure 1.2
non-ferrous and non-magnetic stainless steel.
The control electronics actually split the received signal into

1.4 Balanced Coil System two separate components; these are known as ‘magnetic’
and ‘conductive’, and are at 90º to each other.
1.4.1 Basic Principles of Operation
To prevent airborne electrical signals, or to prevent nearby
Three coils are wound onto a non-metallic frame or ‘former’, metal items and machinery from disturbing the detector,
and each coil is exactly parallel with the other two (Figure 1.1). the complete coil arrangement is mounted inside a metal
The centre coil (the ‘transmitter’) is energised with a high- case. This has an opening (‘aperture’) to allow passage
frequency electric current that generates a magnetic field. of the product.
The two coils on each side of the centre coil act as receivers. The case can be constructed of aluminium or stainless steel
Since these two coils are identical and are the same distance (depending on application). In addition to creating a screen,
from the transmitter, an identical voltage is induced in each. the metal case adds strength and rigidity to the assembly.
When the coils are connected in opposition, these voltages This is crucial for satisfactory operation of the detector.
cancel out, resulting in ‘zero output’.
Several special mechanical and electrical techniques are
essential to the design of stable, reliable metal detectors.
On-line production stability overtime is a key measure
of a metal detector’s performance when comparing one
suppliers offering to another.
Encircling Coil
Encircling Coil
(Cut-away)
(Cut-away)
1.4.2 Mechanical Techniques
The metal case affects the magnetic field’s state of balance,
and any movement relative to the coils can cause a false
detection signal. In addition, microscopic movements of the
coils relative to each other (as small as 1 micron), can cause
Transmitter
Transmitter a signal sufficient to result in a false rejection.
One of the major design problems for metal detector
Receiver
Receiver Receiver
Receiver
manufacturers is to devise a totally rigid and stable system,
unaffected by vibration from motors, pulleys, auto-
Product
Product Flow
Flow
reject devices, temperature changes, transportation and
machinery located in close proximity.
It is vitally important to select the right kind of material
for detector head coil specifications and case design are
equally crucial. To increase mechanical rigidity further,
Output
Output most manufacturers fill the detector case with a material
Figure 1.1 to prevent relative movement of the metal case to the coils
(frequently referred to as ‘potting’).
As a particle of metal passes through the coil arrangement, This helps produce a unit that is able to operate at maximum
the high-frequency field is disturbed at the 1st receiver coil sensitivity under typical factory conditions. The quality of the
(point A) and then at the 2nd receiver coil (point B). This action potting is critical to the performance of the metal detector.
changes the voltage generated in each receiver (though by
a matter of only 1x10-9 nano-volts). Despite the very small
change in voltage, this alteration in balance generates a signal
that can be processed, amplified and subsequently used to
detect the presence of unwanted metal (Figure 1.2).

8
1
1.4.3 Electronic Techniques MFZ

MFZ
Mechanical construction methods will minimise false

signals from coil and case movements; they will also
provide long-term stability in harsh environments. However,
various factors can contribute to an out-of-balance voltage.
These include:
MFZ
MFZ
• Temperature changes MFZ
MFZ MFZ
MFZ
• Metal close to the aperture Figure 1.3

• Ageing of electronic components
• Slow changes in the mechanical structure
The MFZ will normally be specified within the manufacturer’s
These types of factors that create out-of-balance voltage installation instructions. Typical quoted values are in the
can be eliminated by various electronic techniques. region of 1.5 x aperture height for stationary metal and 2.0
Automatic Balance Control continuously monitors out- x aperture height for moving metal. Due consideration of
of-balance voltage and automatically corrects it. This this during installation will provide consistent, reliable metal
removes the need for periodic fine-tuning by an operator, detection performance.
and ensures that the detector is always operating at the
optimum level. Space may be restricted, as where a short conveyor
system is used, or when installation is between a weighing
Quartz Crystal Control, now standard on most metal machine and a vertical form-fill seal bag maker. Under
detectors, accurately controls the frequency of the oscillator these circumstances, a special unit may need to be used
in order to prevent ‘drift‘. However, further electronic if the metal-free zone is unduly small. This is referred to as
compensation is necessary to combat electronic component ‘Zero Metal-Free Zone (ZMFZ) technology. Alternatively there
changes which occur with changes in temperature. are some mechanical techniques, such as the additional of
Automatic Balance Control and Quartz Crystal Control extended flanges that also attempt to control the leakage of
will not, themselves, enable the detector to detect smaller the magnetic field.
pieces of metal. They will, however, enable the detector to
permanently maintain this level of sensitivity without operator 1.5 Ferrous-in-Foil (FIF) Detection
attention and without the generation of false reject signals.
When the product to be inspected is packaged inside
Temperature compensation control circuitry automation an aluminium foil pack or dish, a metal detector using a
manages the effects of external temperatures variation that balanced coil system cannot be used. However, there is an
may effect the performance of the detector by eliminating available detector design which suppresses the effect of
balance drift. the aluminium foil, but is still sensitive to small pieces of
For high performance over an extended period, Automatic ferrous and magnetic stainless steel contamination. Figures
Balance Control, Quartz Control, temperature compensation 1.4 and 1.5 show the basic operating principle.
and ‘potting’ of Detector Heads are all essential.
Magnets
Magnets
1.4.4 Metal Free Zone (MFZ)

Most of the detector’s high-frequency magnetic field is
contained within the metal case of the detector unit. There
is some unavoidable leakage of the magnetic field from
the aperture of the detector – and it is the effect of this
leakage on the magnetic field of the surrounding metalwork
that may influence the detector’s performance, resulting in
inconsistencies in detection capability.
Magnets
Magnets
To achieve optimum metal detection results, the aperture of Figure 1.4 Ferrous
Ferrousinin Foil detector
Foil detector
the detector should be surrounded by an area known as the

‘Metal-Free Zone’ or ‘MFZ’; this area should be kept free of Output
Output
all metals.
The size of the MFZ is dependent upon: Magnets
• The aperture size (Figure 1.3)
• The type of detector Product
Product Flow
Flow
• The operating sensitivity

Figure 1.5 Magnets
Stationary metal can be positioned closer to the detector
than moving metal.

9
1
As a metal particle approaches the detector, it moves into a 1.6.1 Amplitude Detection
powerful magnetic field that magnetises the particle. As this
magnetised particle passes through the single coil (which is When the signal from the metal particle exceeds a
wound around the former), a small voltage is generated and predetermined ‘trigger’ level, the detector operates. Figure
subsequently amplified. 1.6 shows how a large piece of metal breaks the trigger level
and so is detected earlier than a small piece of metal. With
Ferrous-in-Foil metal detectors show much greater Amplitude Detection, a large metal piece is detected earlier,
sensitivity to magnetic material than to non-magnetic and so a greater amount of ‘good’ product is rejected.
material, but in practice, the sensitivity of the detector may
have to be reduced due to some product signal from the
aluminium foil. These kinds of operating conditions often 1.6.2 Zero Crossover Detection
impose a limit on efficient performance. This method gives a ‘detect’ signal from the metal when
The limitations of this technology are clear and unless the signal changes polarity – from a positive to a negative
(through HACCP findings) the only metallic contamination or vice-versa. Figure 1.6 shows that this always occurs
likely to be present is ferrous (or magnetic) it is recommended at the same point (under the centre coil) independent of
that other technology is investigated i.e. X-ray. metal size. With this method, the point of detection can be
accurately determined, regardless of metal size, and the
volume of rejected product can therefore be minimised.
1.6 Detection Modes
As a metal particle passes through a balanced coil 1.6.3 Multiple Metal Pieces
detector, an output signal is generated which increases to The major drawback of the Zero Crossover method is that it
a maximum as it passes under the first coil. It then falls to is not fool-proof. In a typical production line, it is common
zero as it reaches the centre coil, and increases again to a for no contamination to occur for a long period − and then
maximum as it passes under the third coil. several pieces can find their way into product at the same
The signal will start to build up when the metal contaminant time, as when a sieve or mincer breaks up.
is some distance from the coil – and with a large piece of If one metal piece follows a second piece and the metals
metal contaminant, it could be influencing the coil before are of a different size, then the Zero Crossover detector may
it even arrives at the detector. Figure 1.6 shows the signal not detect the smaller piece.
generated by a small and a large piece of metal. This will
be true for all types of detectors. Figure 1.7 shows the signal from a small piece of metal
(‘A’), followed by a larger piece, (‘B’). The detector does not
see the two separate signals, but it does see the combined
signal (‘C’) formed by the signals of both metal piece A and
metal piece B coming together as one.
Before signal C has a chance to change polarity (and
therefore be detected), it is overpowered by the effects of
Large particle
Large particle Small particle
Small particle the second piece of metal − and as a result, the first piece
detected here
detected here detected here
detected here
of metal is not detected.
If a third large piece of metal arrives, the first two pieces of
Trigger
Trigger metal may not be detected − and so on. This is a serious
Level
Level
limitation of the Zero Crossover method.
Amplitude
Amplitude
Detection
Detection
ZeroCross
Zero Cross
Bothparticles
Both particles
detected
detected here
here
Small particle
Small particle A
Figure 1.6
There are, however, two different methods of interpreting Large particle

Large particle B
or processing this output signal, which results in different
detector characteristics; one is known as ‘Amplitude
Detection’, the other is known as ‘Zero Crossover’ (or Resulting Signal
Resulting Signal C
C
‘Narrow Zone’) Detection.
Figure 1.7

10
1
1.6.4 Inverse Detection 1.7.3 Productivity Improvements

Some packaged products deliberately include a metal object Research suggests that on average, plants waste between
as part of the packaging or part of the product itself. This 28% up to 40% of their capacity through stops, speed
might be a specific component or a free gift – but whatever losses, interruptions and defects. Choosing a metal detector
its form or function, metal detectors can also be used to that is simple to set up and operate, offers reliable and
verify that such a ‘required‘ metal object is present in a consistent performance, with low maintenance requirements
packaged product. and easy-clean system design can enable productivity to be
optimised by ensuring costly downtime is all but eliminated.
This is usually achieved by reversing the action of the reject
timer so that product containing no metal is rejected, whilst Overall Equipment Effectiveness (OEE) is an important tool
product containing metal is accepted. in the pharmaceutical, packaging and food processing
industries. In any capital intensive business OEE
With this type of application, it is important to monitor the
improvement is a critical methodology to drive improved
product both before and after the time at which the metal
efficiency, higher quality and reduced cost.
item is introduced. This will ascertain that the detected
metal item at the output point is the ‘required‘ object and Adopting OEE methodology can provide benefits in the
not a metal contaminant. following areas:
• Equipment: Reduced equipment down-time and
1.7 Why Should you Choose the maintenance costs, and better management of the
equipment life cycle
Correct Metal Detector? • Personnel: Labour efficiencies and increased
productivity by improving visibility into operations
1.7.1 Compliance and empowering operators
When installed at critical control points (CCP’s) in your • Process: Increased productivity by identifying bottlenecks
processes, metal detectors enable your business to comply • Quality: Increased rate of quality and reduced scrap.
with the requirements of Hazard Analysis and Critical A number of software tools exist to capture manufacturing
Control Points (HACCP) and the broader needs of food performance data and display OEE performance graphically.
safety regulations and standards. (Refer to Chapter 9 for Selection of an appropriate OEE software tool is critical to
further information). the success of any OEE initiative. A mistake to be avoided
However, it is not enough to simply install a metal detector is the belief that this tool will drive OEE improvement −
at critical control points. Regular testing of the performance remember that any OEE software application is just a tool,
of metal detection equipment is an essential part of any and if not harnessed correctly will merely measure OEE, not
well designed quality management system in the food and improve it. Refer to Chapter 15 for more detailed information
pharmaceutical industry. regarding measuring and calculating OEE, and guidance on
how it can be improved.
1.7.2 Cost Reduction
1.7.4 Improved Competitiveness
Selecting the most stable and reliable metal detector, and
installing it at the most appropriate point(s) within the Ultimately, by complying with regulatory, industry and
manufacturing process enables overall lifetime costs to be retailer standards, reducing your costs and increasingly your
managed and kept to an absolute minimum through: productivity you can improve your competitiveness. This in
turn enables you to protect your brand and your reputation
• Eliminating false rejects and product waste
and be in a better position to win more customers.
• Reducing performance testing requirements

11
2
Key Design Features

In the event of a metal detector failure at a manufacturing facility, the manufacturer may well face a dilemma:
production must either halt until a service engineer visit can be scheduled - or the production line can continue to run,
but with the risk of metal contamination going undetected.
The chances of having to face this dilemma can be significantly reduced by choosing the most reliable metal detection
system available. This chapter provides valuable information on some of the key considerations and design features to
bear in mind when selecting the best metal detection system for a particular manufacturing environment.
2.
Key Design Features Selecting a reliable metal detection system is a major step
towards minimising (or even eliminating) the occurrence of
metal contamination within products being manufactured
2.1 Detector Electronics Design on a production line. However, despite the widespread use
of metal detectors, there are few guidelines available to help
2.2 Detector Mechanical Design
users evaluate the advantages of a particular detector or
2.3 Conveyor System Design compare the capabilities of different brands of machines.
2.4 Non-Conveyorised System Design Drift, erratic detection, complexity of set-up, and random
false rejects are the types of metal detector faults that can
2.5 Reject Mechanism Design
have major negative impacts on the overall success of a
2.6 Hygienic Design metal detection programme. And of course, when metal
2.7 Health and Safety detectors are subject to such faults, metal detectors can be
frustrating to production personnel.
2.8 Fail Safe System Design
Unsurprisingly, such personnel are soon likely to lose
confidence in a metal detection system that rejects product
which is subsequently shown to be good. Equally frustrating
to production operatives is a metal detection system that
requires constant attention in order to maintain appropriate
sensitivity standards.
A metal detection system that is capable of giving consistent,
reliable detection and rejection, without the frustration of
false rejection, will win the confidence of both line operators
and management. It will also provide the best long-term
protection for products, the brand, and the manufacturer.

12
2.1 Detector Electronics Design 2.1.3 Repeatability
Modern metal detectors benefit from advanced digital In addition to false rejections, drift can cause the detection
processing technology, which means that they can provide level to vary over time. However, a detector that detects a
a wide range of features at relatively little cost. However, a test sample repeatedly each time it is used (over a period of
large number of features will not necessarily contribute to a weeks or months) will instil confidence in the user.
metal detector’s overall effectiveness. A detector demonstrating such high levels of reliability also
The danger is that those new to metal detection may draw up avoids the problems of having to re-inspect product; a reliable
a checklist of desired metal detector features, and then make metal detector will identify metal contaminants and will also
comparisons between different brands. They may then make prevent them passing undetected on the production line.
an assumption that the brand with the longest list of features
is the best choice of metal detector for their purposes. 2.1.4 Ease of Set-up and Use
Those new to metal detectors may also believe that the If a detector has a complex or confusing set-up procedure,
greater the sensitivity, the more effective the unit is likely to it is likely that it will not be adjusted correctly. Set-up should
be – and they will use such criteria as a basis to evaluate be a simple and straightforward matter − and after initial
and compare different systems. instruction, the user should be able to adjust all parameters
Judging a metal detector simply by its range of features without reference to an instruction manual.
and/or its sensitivity levels alone is not the right approach Together, a logical set-up procedure and an intuitive Human
when choosing an appropriate system. Users with more Machine Interface (HMI) avoid having to memorise special
experience will know that, whilst sensitivity is important, it sequences; these two factors also permit changes to be
is only one of several key factors that should be borne in made in the correct manner long after initial training is given.
mind in the selection process.
Several detector brands promote ‘automatic set-up’ features;
however, actual ease of operation needs to be considered
2.1.1 Stability in conjunction with the accuracy and ease of the initial
Stability is the distinguishing factor of a top-quality metal set-up. The most modern detectors may include visual
detector, and highlights the difference between sensitivity and representations of complex product signals, which can
performance. In this context, ‘performance’ is a measure of greatly aid user understanding and the auto-set-up process.
equipment capability under real plant conditions. It is of great benefit if a metal detector can deliver an
A stable metal detector can operate consistently without automatic set-up of a similar standard to that which could
false rejects or erratic detections, and should not require be achieved by an experienced user. If this is not the case,
periodic adjustment. Most digital processing controlled there may be a significant loss in detection performance.
units will give similar sensitivity levels when tested side by Metal detectors can be set up on a single pass of a single
side under laboratory conditions; however, over extended product but a set-up taken over many passes of many
operation on a production line, significant differences may products will be more representative of actual production
well become evident. conditions. Leading detector brands now employ software
algorithms that deliver automatic setup equivalent to
An unstable detector, particularly when linked to an experienced operator level set up.
automatic reject device, can quickly attract criticism due to
its inefficient performance. The more product settings that there are, the more likely it
is that an operator will select the wrong setting for a given
product. So in order to minimise the number of product
2.1.2 Electronic Drift settings necessary, some detectors include features that group
Electronic drift is a common cause of instability, and occurs similar products together and establish a common setting.
over a period of time as a result of temperature variations
The benefit of such groupings is the ease with which the
and the ageing of electronic components. These factors can
systems can be operated and the accuracy of the set-up,
result in changing sensitivities (‘drift’), which may well lead
The latest Multi-Simultaneous Frequency (MSF) detectors
to false alarms or unwanted signals.
have taken this a step further and can provide a single
Frequency and phase stability of highly-tuned electronic product setting on which significantly different products can
circuits within the metal detector are key to minimising be inspected at sensitivity levels in excess of conventional
electronic drift – and the degree of stability required becomes single, dual or triple frequency counterparts.
increasingly important as the level of sensitivity is increased.
Design features such as Quartz Frequency Control,
Automatic Temperature Stability Control and Auto Balance
Control will make a major contribution to eliminating
drift – and this will help to ensure that the correct system
sensitivities are permanently maintained.

13
2
Key Design Features
2.1.5 Electrical Noise and Radio Frequency 2.2 Detector Mechanical Design
Immunity (RFI)
2.2.1 Environmental Protection
If the metal detector does not incorporate design features
with a high degree of Electrical Noise and Radio Frequency The selection of the metal detector should be commensurate
Immunity, the system will be vulnerable to false triggers that with the hygiene requirements of the product and the
will ultimately lead to a loss of confidence by production- environment in which it will operate. If the product is high-
line employees; it will also waste time and money through risk, the metal detector should be constructed to withstand
investigation of false product rejections. harsh conditions, deep cleaning and sterilisation routines.
In the manufacturing environment, there are numerous For producers of meat, poultry, dairy and similar products,
Radio Frequency Interference sources, such as fluorescent a metal detector’s inability to withstand frequent heavy
lighting, mobile devices, inverter (and variable frequency) duty wash down is a common problem. Single piece liners
drives etc. These kinds of items all have the potential to with no exposed joints are ideal. The repair of a metal
interfere with the operation of metal detection systems. detector suffering from water ingress is both expensive
and time-consuming.
2.1.6 Modular Electronics Providing the conditions are communicated to the metal
detection system-provider prior to purchase, system
Some metal detectors incorporate a universal quick-change performance should remain unaffected when equipment
electronic module, which is designed for change-out (i.e. is situated in any areas subject to water or steam.
replacement of malfunctioning or obsolete part with a
replacement part). If a metal detection system is to be used in an officially
designated potentially explosive environment, (such as
One of the main benefits of such a module is that it can help a flour mill), the system design should be independently
to reduce service costs and keep lost production time to a certified by an accredited recognised body. Furthermore,
minimum during maintenance and/or repair procedures. the detector manufacturer should be officially approved to
On production lines where downtime is unacceptable, it is make and sell such systems.
important to use detectors with a single electronic module
designed for user change-out. 2.2.2 Balance Stability and
Vibration Immunity
2.1.7 Self-Checking and
Most metal detectors operate on the basic principle of
Condition Monitoring the balanced coil system. Consequently, maintaining
With ever greater emphasis on improving uptime, mechanical stability is important to the ongoing
production efficiency (OEE), many producers are now performance of the metal detector.
focusing on reducing downtime, whilst also trying to reduce Very small movements in mechanical construction (such as
the burden of scheduled testing. expansion due to temperature, mechanical shock, vibration
Metal detection systems with self-checking and continuous etc.) can cause a coil system imbalance that may cause the
condition monitoring features can offer significant benefits by metal detector to false-trigger, drift or go out of balance.
providing early warning of a potential system failure. Such To address these problems, mechanical design and
systems allow preventive action to be implemented, rather construction should be equally as important as electronics
than depending on reactive maintenance and frequent testing. design and construction in preventing and compensating for
In considering the usefulness of such features, it is such movements.
important that the key parameters are continuously Systems prone to vibration, or which require regular manual
monitored through the actual working circuitry of the balancing, are of little value on an automated production
detector, regardless of whether signals are processed line. Good electronics design (such as Automatic Balance
sequentially or in parallel. Control), and good mechanical design (such as enhanced
It is also important that the system should automatically potting techniques) go a long way to minimising these
alert users via an early warning when there has been potential failure modes.
an unexpected change. Furthermore, the system should
trigger an alarm if there is an unacceptable change in the
parameters being monitored.

14
2
Key Design Features
2.3 Conveyor System Design 2.4 Non-Conveyorised

The design of the conveyor system that transports the System Design
product through the detector must meet certain strict criteria
if it is to avoid interfering with the detector in any way. Similar considerations should be given to the design of
metal detection systems which do not incorporate conveyor
A metal detector conveyor is much more than a modified systems. These include systems for the inspection of bulk
transport conveyor; the design of both conveyor and auto- dry powders and granular products, vertical packaging
reject device will have a major impact on the effectiveness applications, and pipeline systems used for liquids, pastes
of the overall metal detection programme. and slurries.
Unless special precautions and design techniques are Incorrectly designed support structures and reject devices
incorporated, the metal detector can become unstable due will have a major negative impact on overall metal detector
to eddy currents in the frame or static build up from the system performance, resulting in reduction of the metal
conveyor belt. These can influence the detector, causing detection programme’s effectiveness.
interference and a downgrading of sensitivity.
Metal detectors emit a high-frequency signal which causes 2.5 Reject Mechanism Design
tiny eddy currents to flow all around the metal structure of the
conveyor. These eddy currents have no effect on the detector Ineffective reject systems are probably the weakest part of
if they remain constant; however, if the conveyor structure has most detection systems, and result in metal-contaminated
an intermittent joint of variable resistance (even at a remote items not being effectively and reliably rejected from the
distance from the detector), the eddy currents change. This production line. A correctly specified system should be fool-
creates an intermittent signal that can be then picked up by proof and capable of rejecting all contaminated product under
the detector and result in a false trigger. all circumstances, no matter how frequent the occurrence,
and no matter what the location of the metal within the
Typical sources of eddy current loops are any intermittent product. (Refer to Chapter 4 for further information).
metal-to-metal contact such as bolted conveyor assemblies
non-insulated supports, pulley shafts and bearings, chain
drives and guards, reject supports and metal conduit clamps. 2.6 Hygienic Design
To obtain the best and most reliable performance, fully All metal detection systems should be designed with due
welded structures are required. These should incorporate: consideration of the environment in which they will operate.
They should also be designed so as to take into account
• Correct metal-free zones
cleaning regimes likely to be encountered.
• Properly insulated rollers and pulleys
• Fully welded cross-structures Hygienic design principles should be applied to every aspect
• Insulated detector head mountings of the system, with the aim of eliminating dirt traps and
ensuring easy cleaning, so design features should include:
Conveyor belting materials should be metal-free and
manufactured to a very high standard, using suitable • Elimination of cavities/bacterial traps
contaminant-free joints. Anti-static belting materials should • Sealing of all hollow sections
be avoided. • Avoidance of ledges and horizontal surfaces
• Use of open-design continuous-welded frames for easy
If these problems are not solved at source, the common access and cleaning
outcome is a gradual increase in false rejects. The easy (but • Hygienic management of electrical cables, trunking and
generally unacceptable) solution is to reduce the sensitivity of pneumatic services
the detector. However, this could result in contravention of the
sensitivity standards that have been specified.
Effective conveyor design considerations, product transfer
methods and recommended types of belts are covered in
greater detail in Chapter 4.

15
2
Key Design Features
2.7 Health and Safety

Health and Safety is an important consideration, so design
and build of metal detection systems should be certified
as being in accordance with statutory regulations and
standards in force at the time of sale.
For example, CE marking in relation to applicable machinery
safety standards will minimise the risk of an employee
being hurt. Any such injuries could result in costly personal
injury claims.
2.8 Fail Safe System Design

Consideration should be given to the implications of a
system failing to function as intended; for example, a
reject device not removing contaminated product or a fault
occurring within the metal detector.
It is good practice to integrate fail safe design features
into the metal detection system, so as to mitigate the risks
associated with system malfunction. For example:
• Reject confirmation systems can confirm that the
contaminated product has been rejected into the reject bin
• In-built condition monitoring systems can be used
to provide early warning of a change in the state or
performance characteristics of the metal detector
The numerous aspects of fail safe design that need to
be considered are covered throughout this guide. The
following list is provided as a quick reference guide to
finding such information.
Relevant sections: 1.2, 4.1.9, 4.2, 4.3.1, 4.4.6, 4.5.3,
9.3, 12.2, 12.6.2, 12.7.2, 12.10, 13.2, 14.2.4

16
2
Key Design Features
Notes

17
3
Factors Limiting Sensitivity

There is often widespread confusion and misinformation regarding metal detector sensitivity specifications and
operating capabilities. Many factors can influence sensitivity performance, and if sensitivity data is to be meaningful,
there needs to be certainty that it is correct and accurate within its area of application. This chapter highlights factors
that need to be considered in order to ensure that newly-purchased metal detector systems provide the right levels
of sensitivity, whilst also delivering proper performance in accordance with operational needs.
3
Factors Limiting Sensitivity 3.1 Factors Limiting Sensitivity
In most markets, sensitivity performance is usually
3.1 Factors Limiting Sensitivity expressed in terms of the diameter of a sphere made from a
specific type of metal.
3.2 Types of Metal
This sphere must be reliably detectable when placed in the
3.3 Shape of Metal and Orientation Effect centre of the aperture of the metal detector. Precision metal
3.4
Aperture Dimension / Position of Metal in Aperture spheres are used for this purpose because they are readily
available in a range of metals and diameters.
3.5 Environmental Conditions
In addition, they have a consistently spherical shape,
3.6 Inspection Speed regardless of how they are presented to the detector. This
3.7 Inspecting Dry Non Conductive Products means that they have no ‘orientation effect‘. (See section
3.3 for more details).
3.8 Wet Product Inspection - a Detailed Look
Japanese metal detector manufacturers state sensitivity
3.9 Automatic Product Compensation
performance using similar spheres and materials; however,
3.10 Product Signal Suppression these spheres are often measured on the conveyor belt and
not in the centre of the detector’s aperture. Using this form of
measurement, the sphere will be closer to the aperture wall
which offers increased sensitivity levels (covered later in this
chapter), so higher rates of sensitivity could be observed
when compared to differing methods, which measures centre
of aperture or worst-case sensitivity performance.
When comparing the performance of metal detectors, it
is therefore important to ensure that the sensitivity of the
machines is being measured in the same way.
Many factors influence the actual operating sensitivity
at which a metal detector is able to reliably perform.
These include:
• Type of metal
• Shape and orientation of metal
• Aperture size/metal position in the aperture
• Environmental conditions
• Inspection speed
• Product characteristics and operating frequency
For these reasons, care needs to be exercised when
comparing metal detectors based only on the information
contained in a specification or within promotional literature.

18
The information contained within a specification may not be Where non-ferrous and stainless steels are specified in the
capable of being reliably achieved once the metal detector detection specification, these should be quantified because
is installed in the intended application and operating there are numerous types of materials – all of which vary in
environment. That’s why sensitivity performance-testing detectability. For example, brass is more readily detectable
under controlled laboratory conditions is not regarded as a than phosphor-bronze, but both are non-ferrous metals.
good indicator of actual achievable performance.
In the food-processing and pharmaceutical industries, the
Actual product testing ‘in situ‘ is essential in order to two most common grades of stainless steel are 304 and
determine the production-line sensitivity of a metal detector. 316. However, these are always the most difficult grades to
Therefore in-line factory tests need to be undertaken to ensure detect, due to the fact that they are non-magnetic and have
that any officially quoted sensitivity performance figures are poor electrical conductivity.
repeatable, without the possibility of false rejecting in the
intended application and operating environment.
3.3 Shape of Metal
3.2 Types of Metal and Orientation Effect
Metals can generally be categorised as being either If a non-spherical particle of metal, such as swarf (a fine
ferrous, non-ferrous or stainless steel. The sensitivity of a piece of metal from machining operations) or wire passes
metal detector will vary dependent upon the type of metal through a metal detector, it will be easier to detect when
contaminant present. The ease of detection depends on the passing in one particular orientation, compared to another
magnetic permeability of the metal contaminant (i.e. how orientation. This is known as the ‘orientation effect‘ and is
easily it is magnetised), as well as the electrical conductivity common to other devices used to detect metal contamination
of the metal contaminant. all detection devices, not just high-frequency metal detectors.
If the contamination is of ferrous metal, then it is both Figure 3.1 shows that a metal detector varies in its ability to
magnetic and a good electrical conductor, so it will be identify wire contaminants – and this variation is dependent
easily detected. Non-ferrous metals such as brass, copper, on the type of metal from which the wire is made, as well as
phosphor-bronze and aluminium are non-magnetic, but are the orientation of the wire.
good electrical conductors. This means that they are relatively Ferrous contaminants are easy to detect when they are
easily detected in dry applications, but are more difficult to presented in an orientation parallel to the direction of travel
detect in wet applications, because they are non-magnetic. ‘A’. However, they are much more difficult to detect stainless
Stainless steel is available in many different grades, some steels (SuS) when they are at 900 (i.e. right angles) to the
magnetic and some austenitic (totally non-magnetic), plus direction of flow ‘B’. Non-ferrous metals are exactly the
their conductivity is also variable. Table 3a summarises the opposite (as seen in Figure 3.1).
key characteristics of various types of metal.
Metal Type Magnetic Electrical Ease of

Permeability Conductivity Detection
Ferrous Magnetic Good Easily

A
(Chrome Electrical Detected1
Steel) Conductor
B
Non-Ferrous Non- Generally Relatively
(Brass, Lead Magnetic Good Easily
Copper) or Excellent Detected2
Stainless Steel Usually Usually Poor Relatively

(Various Non- Conductors Difficult to
Grades) Magnetic Detect A B
Table 3a: Characteristics of various types of metal

Ferrous Easy Difficult
Notes:
Non-Ferrous Difficult Easy
1
Typically the easiest metal to detect in both wet and dry
applications, due to the magnetic properties Stainless Steel Difficult Easy
2
Relatively easily detected in dry applications; however;
more difficult to detect in wet applications due to non- Figure 3.1
magnetic properties

19
3
The orientation effect is only evident when the diameter of As an example, when operating at 1.5mm diameter, the
the wire is less than the spherical sensitivity of the metal piece of tinned copper wire would need to be 9mm long to
detector. For example, with the detector sensitivity set at guarantee detection. At 2.0mm sensitivity, this would increase
1.5mm diameter, only wires thinner than 1.5mm diameter to 26mm in length before detection could be guaranteed. It
will show the orientation effect. If the detector sensitivity is can be seen that a small change in detector sensitivity (in
increased to 1.0mm, only wires less than 1.0mm diameter terms of ball diameter) will make a great difference to its
will cause a problem. If the diameter of a wire is only about sensitivity with respect to wire pieces.
1/3 the diameter of the detectable sphere, the wire may not
Clearly, if wire is identified as a potential contaminant,
be detectable, no matter what its length.
it is best to operate the detector at the highest possible
Table 3b compares a detector’s ability to detect four different sensitivity. However, as sensitivity levels increase, the
wire samples at various detector sensitivities. The left-hand problems of ‘drift‘ become more acute – and with some
column shows four different sensitivities. detectors, nuisance false rejects will increase to an
unacceptable level. Consequently, the design of the metal
detection system becomes more important (Refer to Chapter
2 for further information).
Spherical Steel Tinned Copper Wire Copper Wire Stainless Steel Wire – EN58E (304)
Sensitivity Paper Clip Dia 0.91mm Dia 1.37mm Dia 1.6mm
Dia 0.95mm (0.036”) (0.054”) (0.063”)
(0.037”)
1.2mm 1.5mm long 3.5mm long
1.5mm 3.0mm long 9.0mm long 3.0mm long 8.0mm long
Table 3b: Sensitivity Levels of Different Sized Spheres and Lengths of Wire
3.4 Aperture Dimension/ Typically, the detectable ball size at the centre of a rectangular
aperture is approximately 1.5 to 2.0 times greater than the
Position of Metal in Aperture size of the ball detected at the aperture corners; however,
this may vary dependant upon the manufacturer and specific
A large-aperture detector is less sensitive than a detector with a
design. The detection variance is demonstrated pictorially in
smaller aperture. Both aperture width and aperture height have
Figure 3.3. Detectors with a circular aperture will have the
an influence on the detector’s sensitivity, but changes in the
highest sensitivity close to the detector walls, and a reduced
aperture height (or the smaller aperture dimension) will have
sensitivity towards the geometric centre of the aperture.
a greater effect. Figure 3.2 shows a typical metal detector. The
geometric centre (position 1) is the least sensitive part of the
detector, whilst the corners are the most sensitive (position 3).
All other points will lie somewhere between (e.g. position 2).
This phenomenon is known as the ‘sensitivity gradient‘, and
A
will depend on the design and assembly of the coil system.
BB
2
1 Figure 3.3
3
Figure 3.2

20
3
3.5 Environmental Conditions 3.7 Inspecting Dry Non

Metal detectors can be influenced, to varying degrees, by Conductive Products
adverse environmental conditions such as airborne electrical
interference, plant vibration and temperature fluctuations. Dry products, such as confectionery and cereals, are
These effects become even more acute when operating at relatively easy to inspect – and sensitivity charts can be used
high sensitivities. to calculate the expected operating performance. Detectors
operating at high and ultra high tuned frequencies (typically
Ovens, freezing-tunnels and hot water wash-down all create in the region of 800 Khz and 900 Khz) are available which
thermal shock that can result in false reject signals. Unless deliver high levels of overall sensitivity and are especially
good design techniques are employed to eliminate the good at detecting stainless steel type contamination.
problem, the only solution may be to reduce the sensitivity
of the detector. So, when comparing detector capabilities, When inspecting wet or conductive products such as fresh
testing under controlled laboratory conditions is not realistic. meat, poultry, cheese, fish and metallised film packed
products, the situation is different. The wet product itself
creates a ‘product effect signal‘ in the detector – and this
3.6 Inspection Speed signal needs to be cancelled out before inspection can
begin. It should be noted that the product effect signal tends
Minimum and maximum inspection speeds are seldom a
to reduce the sensitivity of the detector in a way that cannot
limiting factor for metal detectors, particularly on conveyor-
easily be calculated and in most cases if an indication of
type applications. The upper limit of inspection speed will
sensitivity is required then a product test will be required to
vary from manufacturer to manufacturer, but ultimately, it will
give an accurate indication.
be determined by the detector aperture height.
Typically this will be a maximum of around 4m/sec (26
ft/sec) for an aperture of 125mm (5”) in height. Minor
3.7.1 Dry Product Inspection –
modifications are usually possible to extend this range a Detailed Look
further. The limit of performance is often reached when When a wet product passes through a metal detector it exhibits
attempting to inspect on pneumatic pipelines at speeds in a signal which can either be mainly reactive or resistive
excess of 35m/sec (115 ft/sec). depending on the product characteristics (see Section 3.7.1
A uniform sensitivity over the full speed range is more Wet Product Inspection). If the product is dry however it is likely
important than the absolute maximum and minimum to be neither conductive nor magnetic and therefore has a
inspection speed. This is not universal to all detectors, and negligible product signal. Any signal it could possibly have will
Figure 3.4 shows detector ‘A‘ maintaining sensitivity over a be so close to zero (or zero phase angle) it will be insignificant.
very wide speed range, while the profile of detector ‘B‘ shows To explain what is meant by this we utilise vector diagrams
that it is more speed-dependent. and Figure 3.5. The Figure 3.5 shows the signal (vector)
produced by the dry product which is represented by the black
shaded area.
Detector A
Detector A
900
Product Signal
Sensitivity
Sensitivity 00
Min
Min Max
Max
Velocity
Figure 3.5
Velocity
Detector B
Detector B
Sensitivity
Sensitivity
Min Max
Figure 3.4 Velocity

Velocity

21
3
Depending on the application and installation vibration Depending on the operating frequency of the metal detector
or excessive vibration signals can be present which have and its aperture size used signals from pieces of ferrous
to be managed in order for effective metal detection to be metal are larger than signals from pieces of non-ferrous or
undertaken. In Figure 3.6, the vibration signal (vector) is stainless steel metal of the same size – and signals caused
represented by the dark blue arrow and it is noted that the by vibration are always along the horizontal reactive axis.
vibration signal is aligned to the zero phase point along the
To improve the metal detector‘s ability to detect metal and
reactive axis. The position of the vibration signal is actually
to reduce the impact of vibration, special circuits can be
set to this zero phase point during the build and set up of the
used to amplify the signals by differing amounts, according
metal detector.
to phase. This technique is known as ‘Phase-Sensitive
Detection (PSD). This is shown in Figure 3.8
900
900
Stainless Steel
Stainless Steel
00 Ferrous
Vibration Ferrous
Vibration 00
Vibration
Figure 3.6
Non-Ferrous
Non-Ferrous
We use the same vector representation to explain how the

signals from metals are generated and why ferrous metal is
generally easier to detect than stainless steel. Figure 3.8
As mentioned earlier in this chapter, there are two types

of signals created by various metals as they pass through The PSD is shown as a long thin grey oval called the ‘detection
the coils of a metal detector: these are known as ’reactive‘ envelope‘, and for a signal to be detected, it must pass outside
and ‘resistive‘, according to the conductivity and magnetic the detection envelope. As the detection envelope is positioned
permeability of the metal. in the same position as where the vibration signals lie it
requires a large vibration signal before it can pass outside the
When the metal particles are small, the signal from ferrous envelope and be detected; by comparison, only small signals
metal is primarily reactive, while the signal from stainless from ferrous, non-ferrous and stainless steel are necessary –
steel is primarily resistive. Figure 3.7 shows a vector and these are the most satisfactory operating conditions.
diagram of the signals from a number of different metals as
they pass through the detector. They show that: In general achieving high sensitivities when inspecting dry
products is relatively straight forward. If using tuned high
1. The signals increase to a maximum as they pass through
or tuned ultra high frequencies and detectors with the right
the first coil
size aperture for the product being inspected the achievable
2. The signals decay to zero as they pass through the
sensitivity levels will be excellent especially with respect to
centre coil
stainless steel detection levels.
3. The signals again increase to a maximum when passing
through the third coil Table 3c shows the typical sensitivity level for dry
4. The signals have varying ‘phase angles’ primarily determined product inspection when using tuned and ultra high
by the metal type (reactive and resistive components frequency technology.
900
Stainless Steel
Stainless Steel
Aperture size Ferrous Non Ferrous Stainless
Ferrous Metal Metal (brass, Steel 316
Ferrous
copper and grade non-
aluminium). magnetic
Vibration 00
Vibration
350 mm x 0.50 mm 0.40 mm 0.60 mm
Non-Ferrous
50 mm
Non-Ferrous
350 mm x 0.70 mm 0.70 mm 0.90 mm

Figure 3.7 125 mm
350 mm x 0.85 mm 0.95 mm 1.10 mm
200 mm
Table 3c: Typical sensitivity levels for dry product inspection when
using tuned or ultra-high frequency detection

22
3
In Figure 3.9, you can see that both the vibration signal Figure 3.11 shows a metal detector set to inspect a dry
(dark blue arrow) and the product signal (black vector) product (the PSD is set at the zero phase point) but with the
are contained with the detection envelope and as such go signal exhibited from a wet product, which is clearly outside
undetected. The metal signals however all appear outside the confines of the detection envelope and as such would
the boundary of the detection envelope and are therefore are create an unacceptable false trigger from the metal detector.
detected by the metal detector. The actual size detectable
is primarily down to the size and design and operating 900 Product
Product Signal
Signal
Detected
frequency of the metal detector and its ability to operate in Stainless Steel
Stainless Steel
its intended environment. Ferrous

Ferrous
900
Vibration 00
Stainless Steel
Vibration
Product Signal Ferrous
Non-Ferrous
Non-Ferrous
Vibration 0 0
Figure 3.11
Non-Ferrous
By reducing the operating sensitivity of the detector, all the
signals will become smaller until the product signal no longer
Figure 3.9 passes outside the envelope, making inspection possible. For
an application with a small product effect, this is the most
common option. However reducing the sensitivity will clearly
3.8 Wet Product Inspection – impact the operating performance of the metal detector to a
a Detailed Look greater or lesser degree
As discussed in the previous section, when a wet (or An alternative solution is shown in Figure 3.12. The
conductive) product passes through a metal detector it detection envelope can be rotated electronically until it is
will exhibit a signal which can be either mainly reactive or aligned with the product signal. This is known as ‘product
resistive. Depending on the product in question this signal compensation‘ or ‘phasing out‘ the product signal and this
can be large and complex. See Figure 3.10. can be undertaken by the user during the set-up of the metal
detector. The product signal no longer passes outside the
900 Product Signal envelope, so normal inspection is again possible.
StainlessProduct Product Signal

Steel Signal
Undetected
Stainless Steel
900
00
Ferrous
Ferrous 00
Vibration
(limits performance)
Figure 3.10 Vibration
Non-Ferrous
Non-Ferrous
To successfully detect metal, the metal detector has to
ignore this signal at the same time as being capable
of detecting the smallest pieces of metal contaminants Figure 3.12
possible and operate in the intended factory production
environment (i.e. be immune to the effects of external However, using product compensation can have
limiting factor such as plant vibration). disadvantages. It is not uncommon that the signal given off
from the product has a similar phase angle to that given off
by stainless steel, in that they align themselves very closely.
For stainless steel to be detected, the signal from the metal
needs to be larger than the signal from the product. This in
turn means relatively large signals from stainless steel are
needed if the signals are to pass outside the envelope. This
results in the detector becoming less sensitive to these metal
types. At the same time, small signals from vibration may
pass outside the envelope and be detected. Undue sensitivity
to vibration is often the limiting factor when inspecting with
product compensation.

23
3
It is often necessary that the operating sensitivity needs to

be reduced in conjunction to using product compensation to
Metal Metal (brass, Steel 316
ensure effective and reliable metal detection is achieved.
The exact phase of any product cannot be calculated from data aluminium). magnetic
based on salt content moisture levels or pH value, which in
turn means that detection sensitivities cannot be calculated. 350 mm x 0.8 to 1.2 1.0 to 1.5 1.5 to 2.0
Product testing is essential to determine the detector’s 50 mm
sensitivity to a range of metals where there is significant
product effect – and this service is usually available from metal 350 mm x 1.2 to 1.8 1.8 to 2.5 2.0 to 3.0
detector manufacturers. 125 mm
3.9 Automatic Product Compensation 350 mm x 1.5 to 2.2 2.2 to 3.0 2.5 to 4.0
200 mm
It requires considerable experience to accurately adjust the
product phase in order to achieve optimum performance. Table 3d: Typical sensitivity levels when using multi-frequency detection
If a number of different products or pack sizes are to be
checked on the same production line, adjusting the detector
for each new product can be time-consuming. 3.10 Product Signal Suppression
Most modern detectors have an automatic setup or learn In more recent times a new technique has been developed
facility for configuring product settings in preparation for that far more effectively deals with the signal generated from
inspecting product. These routines range from a basic level, the product. Rather than simply masking the signal, the new
where the phase of the detection envelope is pre-set, to technique actually attempts to remove or reduce the product
a much more advanced level routine that sets sensitivity signal and by doing so renders the online achievable
and frequency. These are known as multiple frequency sensitivity considerably better.
machines. This new technique called “Product Signal Suppression”
uses advanced software algorithms to reduce the size
Automatic setup routines normally follow a process of of the active product signal (Figure 3.13) by modifying
requesting a pack or a small number of packs are passed the product signal rather than simply masking it. To do
individually through the aperture within specified time this the metal detector operates with 2 or more active
limits. In general these routines work fine, however in some frequencies simultaneously. Detectors of this type are
cases additional manual adjustment is needed following referred to as having Multi Simultaneous Frequency (MSF)
setup to account for variation in product effect, which is technology. Using product signal data derived from more
not uncommon in wet product applications. The most than one active frequency simultaneously these new MSF
sophisticated detectors on the market today have intelligent metal detectors use various combination of high and low
routines that account for product effect variation during frequencies simultaneously.
setup to deliver a more optimized and trouble free setup.
This is done by the detector allowing a greater amount 900 Product Signal A
of product to pass during setup whilst configuring the 1st frequency
detection envelope in a complex and more efficient manner
to account for product variation.
However the resultant achievable sensitivity will be mainly 00
governed by the product signal and the actual results are
likely to be somewhat different to those achieved when
inspecting dry products – see Table 3d. Product Signal B
2nd frequency
Figure 3.13

24
3
As the resultant product signal (Figure 3.14) is considerably

reduced and much smaller pieces of metal are now
detectable with an online sensitivity far closer to those
achieved when inspecting dry products (Figures 3.13, 3.14
and Figure 3.15 and Table 3e).
900
350 mm x 0.6 to 0.8 mm 0.8 to 1.0 mm 1.0 to 1.2 mm
50 mm
Resultant product
signal after processing
through software 350 mm x 1.0 to 1.2 mm 1.0 to 1.5 mm 1.2 to 1.8 mm
125 mm
00
Product signal
suppression 350 mm x 1.2 to 1.5 mm 1.2 to 1.5 mm 2.0 to 2.5 mm
200 mm
Figure 3.14 Table 3e: Typical sensitivity levels when using Multi-Simultaneous
Frequency Detection and Product Signal Suppression
900
Stainless Steel
Product Signal
Ferrous
Vibration 00
Non-Ferrous
Figure 3.15
This technology can also effectively deal with product

variations which historically has been the cause for a high
false reject rate or a reason to reduce the metal detectors
operating sensitivity. Once a detector has been set up to
inspect a particular product, the product signal suppression
technology is applied to each product that passes through
the detector.
The on-board detector electronics automatically adjusts
for slight variations in the product effect, which in turn
dramatically reduces the occurrence of false rejections.
Table 3e lists sensitivities that are typical when using
Multi Simultaneous Frequency technology in conjunction
with Product Signal Suppression software. The results are
achievable when inspecting wet / fresh products such as
meat, poultry, fish, cheese produce and thawing / semi-
frozen products plus those packaged in metallised film.

25
4
Systems Design and Applications

Having discussed the defining features of a reliable metal detection system in the previous chapter,
it is now important to understand:
• The different types of metal detector systems available

• Where metal detectors can be installed
• How metal detectors can be specified correctly with regard to application, best practice and accepted codes of
practice. Time spent correctly specifying the metal detection system will be rewarded by avoidance of major
modifications after installation; correct specification also ensures ease of verification testing.
4
Systems Design and Applications This chapter provides practical guidance on equipment
selection; it also explains how the adoption of best-practice
techniques and fail safe features can further reduce the likely
4.1 Conveyor Systems occurrence of contaminated products reaching the customer.
4.2 Satisfying Retailer and Food Industry Requirements
4.3 Inspection of Liquids, Slurries and Pastes
4.1 Conveyor Systems
in a Pipeline
4.1.1 Belt Types
4.4 Gravity Feed Inspection of Bulk Powders and
A number of factors need to be considered when choosing
Free-Flowing Solids
suitable conveyor belt material. Static charges can build up,
4.5 Vertical Packaging Applications particularly when conveyor belts are running over plastic skid
plates or plastic-coated rollers and pulleys.
Special anti-static belt materials can cause a problem, since
they can be made of conductive carbon fillers or additives,
which will adversely affect the metal detector’s performance
− particularly when the belt joint passes through the aperture.
With any type of belt, the joint must be metal-free and
made in such a way as to prevent product build-up or an
accumulation of grease or product residue. A vulcanised
or glued joint at 45º or an interlocking finger joint helps
minimise this effect (Figure 4.1). Metal fasteners, or sewn
and laced joints, are unsuitable.
Vulcanised Joint
Vulcanised Joint
Finger Joint
Finger Joint
Figure 4.1

26
The belt material itself must also be totally metal-free, 4.1.2 Product Transfer
since tiny metal particles in the material are extremely
difficult to find if they come loose. Belt manufacturers Packaged products should be transported through the metal
producing consistently high-quality, metal-free belts almost detector in a consistent orientation, centred relative to the
certainly need to use metal detection equipment to inspect detector aperture. The ideal minimum spacing is the length
both their raw materials and finished products. of the product pack.
A wide variety of belt types are available to suit most Transfer onto the conveyor system needs special
applications. These include: consideration when the end rollers are large or when the
product is small. If the distance (‘D’) between rollers is
• Flat, dished, ribbed, and moulded flexible wall belts
more than half the product length, reliable transfer will not
• Solid, plastic modular chain belts and circular section
be possible. Small, non-powered intermediate rollers or a
(round) urethane belts running in grooved rollers – these
dead plate positioned between the two rollers are usually
are ideal where product spillage is likely and frequent
effective to ensure correct spacing (Figure 4.4).
wash-down is required (Figure 4.2)
• Endless ‘double-pass’ belts (Figure 4.3) − these offer a
number of advantages in many applications, including D
rapid replacement. However, the face of the belt passes
over a roller, so these belts are not considered suitable
for transporting wet or sticky product.
SideWall
Side Wall Belting
Belting
Figure 4.4 Idle

Idle Roller
Roller
Single or double knife edges permit transfer of very small

items, under circumstances in which product registration
Round
RoundUrethane Belting
Urethane Belting must be retained, as with rows of confectionery at the
outlet of an ‘enrober’ (a machine which coats confectionery
products in chocolate) (Figure 4.5).
Figure 4.2 D
EndlessDouble
Double Pass BeltBelt Single KnifeEdge
Single Knife Edge
Endless Pass
Double KnifeEdge
Double Knife Edge
Figure 4.5
Figure 4.3
Sticky products (such as raw dough or meat) and bulk

loose product (such as loose peanuts) can be transferred
by cascade. For optimum sensitivity performance and no
false rejects, it is important for the product to be presented
consistently and not in large lumps (Figure 4.6).
Cascade
Cascade Inspection
Inspection
Figure 4.6

27
4
For jar and bottle inspection, the detector system can 4.1.4 Automatic Rejection Systems
be positioned alongside the existing transport conveyor.
Product guides can be used to divert the containers from the The most appropriate choice of reject system will depend on a
line onto the detection system, and acceptable products are number of factors, and the advice of the detector manufacturer
then diverted back onto the main line. When contamination should always be sought. However, some of the common
in a product is identified, the product diverter can be moved types and general applications are described below:
pneumatically so as to allow the contaminated item to be Air Blast
rejected from the line (Figure 4.7). A blast of air blows the product into the reject location
(Figure 4.9).
This type of reject is ideal for light, single-file discrete
products running on a narrow belt width. It is recommended
that a ‘gated timer’ is used in conjunction with the air
blast to ensure the air blast is directed at the centre of the
product, regardless of the location of the contamination.
(See section 4.1.8 for further details).
Diverter
Diverter Rejected
Rejected
Product
Product
Air Blast
Air Blast
Inspecting
InspectingJars andBottles
Jars and Bottles
Figure 4.7
4.1.3 Transfer Speed

To allow easier identification of the contaminated items, it is
often useful to accelerate the product through the detector
so as to create an increase in product spacing.
When packs are very close together, the detector may be
unable to determine which of the packs is contaminated,
so two or three packs may need to be rejected to be sure of Rejected
Figure 4.9 Rejected Product
Product
rejecting the right one. By increasing the detector conveyor
speed, product spacing is increased, permitting the individual
items to be identified, and then rejected when necessary. Punch/Pusher
This device operates at high speed, and pushes individual
When inspecting bulk and loose product, the height of the
product into the reject location (Figure 4.10). High belt
loose product can be reduced by increasing the speed of the
speeds are acceptable − but on closely-spaced items,
product when it is transferred from the production-line conveyor
recovery time needs to be extremely rapid.
to the metal detector conveyor. This minimises the volume
of rejected product and permits the use of a lower detector This type of reject is suited to light-weight to medium-weight
aperture, so as to ensure higher sensitivity (Figure 4.8). discrete packs, spaced and oriented on a narrow belt
width. The punch/pusher must always be ‘gated’ to ensure
that the punch strikes the centre of the product every time,
Reduced Product
Reduced Product Height
Height regardless of the location of the contaminant. This type of
reject is unsuitable for loose or fragile products.
Punch/Pusher
Punch/Pusher
V1 V2 V3
V3
Figure 4.8
When installing systems in a line which may operate

at various speeds, it is not always necessary to have
the detector system run at variable speeds as well. The
additional expense and complexity of such a process can Figure 4.10 Rejected
Rejected Product
Product
often be avoided by fixing the detector system so that it can
run at the normal line maximum speed.

28
4
Sweep/Diverter Arm Retracting Belt

Retracting Belt
An arm moves at an angle across the belt to divert items
(Figure 4.11).
This type of reject is suitable for light-weight to medium-
weight, discrete or random non-oriented product running on
a narrow belt, typically up to 350mm wide. Care must be
taken that the product enters the reject bin correctly, since it
will usually enter diagonally.
Sweep/Diverter ArmArm
Sweep/Diverter
Figure 4.13 Rejected

Rejected Product
Product
Reverse Belt
Two types are available. (Figures 4.14 and 4.15) On
detection of metal, either the inspection conveyor or the
feed-off conveyor is reversed for a short time, so as to divert
contaminated product into a reject container.
This type of reject is ideal for bulk loose, dry or sticky
product, or multiple random items.
Rejected Product
Product
Reverse on
Reverse on Normal
Normal
End Flap/Dump Detection
Detection Product Flow
Product Flow
This type of system necessitates a drop in production-line
height (Figure 4.12), if necessary this can be overcome by
an incline on the conveyor. The point of pivot can be varied
according to the application.
This type of reject is suited to small discrete items at random
or loose bulk items (dry or sticky) running on a wide, flat or
dished inclined conveyor belt.
Rejected Product
Rejected Product
End FlapReject
End Flap Reject
Figure 4.14
Rejected
Rejected
Product
Product
Reverse on
Reverse
Figure 4.12 on Detection
Detection
Normal
Normal
Retracting Belt Direction of
Direction of
Product
ProductFlow
Flow
The end roller moves back to create a gap in the flow, which Product Flow
Product Flow
allows the product to drop through (Figure 4.13). After

Figure 4.15
product rejection, the roller moves forward to the closed
position faster than the belt speed, avoiding the danger of
trapping product. End rollers can be made in a knife-edge
format to ease the transfer of small items.
This type of reject is very reliable for most applications.
Where more than one product passes in-line across the
width of the conveyor, a retracting belt rejection mechanism
should be used.

29
4
4.1.5 Stop Alarm Systems It is important that the timers are capable of being re-set
immediately, and that the detector is still operational while
A simple stop alarm system would probably only be timing-out. The detector must also be capable of detecting
considered acceptable by most manufacturers and major a second particle in a following pack; then it must either
retailers under circumstances in which automatic rejection re-set or extend the timer to ensure that the second pack is
is not possible or practical. Typically, they are used on large also rejected. A continuous stream of metal particles should
bags or boxes where rejection is difficult. result in the reject device operating continuously until all the
When metal is detected, the conveyor should stop metal contaminated products
immediately and all products on the conveyor belt should are removed.
be removed for investigation. The system should also be Variable Speed/Stop-Start Applications
fitted with an audible and/or visual alarm. It should only
be possible to re-start the system using a key held by a Accurate rejection and timing become more complex if the
nominated person or have a controlled reset as an option. transport conveyor has variable speeds or can be stopped
when product is located between the detector and the reject
These solutions are generally considered high-risk, and system. The time taken for the product to move to the reject
depend entirely on the competence of the line operator. position is not constant and so a simple time-delay method
cannot be used.
4.1.6 Metallised Film Applications The normal solution is to use a speed encoder which can
To achieve good detection of metal in a metallised film monitor both belt movement and the position of product on
pack, it is necessary to remove the signal generated by the belt. A shift register is a device which will give an output
the fine aluminium coating present on the plastic film as signal after it has received a pre-determined number of input
discussed in section 18.2.6. pulses. It does not matter if these pulses are received rapidly
or are received in a spaced-out pattern over a long period.
It is preferable to have a space between the product and
the aperture of approximately 50 to 60mm all around; and The input pulses are produced by an encoder fitted to
extremely high levels of sensitivity can be achieved. the shaft of a roller on the conveyor system. The pulse-
generator is normally made from a metal disc with teeth
or holes cut into it. Each time a tooth on the disc obscures
4.1.7 Aluminium Foil Applications the photo-electric device or passes close to the proximity
When the packaging material includes aluminium foil, it is sensor, a pulse is generated. In the example shown (Figure
possible to undertake metal contamination checks before 4.16), each revolution of the disc generates fifteen pulses.
packing, using a balanced coil-type metal detection system.
Alternatively, a ‘Ferrous-in-Foil’ type detector can be used Proximity Sensor
Proximity Sensor
after packing. However, as stainless steel and non-ferrous
metals are not detectable using the ‘Ferrous-in-Foil’ unit, this
is only recommended when no other alternative is available.
4.1.8 Reject Timing

A time-gap is usually required between the moment of
detection and the moment of rejection, so as to allow the Pulse
Pulse Generator
Generator
metal contamination to move to the reject point.
Figure 4.16
This time-gap can be just fractions of a second on high-
speed applications, where the detector and reject device are
A higher number of teeth will produce more pulses per
close together; by contrast, the time-gap can be as much
revolution, which will ultimately allow more accurate
as 30 seconds, when rejection is planned (manually or
registration of the product.
automatically) at some remote point.
The signals from multiple or consecutive metal particles
A second independent timer is also required in order to
will be stored in the shift register and sequenced through,
control the length of time during which the reject device
ensuring that each one is subsequently rejected.
operates. This is usually adjustable from approximately 0.5
seconds to 10 seconds. A punch-type reject would require
the shortest time, whilst a retracting band system would
normally operate for several seconds to remove larger items
from a slow-moving belt. Both timers are normally available
as standard features from the detector system manufacturer.

30
4
4.1.9 Photo-Gated Timing When punch rejects (or air blasts) are used, the best
solution is to accurately monitor the position of the pack
Figure 4.17 shows a typical reject without photo-gating, and operate the reject device when it has reached the
with the timing adjusted to accurately reject a known metal correct position. The technique is known as ‘photo-gating.’
particle type and size located in the centre of the pack. If a This will ensure accurate rejection, independent of the
particle occurs at the leading or trailing edge of a different location size and metal type in the pack.
size and type, the reject may operate too early or late,
possibly missing the pack or disturbing the neighbouring
packs and causing a line blockage (Figures 4.18 and 4.19). 4.1.10 Typical Reject Problems and
If using air-blast or diverter arm rejection, a possible solution
Fail Safe Design
is to adjust the timers so that they operate early and for an Ineffective reject systems are probably the weakest link
extended period. However, this would remove several good in most detection systems, and they can result in metal
packs and would also be likely to spin or disturb others. contamination not being effectively and reliably rejected from
the line. A correctly specified system should be fool-proof
Punch
Punch rejectwithout
reject without Photo-gating
Photo-gating and capable of rejecting all contaminated product under all
circumstances, however frequently contamination occurs,
and whatever the location of the metal inside the product.
The following are common application problems which
should be taken into consideration when specifying a metal
detection system:
• Reject not suitable for the application i.e. Air blast
specified on 2 kgs pack (pack too heavy)
• Reject not photo-gated. The potential problem is
associated with the location of the metal in the product.
The longer the packs the greater the risks. If such a
Rejected Product
Product system is not used, the wrong pack can be rejected or
the item may not be rejected properly, causing a possible
Punch
Punch reject operates
reject operates tootoo
latelate line blockage.
• System not capable of removing consecutive
contaminated packs. When a number of consecutive
contaminated items occur, the reject device must be
capable of accurately rejecting each one, without a
line blockage.
• Failure of the reject to operate to low air pressure,
insufficient air volumetric fault.
• Downstream product back-up through the detector
include build back sensor
• Conveyor speed changed without due consideration for
changing reject timings including encoder.
Rejected Product
Product • Increase belt specifications
One of the benefits of a single-source responsibility for
Punch
unch rejectoperates
reject operates too
tooearly
early conveyor, reject and metal detector is that these issues can
be addressed at the design stage, if necessary.

Rejected Product
Product

31
4
4.2 Satisfying Retailer and 4.3 Inspection of Liquids,

Food Industry Requirements Slurries and Pastes in a Pipeline
The metal detection system design can include simple Inspection of pumped liquids, slurries and pastes can be
additional control devices that will ensure a reject device achieved by replacing a short section of metal the transport
is operating properly, that contaminated packs are pipe with a food-grade non-metallic pipe, and then passing
accurately rejected − and that the metal detection system it through a metal detector (Figure 4.20). The choice of the
is operating in a fail safe mode. Implementation of the pipe will be influenced by:
following design requirements generally represents good • The style of the pipe connection required
practice, and will probably satisfy most brand retailer • Product type and viscosity
and food industry requirements: • Nature of the product
• An automatic reject system to effectively remove product • The product temperature
from the production line • The pipe pressure expected
• A lockable reject bin (with optional bin door lock monitor)
Care should be taken to design the installation so that the
which receives rejected product. Only authorised, trained
pipe will not be under strain from the weight of the in-
personnel should have access to the reject bin. This
coming and out-going stainless steel transport pipes.
helps to avoid the risk that if product is rejected into an
open container and is readily accessible, it can easily When metal contamination is detected, a sanitary three-way
be returned to production in error. valve can operate to divert the contamination. Alternatively,
• A warning device to indicate when the reject bin is full the pump can be stopped and the contamination flushed out
of product manually. The choice of the valve will be influenced by product
• A full enclosure between the detector head and reject bin type (inclusion of product solids) temperature and viscosity.
• An audible and visual indication of system status e.g.
Some valves are best suited to low-viscosity products such
detect head fault, reset fault etc.
as juices etc. If the pipe-cleaning procedure includes a
• A photocell to detect each pack passing through the
cleaning plug being flushed down the pipe, (sometimes
system (to facilitate the correct timing of the reject
referred to as a ‘pig’), the selected valve must be of a
mechanism, irrespective of the position of the metal
‘straight-through’ non-restrictive design.
within the pack)
• An automatic belt-stop fail safe system in response to Typical products suitable for pipeline inspection include
the certain events: liquid chocolate, ice cream, soup and meat slurry.
− Reject bin full This type of application should be considered where the
− Loss in air pressure extra-sensitivity capability of a relatively small aperture
− Reject confirmation system fault etc. outweighs the benefit of final package inspection. This is
• A reject confirmation error highlighting when products are especially true if the final packing material contains metal
not correctly rejected. (as in a canning line).
• R eject check sensor to confirm operation of the photo
gating system Pumped product is seldom totally homogeneous. Voids and
• Drive roller shaft encoder to manage variable speed and bubbles frequently occur.
stop / start systems. Under normal conditions, product passes through the detector
• Build back sensor to prevent products backing up though coils and any product effect tends to cancel out signals, and
the Metal Detection system the detector can be adjusted to give high-sensitivity readings.
It should only be possible to re-start the system via a (Figure 4.20) If, however, a void or bubble appears as it
security password or a key held by a nominated person. passes through the first coil (Figure 4.21), the detector will
Suitable procedures should be in place with respect to any sense a large product difference and a false reject may occur
products in the event of metal detector conveyor stopping with conventional balanced coil metal detectors.
for and reason, all products should be removed and Product Flow Accept
subsequently re-inspected.
A full pipe tends to cancel out product signals

Reject
Figure 4.20

32
4
4.4 Gravity Feed Inspection of Bulk

Powders and Free-Flowing Solids
Any free-flowing powder or granular product (such as flour,
Catch grid valve at end of product flow acts as peanuts, rice, plastic pellets, milk powder, ingredients and
fail safe should the test piece fail to be rejected cocoa beans) can be inspected under free-fall conditions
Figure 4.21 using a gravity-fed free-fall detector and a high-speed
diverter system (Figure 4.23).
With the development of MSF technology and Product Signal
Suppression this effect is reduced and the instance of false Dry Product
Dry Product Flow
Flow
rejects is virtually eliminated.
The product speed in the pipe will ultimately determine the
position of the reject valve relative to the metal detector.
Since the valve has a minimum divert response time,
the distance between the valve and the detector must be
increased, directly proportional to the product speed and the
valve response time.
For product likely to solidify if pumping stops, such as liquid
chocolate, the throughput pipe can incorporate a hot-water-
jacket heating system; this is because electrical heating
wires cannot be passed through a detector.
4.3.1 Fail Safe Design for Pipeline

Systems
The following system design features are generally Good Product
Product Rejected Product
Product
Good Rejected
considered good practice, and will probably satisfy most
brand retailer and food industry requirements: Figure 4.23
• A reject mechanism which can isolate a plug of product

which may contain metal contamination Under normal operational conditions, the product falls under
• Rejection of the contaminated product to a suitable gravity and with the relatively high volumes that can pass
secure container through a small detector aperture delivering, very high
• An audible and visual indication to show that product sensitivity can be achieved.
has been rejected The detector and auto-reject should be mounted on a rigid
• A reject confirmation system that will stop the flow of framework with sufficient space between them to ensure that
product if there is a failure of the reject system. It should metal contamination is always rejected. Consideration should
only be possible to re-start the system via security be given to the design of the reject mechanism with regard to
password or a key held by a nominated person. the potential for product leakage through the reject position.
In certain applications (such as fine powder), there can be
4.3.2 Testing Considerations an accumulation of product dust in the reject device that
Testing access and recovery should be built into the system, can potentially leak out of the reject position, resulting in
so that testing of the detector and reject device can be unacceptable product waste. In such applications, a sealed
performed quickly and reliably. If possible a test sample reject type is recommended.
access port to allow the introduction of a test sample Product flow should be continuous free-fall or batch free-fall.
upstream of the metal detector. Systems of this type are not considered suitable where product
The location and design of the access port should allow is likely to back up in the throughput pipe and move slowly.
the sample to travel at normal speed through the metal The system should have a fixed speed of response,
detection system. There should also be some means of whatever the frequency of operation. The system should
catching the test sample, via a catch grid or open catch also be capable of moving to the reject position more
valve, if it is not detected. quickly than the time taken for a metal particle to fall from
the detector to the reject device.
Frequently, the overall system height is a limitation to the use of
gravity-feed systems, particularly where little headroom exists.
The following limiting variables have a direct relationship on
the overall system height.
Figure 4.22 A Full Pipe Tends to Cancel Out Product Signals

33
4
4.4.1 Initial Fall Height of the Product 4.4.6 Fail Safe Design for
The fall-height of the detector is normally expressed from Gravity-Fall Systems
the point at which the product begins to fall to the top of The following system design features are generally
the detector flange. This height will determine the product’s considered good practice, and will probably satisfy most
velocity at the point of inspection. Ideally, the fall height brand retailer and food industry requirements:
should be reduced to a minimum by locating the equipment
as close as possible to the point of initial fall, without • A reject mechanism which can isolate the actual product
infringement of the ‘metal-free zone’ (see Section 1.4.4). which may contain metal contamination
• An audible and visual indication of system status e.g.
By way of general guidance, the maximum fall-height for a product has been rejected
150mm diameter-aperture detector would be approximately • A reject confirmation system that will stop the flow of
800mm; however, this may vary depending upon the actual product if there is a failure of the reject mechanism.
detector specification. As the fall-height is increased, the It should only be possible to re-start the system via a
distance between the detector and the reject valve must also secure password or a key held by a nominated person.
be increased in order to maintain adequate time for the valve However in some cases a controlled push button reset
to respond. may be the more standard practice.
• Fail safe design i.e. Fail to safety on power outage
4.4.2 Detector Aperture • An audible and visual indication of system status e.g.
product has been rejected
The size of the aperture will determine the operating
sensitivity of the system (subject to the frequency of
operation). The maximum thru put of the system and the 4.4.7 Static Considerations
minimum overall system height. Special technology (such as Falling dry powders and granules can generate static
Zero Metal-Free Zone technology) will keep this distance to a electricity. The build-up of large static charges could have a
minimum. The aperture size will also determine the distance detrimental effect on the performance of the metal detection
that the reject diverter must travel in order to reject product. system or even pose a safety hazard. Some products will
be more prone than others, and environmental conditions
4.4.3 System Response Time (such as humidity) will be a contributing factor. In order to
prevent the build-up of large static charges, the following
This covers the speed of response of the relay or solid-state measures should be considered:
output, air solenoid or air cylinder. It also covers the time
• All metal near the metal detection system (pipes, flanges,
taken to move the reject diverter to the reject position.
supports) should be properly grounded
• Plastic throughput tubes should be made from food-
4.4.4 Reject Angle approved conductive plastic (e.g. FDA approved), which
The reject angle must not be so large that it creates a blockage should be grounded.
or bridges the product. As the length of the reject flap is • The system should have a single-point earth
reduced, the reject angle increases. An angle of between 25º
to 30º is considered a maximum for most products. 4.4.8 Testing Considerations
Testing access and test sample recovery should be built into
4.4.5 Reject Design the system so that testing of the detector and reject device
The speed of response can be slowed by factors such can be performed quickly and reliably. There should be a
as product build-up on the reject device, a drop in air test access port to allow the introduction of a test sample at
pressure and ageing of bearings. A sufficient safety margin the point at which the product begins to fall, so that the test
is needed in the design to ensure that metal is rejected with sample speed will be the same as the product.
100% accuracy. A safety catch grid should be inserted into the normal
product flow below the valve accept position, so that the
test sample can be safely recovered if it is not detected – or
if the valve fails to operate.
The test grid should be capable of being inserted quickly
during test and removed from the product flow afterwards.
The use of a test grid on the reject side of the valve is
recommended; this helps to ease the recovery of the test
sample when it is rejected.

34
4
4.5 Vertical Packaging Applications 4.5.3 Rejection and Fail Safe Design for
Installing a metal detector directly onto, or into, a packaging Vertical Packaging Applications
or process machine, can have a number of advantages for
both the user and the supplier of the original machinery. i.e. The following system design features are generally
• Cost reduction, considered good practice, and will probably satisfy most
• Sensitivity performance increase brand retailer and food industry requirements:
• Overcomes the limitations of metallic packaging • A reject mechanism which can isolate the actual
• No moving parts that wear product which may contain metal contamination. If a
Where installation is required in a restricted space, special reject mechanism is not practicable, the packaging
technology (such as Zero Metal-Free Zone technology) can machine should have the capability of forming a double
be used. This allows metal structures and components to be pack and stop.
positioned very close to the detector without interference. • An audible and visual indication to show that the
packaging machine has stopped.
• A reject confirmation system that will stop the flow of
4.5.2 Vertical Form Fill Seal product if there is a failure of the reject mechanism
or failure of the bag maker to stop. It should only be
A Zero Metal-Free Zone detector can be fitted between a possible to re-start the system via a security password
scale e.g. a multi-head weigher and a vertical form-fill or a key held by a nominated person. However under
seal bag-maker (Figure 4.24). Often, a large detector, controlled conditions a push button reset could be the
e.g. 175mm / 200mm (7” / 8”) is required, which would adopted practice.
demand a larger metal-free zone making installation difficult
or often impossible due to the space needed.
Scale Discharge
Chute
Plastic
Flanged
Funnel
Detector
Plastic Head ‘ZMFZ’
Throughput
Tube
V.F.F.S. Bagmaker
Figure 4.24
Patented Zero Metal-Free Zone technology allows high

sensitivity levels to be maintained (without false rejects) in
a minimum space, making installation easier and avoiding
the danger of product breakage.

150 mm 0.6 to 0.8 mm 0.8 to 1.0 mm 1.0 to 1.2 mm
200 mm 0.8 to 1.0 mm 1.0 to 1.2 mm 1.2 to 1.5 mm
250 mm 1.2 to 1.5 mm 1.5 to 1.8 mm 1.8 to 2.0 mm
Table 4a: Typical Sensitivity Ranges in Vertical Packaging Applications

35
5
Metal Detection, X-ray Inspection or Both Technologies?

The quality and safety of food and pharmaceutical products depends on the due diligence exercised during the
production process, in order to exclude contaminants from finished products.
5
Metal Detection, X-ray Inspection or Food manufacturers must comply with regulations and
standards to prevent contamination, such as the Food Safety
Both Technologies? Modernization Act (FSMA), Global Food Safety Initiative
5.1 The Capabilities of Metal Detection and (GFSI), the British Retail Consortium (BRC), Food Safety
X-ray Inspection Systems System Certification 22000 (FSSC22000) and International
Feature Standard for Food (IFS).
5.2 Product Effects
Pharmaceutical manufacturers have their own compliance
5.3 Packaging Effects
requirements. The choice of protection and inspection
5.4 Which Technology - Metal Detection, equipment hugely affects product quality, product safety and
X-ray Inspection or Both? consumer confidence. Manufacturers must decide whether to
install a metal detection system, an x-ray inspection system,
or both – and this chapter compares the two.
5.1 The Capabilities of Metal Detection

and X-ray Inspection Systems
Metal detection and x-ray inspection systems can be installed
at Critical Control Points (CCP) to inspect in-coming raw
materials prior to processing – or at numerous other points in
the manufacturing process. Inspection systems can also be
installed at the end of the production or packing line.
As discussed in Chapter 1 modern metal detectors can
identify all metals, including ferrous (chrome, steel, etc.),
non-ferrous (brass, aluminium, etc.), as well as both
magnetic and non-magnetic stainless steels in food and
pharmaceutical products.
Metal detection and x-ray inspection systems can be used to
inspect unpackaged or packaged products, including tall rigid
containers such as glass jars, bottles and plastic containers.
X-ray technology is commonly used to inspect metal cans
and products packed in aluminium foil. However, the latest
metal detection technology (See Chapter 3 detailing Multi
Simultaneous Frequency technology) now makes it possible
to inspect products packaged in metallised film and achieve
similar detection levels to that of an x-ray system.
Both metal detection and x-ray inspection systems
can be used to inspect liquids, pastes and slurries in
pipeline applications.

36
For the inspection of gravity-fed, free-falling products, An application area in which x-ray inspection excels, over
metal detection is the only choice. Gravity-fed, powdered traditional metal detection technologies, is when inspecting
or granular products don’t travel at the same speed; they products packed in aluminium foil. Due to the way in which
accelerate as they fall, plus the direction of travel is not the x-ray system works, this type of packaging material has
uniform, since they bounce off each other. X-ray inspection a negligible impact on detection levels.
systems can’t offer a satisfactory solution when handling
this type of product. As the product tends to be dry and 5.3.1 Metallised Film Packaging
non-conductive the sensitivity levels achieved with using a
metal detection system are extremely high. Products packed in metallised film packaging have
historically been inspected by metal detectors using low-
frequency techniques (depending on the film thickness).
5.2 Product Effects However, as mentioned in section 5.2, new technology
The effect of the products being tested will depend upon the is now available from some metal detection companies
chosen inspection technology; both metal detectors and enabling the sensitivities achieved with a metal detector to
x-ray systems have differing inspection capabilities which be very similar to that achieved when using an x-ray system
directly impact sensitivity. and in some cases the detection capability of the metal
detector can be greater.
Historically products with high moisture content have
been more of a challenge for a metal detector to detect In some cases however, if the application allows it is
the smallest contaminants as the signal given off from preferable to inspect these products prior to packing by
the product, known as the product effect has masked the using a “throat” type metal detector, a good example of
signal given off by the metal contamination, however now this is seen in the snack food sector where throat metal
with development of Multi Simultaneous Frequency (MSF) detectors are seen as the solution of choice due to the high
technology this effect is considerably reduced and in many sensitivity levels achieved and the relatively low cost of
cases what was previously considered a challenging investment and ownership.
product is now considered a relatively easy product for
a metal detector to inspect with the result being a vastly 5.3.2 Aluminium Foil Packaging
improved level of achievable sensitivity.
Aluminium packaging, such as foil wraps and product trays,
X-ray inspection technology can also be impacted by are a bigger challenge for metal detectors. Detectors using
certain characteristics of the products being inspected. balanced coil technology are unable to inspect products in
The ease with which contaminants can be identified by aluminium packaging – so a different technology, known as
x-ray inspection depends on various factors such as ’ferrous-in-foil’ detection, must be used. The drawback here
product density, product depth and product homogeneity. is that this technology can only detect magnetic metals and
For example, when using an x-ray system if the product may not be an acceptable solution. This is a good example
includes free salt crystals such as dry crackers and where the choice for using x-ray is clear.
biscuits these salt crystals due to their density can limit the
performance achieved by the X ray system when compared An application area in which x-ray inspection excels, over
to the same product that doesn’t include free salt crystals. traditional metal detection technologies, is when inspecting
Also products with varying texture can affect how the X ray products packed in aluminium foil. Due to the way in which
performs; the more homogeneous the product the better the the x-ray system works this packaging material has a
overall sensitivity. negligible impact on detection levels.
5.3 Packaging Effects 5.3.3 Aluminium Contaminants in

Non-Metal Packaging
The packaging material used can impact upon detection
levels to a differing degree, depending upon the inspection Aluminium is a lightweight metal and a good electrical
technology used. conductor. Since its density is lower compared to other
metals such as ferrous and stainless steel this causes a
A vast array of packaging materials is currently used in the reduction in the sensitivity on an x-ray inspection system.
food and pharmaceutical industries: In these cases, aluminium is detected at twice the size of
• Plastic trays or overwrap ferrous or stainless steel. In contrast, due to its excellent
• Paper conduction properties, aluminium can often be detected at
• Metallised film smaller sizes using metal detection technology so metal
• Aluminium foil detection is generally the better solution.
• Glass
• Metal cans
• Ceramic pots
• Doypacks
• Composite cartons/ tubes

37
5
5.3.4 Metal Contaminants in 5.4.1 Installation and

Non-Metal Packaging Testing Requirements
For cost-effectiveness, metal detectors are the best solution Metal detection and x-ray inspection systems can be
when looking for metal contaminants only once all the supplied with a variety of product-handling devices,
factors above have been taken into consideration. However, including an array of fully-automatic reject devices.
if the requirement is to detect metal contaminants and non- Both metal detection and x-ray inspection systems also
metallic foreign bodies, x-ray systems are the appropriate require regular performance test checks to be carried out
solution. If there are any doubts a product test would at prescribed intervals. Installation, commissioning and
always be advisable. training is explained in more detail in Chapters 11 and 12.
However, recent advances in metal detection technology
5.3.5 Non-Metal Contaminants in such as the development of Predictive Analytics, makes it
possible to extend the interval between scheduled tests of
Any Packaging the metal detector. This in turn can make it very attractive
X-ray inspection is the only solution, and has the ability to to the user as it gives potential for an increase in the user’s
detect non-metallic contaminants such as glass, mineral OEE percentage.
stone, calcified bone, and high-density rubber and plastic.
5.4.3 Fast/Variable Line Speeds
5.3.6 Product Size Limitations Metal-detection and x-ray inspection systems are both
Both x-ray and metal detectors can be designed to suitable for variable and fast production lines. Metal
accommodate any product size. For larger packs and detectors will detect contaminants in products moving
products, the aperture height or opening must be increased at low and high speeds, including conveyors running at
and, as a general rule, the larger the aperture height and speeds above 400m/min (although very few conveyorised
product, the lower the sensitivity. processes run at such high speeds).
X-ray inspection systems can monitor conveyor lines
5.4 Which Technology – Metal running at up to 120m/min. Even higher inspection
volumes/ speeds can be achieved in pumped and bulk
Detection, X-ray Inspection or Both? applications for both metal detection and x-ray technology.
Metal detection and x-ray inspection offer different The choice of technology is dependent on multiple factors
capabilities – and in order to assess those capabilities fully, such as types of contaminant, product type and packaging
the first step is to carry out a Hazard Analysis and Critical material; speed generally isn’t a deciding factor.
Control Points (HACCP) audit. This will help to understand
the requirements of any customer or compliance-related 5.4.4 Limited Space
issues driven by the GFSI and/or major retail groups.
A metal detection Detector Head takes up less space than
A HACCP audit will identify the risks of contamination being an x-ray inspection unit, so in situations where installation
introduced into the manufacturing process, and the types space is limited and metal is the likely contaminant, a metal
of contamination likely to be encountered. CCPs should be detector may be the best solution. If packed products are
established to mitigate the risks, and product inspection being inspected, both systems will normally need a conveyor
equipment needs to be installed at these points to reduce system and an automated reject system. In some situations,
the risk of contamination to acceptable levels. the differences in overall system length can be very small.
Some metal detector companies offer what is referred to as
If the HACCP audit determines that metal is the only likely
Zero Metal Free Zone (ZMFZ) technology. This allows the
contaminant to be found, then a metal detector is the best
overall size of the metal detection system to be drastically
solution. Likewise, if other contaminants like glass, mineral
reduced and it is common place to find metal detection
stone, calcified bone or high-density plastics and rubber are
systems that take up less than 1000mm of line space.
identified as likely to be encountered, then x-ray is a more
suitable solution. Recalling the packaging and product effects
mentioned previously, it would be advisable to conduct
product testing to establish the most appropriate technology.
In many cases, there’s only one suitable solution – either
metal detection or x-ray inspection. However, there are also
occasions when it could be necessary to install both metal
detection and x-ray inspection at different CCPs on the
same production line.
HACCP and its principles are explained in the Chapter 9
Selecting Control Points).

38
5
5.4.5 Industry Standards and

Codes of Practice
Recent changes in food and pharmaceutical industry
safety standards are resulting in the increased adoption
of metal detection and x-ray inspection systems by
manufacturers. A growing number of major retailers are
setting their own codes of practice, which contain specific
advice regarding product inspection equipment based on
the Global Food Safety Initiative (GFSI), the British Retail
Consortium (BRC), Food Safety System Certification 22000
(FSSC22000) and International Featured Standard for Food
(IFS). In addition, pharmaceutical manufacturers have their
own compliance requirements.
5.4.6 Simplifying the Choice

Following the development of both metal detection and
x-ray inspection technologies the choices are no longer
made by simply choosing one or the other. This chapter is a
good starting point for choosing an appropriate technology,
but it can’t provide all the answers. There’s often an area
of indecision that requires further levels of discussion with
product inspection experts.
If cost is the sole criterion for deciding, metal detection is a
more suitable solution. However, product safety decisions
are rarely that simple. The performance of each solution
is affected by the size of the product to be inspected, plus
it’s important to compare lifetime costs, not just the upfront
capital costs.
The type of product and the likely contaminants will also
affect the choice; consideration must be given to the HACCP
audit and CCPs on the production line. Sometimes, the
answer is to install more than one detection system at
different CCPs on the same production line.
For example, a metal detector or a bulk flow x-ray
inspection system placed early in the processing line can
remove large metal or non-metallic contaminants before
they reach delicate machinery downstream, where they
could damage the machine or become fragmented into
multiple, smaller, more difficult to detect contaminants.
An x-ray inspection system can be installed at the end of
the line to detect a wider range of contaminants and to carry
out other quality checks, such as confirming pack integrity
and checking the contents of the pack before the product
leaves the factory.
There is an area of overlap between the two technologies
where you could choose either one. Then it’s not so
much a question of which technology is better, but
which technology is most appropriate for your particular
application and budget.

39
5
5.4.7 Summary Table

The following table summarises the key differences between the two technologies:
Metal Detection X-ray Inspection
Product formats Packaged, conveyorised products, Packaged, conveyorised products, loose,

loose, bulk products, free-falling and bulk products, pumped liquids, pastes
vertically-packed products (including and slurries, continuous web products
powders and granular products),
pumped liquids, pastes and slurries,
continuous web products
Contamination detection Detection of all metal contaminants, Detection of dense contaminants like
including ferrous, non-ferrous (including ferrous, non ferrous and stainless-steel,
aluminium) and magnetic and non- as well as other contaminants like
magnetic stainless steels glass, stone, bone, high-density plastics
and some rubber compounds
Detectable contaminants Contaminants must be austenitic Contaminants must be high-density

(magnetisable) or electrically conductive or have a high atomic mass number
Aluminium contaminants Easily detected Detectable, but not as easily detected

as other metals
Quality checks Detection of metal contaminants Detection of dense contaminants and

simultaneous quality checks for mass
measurement, seal inspection, fill-level
control, component count,detection of
missing and broken products, as well
as packaging
Product texture No effects May limit performance
Conductive product Can be inspected Can be inspected
Metallised film-packed products Can be inspected Can be inspected
Aluminium foil-packed products Cannot be inspected effectively Can be inspected
Pack size effects The larger the pack, the less sensitive The larger the pack, the less sensitive
Increased aperture size Sensitivity can decline, and costs Sensitivity can decline, and costs
increase moderately increase significantly
Short Conveyor length Short conveyor lengths or space Short conveyor length may need special
required for insertion guarding for radiation safety
High line speeds Operates at high line speeds Operates at high line speeds
Variable line speeds Operates at variable line speeds Operates at variable line speeds
Gravity Fed Production Can be inspected Cannot be inspected
Table 5a

40
5
Notes

41
6
Reasons for a Metal Detection Programme

For many companies, the purchase of a metal detection system can represent a significant capital investment,
so it is important that the equipment is reliable, appropriately designed for the intended application – and used in the
most effective way. Adherence to these principles will ensure that the metal detection system generates a good Return
On Investment (ROI) by minimising costs and maximising product safety.
6
Reasons for a The most effective use of a metal detection system is as
part of a wider-reaching metal contaminant reduction
Metal Detection Programme programme designed to achieve several aims:
6.1 Minimising Metal Contamination 1. To detect contamination in product, and…
6.2 Minimising Costs 2. To carry out appropriate preventive actions when
contamination is detected, and…
6.3 Protection of the Customer and Consumer 3. To take appropriate measures which prevent metal
6.4 Protection of Brand and Reputation contamination occurring in the first place
6.5 Certification The justification for purchasing a well-designed metal

detection system (and the reasons for its implementation)
6.6 Employee Buy-In can be demonstrated by considering the following benefits,
6.7 Due Diligence and Regulatory Compliance which are discussed in detail in this chapter:
6.8 Retailer and Consumer Brand Codes • Minimising metal contamination
• Minimising costs
6.9 References • Protection of the customer and consumer
• Protection of brand and reputation
• Certification
• Employee buy-in
• Due diligence and regulatory compliance
• Conforming with retailer and consumer brand codes

42
6.1 Minimising Metal Contamination 6.3 Protection of the Customer
Metal contamination can still be a cause of consumer and Consumer
complaints, even when metal detection systems are used.
However, such complaints are not normally due to the metal Although modern manufacturing techniques constantly
detection system failing; they are usually associated with strive to eliminate the occurrence of metal contamination in
lack of effective controls, poor working methods, incorrect products, there will always be occasions where processes
system specification and incorrect design. or procedures break down, resulting in its occurrence.
Many incidences of metal contamination do not result from Manufacturers and their employees have an obligation to
the presence of tiny metal pieces but from the presence customers and end consumers to minimise instances of
of much larger items such as washers, bolts, and pieces contamination; they are also obliged to ensure consistent
of blades or screens. These kinds of items should be quality is maintained and that all possible steps are taken to
detectable by even the most basic type of detector. protect the welfare of the end user.
A well designed metal detection programme should be Failure to achieve these aims can create potential animosity
able to address these wider-ranging issues − although between manufacturer, retailer and customer; they can also
the programme should also focus on how to minimise result in the potential breakdown of the customer relationship,
instances of contamination in the first place. as well as the loss of future business opportunities.
Contamination prevention can be achieved through factors

such as: 6.4 Protection of Brand
• Good manufacturing practice and Reputation
• Prerequisite programmes
Powerful product branding gives customers a perceived
• Selection of the correct equipment and the use of certified
assurance of safety and quality. Effective, visible and
test samples
memorable branding is frequently responsible for driving
• Effective testing
consumer repeat purchases − and so is an important tool
• Gaining a greater understanding of how industry
in maximising sales and justifying premium product pricing
standards, customer requirements and legislation impact
charged by manufacturers and retailers.
on manufacturers
For this reason, an organisation’s responsibility is not only
related to protection of the end user, but also to the brand and
6.2 Minimising Costs to the on-going reputation of the company. Product brands
The costs associated with implementing and maintaining an are important assets that need to be managed carefully, and
effective metal detection programme are significantly lower must be protected, at all costs, from adverse publicity.
than the potential costs of failure to do so.
Contaminated product found by consumers can have a
A metal-contaminated product found before shipment will serious negative impact on any organisation, resulting in
inevitably result in costly product and packing wastage, damage to the brand and potentially costly recalls. In the event
possible machinery damage, and loss of output. Whilst of a company being investigated as a result of a customer
costs can easily be allocated to such events, they can be complaint, documentation will provide invaluable evidence of
particularly high when there is loss of output − particularly the correct operation of the metal detection programme.
on high-volume automated production lines.
However, even these costs can be overshadowed by 6.5 Certification
instances of contamination discovered after shipment; these
It is highly likely that metal detection systems will be the
can result in loss of customer satisfaction, product recall,
focal point of any customer/retailer audit because of their
adverse publicity, and potential legal action.
contribution to the safety of the manufacturing process. In
Time and money spent reducing the incidence of metal addition, the presence of metal detection systems provides
contamination in the first place will result in less internal evidence of good manufacturing and product safety
waste, reduced loss of output and fewer complaints. This practice. Such evidence will undoubtedly be requested (if
will yield a better return than money spent responding to not immediately, at some time in the future) by any one of a
contamination after it has occurred, and dealing with its number of audit processes, such as:
many costly consequences. • Internal food safety and management system audits
As well as leading to reduced incidence of contamination and • Customer audits
reduced failure costs, a correctly implemented metal detection • Quality management system audits e.g. ISO9001:2000
programme will undoubtedly lead to improved customer and • Food safety management system audits e.g.
consumer satisfaction, as well as greater profitability and ISO22000:2005, SQF1000/2000 Code
better protection of the manufacturer’s brand. • Regulatory audits e.g. FDA, USDA, International Food
Standard (IFS), British Retail Consortium (BRC)

43
6
6.6 Employee Buy-In 6.8 Retailer and Consumer

Formalised procedures and working practices relating to Brand Codes
product safety and brand protection will help support overall
quality throughout the manufacturing organisation. These Major retailers, and custodians of leading consumer
procedure and working practices can be reinforced by key brands, have also played a major part in developing their
customer training sessions and awareness seminars. Such own codes of practice – and these need to be adhered to
training can be organised and run by the metal detector in order to satisfy supply agreements. Such standards can
manufacturer or appointed representative. vary considerably across geographical territories – and,
increasingly, the implementation of a formal metal detection
programme is expected before supplier approval is granted.
6.7 Due Diligence and
Regulatory Compliance 6.9 References
Currently, there is no broad-based legal requirement Links to various sources and types of information are
which forces manufacturers to install metal detection included below for reference:
equipment or implement a metal detection programme.
ANVISA − Brazilian Health Surveillance Agency
However, where proceedings result from the presence of
metal contamination in a food or pharmaceutical product, http://www.portal.anvisa.gov.br/wps/portal/anvisa-ingles
manufacturers could be called upon to prove that they have British Retail Consortium (BRC)
exercised due diligence in their processes; failure to do so
http://www.brc.org.uk
could result in serious consequences.
Due diligence is easier to prove when an organisation has a Codex Alimentarius
documented system which continually assesses the risks to http://www.codexalimentarius.net
food safety and allocates resources to minimise these risks.
European Food Safety Authority (EFSA)
In the absence of any definitive metal detection legislation, http://www.efsa.eu.int/
several regulatory bodies have emerged with standards and
codes of practice to which manufacturers can adhere. These Food and Agriculture Organisation (FAO)
codes advocate universal inspection of all food and allied of the United Nations
products by metal detection equipment. A few examples are http://www.fao.org/
shown below.
Food Safety Inspection Service (FSIS)
‘Every company needs to perform a hazard analysis for
every product it produces to assess the risk of metal http://www.fsis.usda.gov
contamination in their products. If the hazard analysis Food Standards Agency (FSA)
shows there is risk of metal contamination, then a metal
http://www.food.gov.uk/
detector will be required.’
BRC Global Standard Guidelines Food Safety Modernization Act (FSMA)
http://www.fda.gov/food
‘Effective measures shall be taken to protect against the
inclusion of metal or other extraneous material in food. Global Food Safety Initiative (GFSI)
Compliance with this requirement may be accomplished http://www.mygfsi.com
by using sieves, traps, electronic metal detectors,
or other suitable effective means.’ Food and Drug Hazzard Analysis and Critical Control Points (HACCP)
Administration (FDA) http://www.fda.gov/food/guidanceregulation/HACCP/
GMP: 21CFR 110.80(b)(8)
International Committee of Food Retail Chains (CIES)
Some of these standards are beginning to play a part in http://www.life-sciences-france.com/
supplier selection and the specification of metal detection
standards for manufacturers. They generally require control
in the form of a documented programme covering protective
and safety-related actions.

44
6
International Features Standards (IFS)

http://www.ifs-certification.com
ISO 22000:2005 - Food Safety Management
System Standard
http://www.lrqa.co.uk/products/otherproducts/iso22000/
Safe Quality Food Institute SQF Code
http://www.sqfi.com/standards
United States Department of Agriculture (USDA)
http://www.usda.gov/wps/portal/usdahome
United States Food and Drug Administration (FDA)
http://www.fda.gov
World Health Organisation (WHO)
http://www.who.int/en/
World Food Safety Organisation
http://www.worldfoodsafety.org/

45
7

When a manufacturing organisation has decided to implement a new metal detection programme, or improve upon an
existing metal detection programme, it is important to ensure that the plan is initiated as efficiently and effectively as
possible. This chapter provides practical guidance on how to build an effective programme.
7
Building an Effective Programme 7.1 Programme Requirements
The decision to adopt a metal detection programme needs
7.1 Programme Requirements to be a major strategy for the organisation; otherwise,
there is a risk that it will lose its importance and will not be
7.2 Key Elements and Controls effectively maintained.
7.3 Documenting the Programme The design and implementation of the programme should
7.4 Competence, Awareness and Training be governed by:
• The varying needs and objectives of the organisation
• The product range
• The processes employed
• The size and structure of the organisation
The programme needs to be proactive rather than reactive, and
should be used to prevent the occurrence of contamination in
the first place, rather than simply detecting it when it occurs.
The aim should be to maintain control over the complete
process, from ascertaining the quality of supplied ingredients
through to dealing with customer and consumer complaints.
7.2 Key Elements and Controls

It is important that those responsible for defining and
documenting the metal detection programme should have a
good understanding of the basic principles of operation and
equipment capabilities. This will help to avoid disappointment
in performance once the equipment is operational (Refer to
Chapters 1 to 4 for further information).
If the correct metal detection solution is not identified in the
first instance, subsequent efforts to implement it may be of
little benefit.
Having understood the basic principles of operation, and having
selected the best metal detection solution for the application, it
is important to understand the wider issues and key elements
that need to be implemented if the programme is to be effective.
The specific controls contained within the programme
should be based on an analysis of hazards and frequency
of occurrence; the nature and size of the business should
also be taken into account.

46
Table 7a highlights the key elements (and chapter references 7.3.2 Responsibilities and Authority
covering this guide) that review the requirements in greater detail:
To ensure the effective operation and maintenance of its
metal detection programme, management should ensure
Key Element Chapter
that responsibilities are clearly defined and communicated
within the organisation.
Prevention of Metal Contamination 8
All company personnel should be responsible for reporting
Selecting Control Points 9 potentially hazardous situations associated with effective
operation of the metal detection programme; they should
Operating Sensitivity 10 also know to whom they should report these occurrences.
Installation and Commissioning 11 7.3.3 Documented Procedures

In order for the programme to be effective, procedures
Performance Validation, 12
should be:
Verification and Monitoring
• Appropriate to the organisational needs of the facility,
Dealing with Suspect and Rejected Product 13 with application to food safety
• Appropriate to the size and type of the operation
Data Analysis and Programme Improvement 14 • Appropriate to the nature of the products being
manufactured or handled
Data, Connectivity and Improving Performance 15 • Implemented across the entire production system, either
as generally applicable programmes or as programmes
Table 7a: Elements of an effective metal detection programme applicable to a particular product or production line
• Approved by those responsible for food safety
7.3 Documenting the Programme 7.3.4 Records

The metal detection programme should be documented
Records should be established and maintained to provide
as a set of controlled policies and procedures. The scope
evidence of compliance with requirements; these records
and detail of these procedures should match the size
should also provide evidence of the metal detection
and complexity of the organisation, whilst also being in
programme’s effective operation.
accordance with the organisation’s lines of communication.
Records should remain legible, readily identifiable and
For a small organisation, it may be possible to establish
retrievable, regardless of whether they are in hard copy or
all necessary controls in one single operational procedure;
electronic format.
in a larger organisation, it may be advisable to integrate
the requirements into the existing food safety or quality A documented procedure should define the systems needed
management system. for the proper maintenance and control of records, covering:
The most effective metal detection programmes are • Identification
established, documented, operated and maintained within • Storage
the framework of a structured food safety management • Protection
system. This should be supported by the overall • Retrieval
management activities of the organisation. • Retention time
• Disposal
Relevant and meaningful documentation is crucially
important if a company is being investigated as a result
of a customer complaint. Under these circumstances, 7.4 Competence,
appropriate documentation will provide evidence of safety
measures employed within production processes.
Awareness and Training
Personnel carrying out activities that have an impact
upon the effectiveness of the metal detection programme
7.3.1 Metal Detection Policy should demonstrate a level of competence based on their
Senior management should define and document a education, training, skills and experience.
company’s metal detection policy. The policy should be:
Personnel should be aware of the relevance and importance
• Appropriate to the role of the organisation, in relation to of their activities; they should also be aware of how those
its position in the food chain activities contribute to the achievement of food safety.
• Supportive of applicable regulatory, retailer, customer or
corporate food safety and quality requirements Appropriate records of education, training, skills and
• Communicated, implemented and maintained at all levels experience should be maintained.
within the organisation
• Reviewed for continued suitability
• Supported by measurable objectives
• A source of what actions to take in the event of product
rejection and metal detection system faults

47
8
Prevention of Metal Contamination

Every piece of metal prevented from entering the production process represents a 100% success story.
However, no detection system can ever reach a 100% level; contaminated ingredients, lack of preventative
maintenance, poor working practices during installation and daily operations are all common causes of metal
contamination, so this chapter provides practical guidance on how to prevent metal contamination in the first instance.
8
Prevention of Metal Contamination 8.1 Contaminated Ingredients
Inspecting suppliers’ raw materials when they arrive at
8.1 Contaminated Ingredients the production-line facility will eliminate many large,
easily detected pieces of metal before they are broken into
8.2 Maintenance Procedures numerous smaller and more difficult-to-detect fragments by
8.3 Good Manufacturing Practice the production line process.
Further large pieces of metal (in the form of broken blades
and other major items) can also enter the food production
process on the manufacturer’s own production line, causing
additional metal contamination.
Therefore, the best approach is to impose two distinct
metal detection programmes – one on the manufacturer’s
own production line, and the other governing all those raw
materials delivered by external suppliers.
Each supplier should therefore take full responsibility for the
quality of the products they supply by operating their own
effective metal detection programme.
Supplier agreements or individual ingredient specifications
should clearly state applicable operational sensitivity
standards, as well as any other specific precautions that the
supplier should take. These will depend on the product type,
so can include instructions such as:
• Material in powder form to be passed through a metal
detector system
• Carcass meat not to be labelled with metal tags
• No stapled containers to be used

48
8.2 Maintenance Procedures 8.2.1 Planned Preventive
There is an inherent risk of metal contamination every time Maintenance Programme
a product is transferred from one production process to The planned preventive maintenance programme should
the next. Items such as crushers, mixers, blenders, slicers, aim to limit the wear and tear on equipment that might
sieves and transport systems are all potential sources of otherwise result in metal contamination, or contribute to a
contamination if they are not properly maintained. reduction in on-line performance. For such a programme
There is also the potential for creating contamination when to be effective, the degree and frequency of maintenance
conducting maintenance routines or when performing new activity should be based upon:
installations. Therefore, the importance of carrying out • Plant breakdown history
preventive maintenance under controlled conditions is essential • Equipment manufacturer’s recommendations
to the effective operation of any metal detection programme. • Lubrication requirements
• Importance of the equipment in the manufacturing process
Procedures for maintenance should ensure the following:
• Risk assessment of critical locations where metal
• Product safety and quality is not jeopardised during contamination might occur
maintenance operations and installations • Equipment known to be vulnerable to wear and tear,
• A documented, company-wide, planned maintenance e.g. bearings, slicer and mincer blades, mixing vessels,
programme is in place sieves etc.
• Instructions are available to maintenance personnel • Predictive modelling (where appropriate)
indicating what is to be done during planned maintenance
(including strip-down and re-build procedures)
• Personnel are trained with regard to these instructions. 8.2.2 Documentation and Records
This training should be provided by the equipment Records of maintenance undertaken (and any subsequent
manufacturer or by the organisation’s own staff who corrective actions) should be recorded. This information can
have been trained by the manufacturer be used to good effect when reviewing the effectiveness of
• All outside contractors and engineers are made aware of the planned maintenance programme and incident resolution.
(and adhere to) the company’s manufacturing practices
and hygiene standards Ideally, the maintenance status should be indicated on
• Arrangements for ensuring jobs are raised and completed the equipment itself for maximum visibility. Typically, the
on time, and are highlighted if they are not carried out for information should include the date last checked, who
any reason made the check, and when the next check is due.
• A full test of all applicable systems is carried out
following repairs, maintenance or adjustments 8.2.3 Good Engineering Practice
• Provision is made for the management of spare parts
Pieces of metal, e.g. swarf, metal filings etc. can be
and replacement equipment
produced when repairing, modifying or installing
It is important that potential hazards (such as defective equipment. There is always the risk that this metal may
machinery) are reported as soon as identified, so it needs get into (and contaminate) the product. However, this risk
to be clear to whom such instances should be reported. can be significantly reduced if maintenance personnel are
Once feedback concerning hazards has been received by trained in food safety and hygiene, and if the work is carried
the relevant parties, it is important that necessary remedial out in accordance with good engineering practices.
action is promptly taken.
The following constitute good engineering practice:
In addition, maintenance procedures need to be reviewed • Wherever possible, engineering work should take place
in light of these recently reported events − especially with outside production areas − preferably in the engineering
regard to making appropriate revisions so that they do not workshop. Welding, drilling, riveting and soldering
occur again. This process will keep necessary procedures should never take place on equipment being used for
and work practices live and effective. production; neither should such processes take place
on any equipment immediately adjacent to production
equipment, unless suitable hygienic screening is in
place. For major work or new installations, complete
floor-to-roof screens may be necessary.
• Workshops should be kept clean and tidy by being
swept or vacuumed at least daily, with a ‘clean as
you go’ approach being the preferred methodology.
Engineering spares and equipment should be stored
above floor level to allow for cleaning access. The
equipment used within the workshop should be
maintained in good working condition, and should be
subjected to the same regular cleaning.

49
8
• Using the appropriate methods (such as magnets, • Paper clips and staples should not be used on
vacuum cleaners etc.), any equipment that has been documents in production areas
maintained or repaired in the workshop should be • Pins should not be used on notice-boards
thoroughly cleaned to remove all debris, before being • No hair clips, watches, jewellery should be allowed in
returned to the production area. production areas (sometimes exceptions can be made
• If the workshop is within the production environment, a for plain wedding bands)
suitable foot-scraper mat (or other similar trap) should • Protective clothing should have no outside pockets
surround the workshop, accompanied by a clear notice • Only ‘metal-detectable’ plasters or wound dressings
which requests staff to scrape their footwear before should be used by personnel, to aid detection if lost in
leaving the workshop. production processes
• Personnel carrying out repairs on production lines • Only ‘metal-detectable’ pens, hairnets, ear-defenders
should be provided with an enclosed tool-box for tools, and ancillary equipment should be used to aid detection
nuts, bolts and screws etc. Magnetic trays or other of lost items
clearly marked containers should be used to contain • Product-holding containers should be covered at all times
fixings and other items removed or replaced during • Conveyor lines carrying open containers should be
engineering work. The tool boxes should be kept clean covered until the containers are closed or capped.
and free of any unnecessary contents which could be
hazardous to production.
• When repairs, installation and commissioning have been
completed in the production area, the equipment and the
surrounding area should be independently inspected, so
as to confirm that cleaning has occurred in accordance
with agreed procedures. Documentation should be in
place indicating that designated personnel have checked
that production lines are clean and that production may
re-commence ( i.e. use of a ‘positive release’ system).
• Tape or wire (i.e. temporary engineering solutions) should
not be used to repair equipment. Damaged fittings and
missing or loose screws should be repaired promptly and
permanently. Any metal debris (along with other potential
contaminants) should be disposed of safely and promptly.
Missing fixings on equipment should be accounted for
and/or replaced. Use should be made of ‘nylock’ nuts, or
similar secure fixings, wherever possible (a nylock nut
includes a nylon collar screw-fitting insert).
• Wherever possible, nuts, bolts and washers, sieve mesh
etc. used on food processing equipment should be made
from magnetic stainless steel.
8.3 Good Manufacturing Practice

Personal effects and operational items, such as tools and
parts, present a real contamination risk if there is poor
awareness and lack of good working practices. Time spent
identifying potential risks, defining good working practice
and provision of correct equipment will be rewarded by
minimising risk of contamination.
Clear, concise staff policies should be implemented and
communicated on a regular basis to ensure that personnel
remain informed about correct procedures – and to ensure that
they put these procedures into practice on a continuous basis.
Listed below are examples of what constitutes good
manufacturing practice. There are, undoubtedly, many
more specific control measures that are relevant to specific
industries, companies and manufacturing processes;
however, the following practices effectively demonstrate
risks that could easily be overlooked:

50
8
Notes

51
9
Selecting Control Points

HACCP (Hazard Analysis and Critical Control Points) is a systematic and preventive approach to protecting products
from biological, chemical and physical hazards. This protection is provided within the context of production processes
that can cause the finished product to be unsafe; HACCP offer guidance on how to reduce risks to a safe level.
9.
Selecting Control Points HACCP techniques are considered to be a key contributor to
the establishment of an effective metal detection programme
– and a Hazard Analysis can assist greatly in identifying
9.1 Conduct a Hazard Analysis potential sources of contamination.
9.2 Determine Critical Control Points (CCP) In addition, Hazard Analysis provides information needed to
establish necessary inspection points, whilst also providing
9.3 Establish Control Limits
guidance as to the best metal detection solution for the risks
9.4 Establish Monitoring Processes that have been identified.
9.5 Establish Corrective Actions This chapter does not attempt to teach the fundamental
principles of HACCP; instead, it aims to give practical guidance
9.6 Establish Documented Record Keeping Procedures
on where to use metal detection systems and how to use this
9.7 Verification guide to support the HACCP process. Some links to useful
9.8 HACCP Reference Sites HACCP information sites are included at the end of this chapter.
Conduct Hazard Analysis
Determine Critical Control Points
Establish Control Limits
Establish Monitoring Processes
Establish Corrective Actions
Establish Documented Record Keeping Procedures
Verify HACCP system is working as intended
Table 9a: Key HACCP 7 Analysis stages

52
9.1 Conduct a Hazard Analysis As a minimum, the end of every production line should be
considered as a Critical Control Point. The ideal point is
Every company needs to perform a Hazard Analysis for every immediately after the packing point, in-line with the main
product it produces, in order to assess the risk of metal production flow. Thus the potential for metal inclusion
contamination occurring during the production process. afterwards is significantly reduced.
Good practice requires that all hazards that may be reasonably If the above Critical Control Point is not possible, the metal
expected to occur (including hazards associated with the detection system should be located as close as possible to
processes and facilities used) are identified and assessed. the finished packing point, in-line with the main production
A thorough Hazard Analysis should identify potential flow. In such instances, the producer may be required to
sources of contamination and metal types likely to be obtain approval from their customer.
found. Such information will assist greatly in selecting In situations where it is impractical to carry out finished
the correct metal detection system – so, for example, if a pack metal detection (such as canned foods), alternative
producer makes beef and onion pies, the Hazard Analysis control systems must be in place (see Chapter 4) and
may show a potential risk from: agreed with the customer. These kinds of control systems
• Broken cutting blade contamination from the meat and include pipeline metal detection/rejection systems.
onion preparation If good product is to be manually removed from the
• Mixing blade contamination from the meat sauce and conveyor for hand-packing processes, a suitable clear
pastry mixing guard should be used to cover the area from the detector
• Filling heads falling into the pies through to the reject point. The guarding should extend
• Swarf from aluminium food trays sufficiently along the length of the conveyor and reject
• Sieve damage from dry ingredient lines system to prevent operators from removing items for
This is just a simple example of how Hazard Analysis can packing before they have passed the reject device.
identify potential contamination risks; it should highlight
the type of metal that could cause potential contamination.
Where the potential for metal contamination is identified and
9.3 Establish Control Limits
a metal detection system is defined as the necessary control Having identified the Critical Control Points, it is important
measure, then this should be considered a Critical Control to define critical limits. In the case of the metal detection
Point (CCP) and should be included in the HACCP plan. system, these relate to the operating sensitivity, the
operation of the reject mechanism and any built-in fail safe
features. Chapter 3 of this guide explains the factors limiting
9.2 Determine Critical sensitivity, whilst Chapter 10 explains how to define and
Control Points (CCP) document the actual operating sensitivity standard.
When determining the Critical Control Points (CCP),

consideration should be given to identifying and rejecting 9.4 Establish Monitoring Processes
the contamination as early as possible within the Having established the operating sensitivity limits, it is
manufacturing process. Such an approach is consistent important to periodically verify how well the metal detection
with good manufacturing practice and HACCP programmes. system detects and rejects contaminated product at (and
HACCP does not rely solely on end-product testing to ensure above) the operating sensitivity standard. Chapter 13 of
that the food is safe. Instead, it builds food safety into the this guide provides practical guidance on how to define the
manufacturing process and relies on process controls to appropriate test and audit routines.
prevent or reduce the presence of known food product
hazards to an acceptable level.
9.5 Establish Corrective Actions
If contamination is knowingly allowed to travel through the
If the monitoring process identifies that the Critical Control
manufacturing process, there is a danger that it may cause
Point is not operating to the agreed critical limits, then there
damage to downstream processing equipment. Alternatively,
needs to be a clearly defined process for corrective action.
travel through the manufacturing process could make the
Chapter 14 of this guide provides guidance on what actions
metal pieces even smaller, making it more difficult to detect
should be taken if there is a failure of the metal detection
further down the production line. These circumstances
system or if metal contamination is detected.
could result in higher than necessary scrap costs due to the
amount of processing that has gone into the product by time
it reaches the end of the production process.
Inspection requiring additional handling is never totally secure.
Whenever possible, the metal detection system should be
integrated in-line with the normal production flow. This avoids
possible confusion over what has been inspected; it also
prevents any bypass of the inspection process.

53
9
9.6 Establish Documented Record Finally, your HACCP plan should be up-to-date at all times
and reflect any change. A change is anything in the HACCP
Keeping Procedures plan that is different to when the study was last carried out.
A review should be both scheduled and triggered. Triggers
Efficient and accurate record keeping is essential to the
could include but not be limited to:
application of a HACCP system. According to the Codex
guidelines, documentation and record keeping should be • Changes in raw materials
appropriate to the nature and size of the operation and • Introduction of new product to the line
sufficient to assist the business to verify that the HACCP
controls are in place and being maintained. Although it • Change of raw materials supplier
requires considerable effort, the record-keeping program • Modification to the layout, environment or equipment
gives references available to trace the production history
of a finished product. Records of validation and verification • Product recalls or changes in legislation.
studies should be kept as evidence that they have been
carried out correctly. Such records can be used as a tool 9.8 HACCP Reference Sites
to alert the operator to potential problems before they lead
to the violation of a critical limit and records can serve as Dutch HACCP
evidence that proper procedures are being followed. http://www.foodsafetymanagement.info
Food Standards Agency
9.7 Verification http://myhaccp.food.gov.uk/
HACCP needs to be considered in the context of Validation
HACCP Principles Side by Side
and Verification of the metal detection system. ‘Validation’
can be defined as the assessment of whether the plan or http://www.fsis.usda.gov/OPPDE/rdad/FRPubs/00-014R/
operation is scientifically and technically sound; it also HACCPPrinciples.pdf
confirms that the hazards, critical limits, monitoring, and USDA Seven HACCP Principles
corrective actions have been correctly established. Re-
http://www.fsis.usda.gov/oa/background/keyhaccp.htm
validation is often necessary when manufacturing methods
or processes change. i.e. Will the HACCP plan ensure US FDA HACCP Guide
that safe food will be produced”? The original equipment http://www.cfsan.fda.gov/~lrd/haccp.html
manufacturer (or their representative) can offer valuable
support in the validation process, ensuring that the correct WHO / CODEX HACCP
equipment specification is defined (see Chapters 2 to 4), as http://www.who.int/foodsafety/fs_management/haccp/en/
well as providing professional installation and commissioning
services (Refer to Chapter 12 for further information).
The HACCP plan and its operation should be verified annually
‘Verification’ can be defined as the assessment of whether the
plan is being correctly adhered to, as well as confirmation
that the monitoring procedure is being followed. This relates
to the application of methods, procedures and tests (plus
other evaluations in addition to monitoring) to determine
compliance with the agreed plan. The original equipment
manufacturer (or their representative) can offer advice and
guidance on the best ways to verify the metal detection
system and provide verification services i.e. “Is the HACCP
plan working, is it producing safe food”? (Refer to Chapter 12
for further information).
If a specific type of contamination is common, this should be
discussed with the detector manufacturer (preferably during
a site visit), together with all other relevant information, since
this may have a bearing on the type of detector most suited
to the application.

54
9
Notes

55
10
Operating Sensitivity
The factors limiting sensitivity have been explained in Chapter 3, so this chapter highlights the need for
maximum operating sensitivity; it also provides practical guidance in relation to defining company operating
sensitivity standards.
10
Operating Sensitivity 10.1 The Need for Maximum
Operating Sensitivity Performance
10.1 The Need for Maximum Operating Ideally, metal detection systems should be set for maximum
Sensitivity Performance sensitivity performance; at the same time, they should remain
10.2 Establishing the Operating Sensitivity Performance stable and reliable in order to ensure optimum consumer
protection. The overall aim should be to consistently improve
10.3 Establishing a Sensitivity Standard
metal detection capabilities wherever possible.
10.4 Documenting the Sensitivity Standard
Slight reductions in operating sensitivity can have a
significant effect on the performance of a metal detection
system – though this is a fact seldom appreciated by
many users.
If wire is identified as a potential contaminant, it is best to
operate the detector at the highest possible sensitivity, so
as to minimise the impact caused by the ‘orientation effect’
described in Chapter 3.
Retailer and consumer brand codes imposed by external
organisations may dictate operating sensitivities – and these
should always be considered as the minimum acceptable
standards. If more stringent standards can be practically
applied, then this is considered good manufacturing practice.
It is important that the metal detection system can
deliver long-term, effective and reliable operation at the
programmed operating sensitivity; if not, operators will
lose confidence in the control point, and there may be
a tendency to turn down the sensitivity setting to avoid
nuisance false rejections.
Maximum attainable sensitivity and reliability both ultimately
depend upon the quality and reliability of the detector type
(Refer to Chapter 2 and Table 3b for further information).

56
10.2 Establishing the Operating Company-Wide Sensitivity Standard
It is normal for producers to apply a common company
Sensitivity Performance standard across many different production lines and
The best attainable sensitivity will depend upon product products. This common sensitivity standard will apply to
size, type and packaging material; optimum sensitivity many different detectors of differing type, age and reliability
should be selected in consultation with the metal detector levels – and made by various different manufacturers.
manufacturer’s representative. The disadvantage of a common sensitivity standard
When determining operating sensitivity (or when comparing under these circumstances is that sensitivity will not be
capabilities of different metal detectors), the following maximised for a given application or product; worse still,
factors are important: the company standard is likely to be dictated by the lowest
common denominator, i.e. the worst sensitivity performance
• The sensitivity performance should be maintained
or the least efficient detector.
permanently without the need for operator attention. An
unstable unit that requires constant attention is of no value Agreeing to a minimum company standard for finished
• The detector should not reject any ‘good’ product due product inspection will, however, help to overcome the
to false reject signals from the product itself, local plant possibility of a detector being installed at the wrong place
vibration and/or other outside influences in a production line – such as where the inspection of
finished cases is being considered instead of inspecting
To achieve the best sensitivity performance for packaged
each individual item to be placed in a case; the larger
product, the packaging materials should be free from metal
detector required to accommodate the outer pack would be
components such as those types of staples and metal
of reduced sensitivity because of its larger size.
contamination that can be found in low-quality recycled
carton board. Product Specific Sensitivity Standard
The best attainable sensitivities should be established and In order to maximise operating sensitivity, consideration
set for each product setting. should be given to defining sensitivity standards at product
level, though the number of settings for different products
If the signal from the test sample is large, when compared to should ideally be kept to a minimum. The more options
the point at which the metal detector triggers (the ‘detection available to an operator, the more likely it is that a mistake
threshold’), there is a ‘good margin of detection’. This can will be made in selecting the correct product settings.
usually be verified by observing the level of detection on the
metal detector user interface – assuming that the display Product Group/Production Line Sensitivity Standard
provides an accurate representation of the detection signal. Where products are similar, it is common practice to define
the sensitivity standard at a product/group level or for
10.3 Establishing a individual production lines.
Sensitivity Standard Defining sensitivity standards at product/group and/or

production-line level may give assistance in identifying
The product manufacturer needs to balance the desire poorly performing detectors.
for maximum operating sensitivity performance with the
Newer technology allows for multiple products to be
practicalities of implementation and enforcement. For these
“automatically” clustered on one or a minimum number of
reasons, the performance level should be based upon
product settings. As well as avoiding the risk of the wrong
risk assessment, and is ultimately the decision of the
product being selected during product changeovers, this
product manufacturer.
clustered approach helps maintain/improve OEE rates as
The sensitivity standard is usually set at one or more of the the time taken between product changeovers is kept to an
following levels: absolute minimum.
• Company-wide
• Product-specific
• Product group – or Production line-specific
Establishing a sensitivity standard can be relatively easy
for producers of small, dry items such as confectionery,
but establishing such a standard can become more difficult
when a wide range of ‘product effect’ items are produced.

57
10
10.4 Documenting the Advances in technology and more specifically the

development of ultra-high tuned and Multi-Simultaneous
Sensitivity Standard Frequency (MSF) technology have seen a considerable
improvement in detection capability for “dry products”
The sensitivity standard should be expressed as the
(Table 10b) and “wet products” / metallised film-packed
minimum detectable ball size. This should be denoted by
products (Table 10c).
the nominal spherical ball diameter and the material type,
e.g. 1.0 mm diameter, 316 stainless steel. As discussed in
Chapter 3, it is important to state the actual material type
Product Ferrous Non- Stainless
(e.g. ferrous metal, non-ferrous metal or stainless steel)
Height Ferrous Steel (316)
and not simply the generic name (e.g. metal) because of
(Brass)
the differences that will exist in magnetic permeability and
electrical conductivity of each individual type of material
Up to 0.5 – 0.6 – 0.8 –
which falls within the generic name classification.
25 mm 0.6 mm 0.8 mm 1.0 mm
The minimum detectable ball size should be qualified with
regard to the metal detector aperture height and the product 25 mm to 0.6 – 0.8 – 1.0 –
type/application. Typical quoted product/application types 75 mm 0.8 mm 1.0 mm 1.2 mm
are as follows:
• Dry product 75 mm to 0.8 – 1.0 – 1.2 –
• Hard-frozen 125 mm 1.0 mm 1.2 mm 1.5 mm
• Fresh/conductive product
• Wet/de-frosting product 125 mm to 1.0 – 1.2 – 1.5 –
• Metallised film-packed product 175 mm 1.2 mm 1.5 mm 1.8 mm
The sensitivity standard should be formally documented Table 10b: Typical sensitivity standards on dry products using ultra
(issue control and authorised) and effectively communicated
high tuned frequency technology
throughout the organisation. It should also be readily
available to appropriately trained verification personnel.
10.4.1 Retailer Sensitivity Standards Height Ferrous Steel (316)
Retailers and leading consumer brand codes often define (Brass)
minimum sensitivity standards that their suppliers should
achieve for their products. Table 10a provides a typical Up to 0.8 – 1.2 – 1.5 –
expression of a major retailer’s sensitivity standard for dry, 25 mm 1.0 mm 1.5 mm 1.8 mm
wet, and metallised film-packed products.
25mm to 1.0 – 1.5 – 1.8 –
75 mm 1.2 mm 1.8 mm 2.2 mm
Height Ferrous Steel (316) 75mm to 1.2 – 1.8 – 2.2 –
(Brass) 125 mm 1.5 mm 2.2 mm 2.5 mm
Up to 1.5 mm 2.0 mm 2.5 mm 125mm to 1.5 – 2.2 – 2.5 –

25 mm 175 mm 2.0 mm 2.8 mm 3.0 mm
Table 10c: Typical sensitivity standards on wet/metallised film-packed

25 mm to 2.0 mm 2.5 mm 3.5 mm
products using MSF technology
75 mm
75 mm to 2.5 mm 3.0 mm 4.0 mm

125 mm
125mm to 3.0mm 3.5mm 4.5mm

175mm
Table 10a: Example of a major retailer’s sensitivity standards

58
10
Notes

59
11
Installation and Commissioning

Once a manufacturer has identified the need to purchase a metal detection system, due consideration should be
given to ensuring that the equipment is correctly installed. In addition, it should be properly serviced, checked and
maintained throughout its operational lifetime. Objective records of such activities should be generated and stored
for future use.
11
Installation and Commissioning 11.1 Installation
The intended installation location and environment for a
11.1 Installation metal detector could potentially have an adverse effect
on its operational performance – so the manufacturer‘s
11.2 System Validation, Verification and Maintenance installation instructions should be consulted before and
during installation. This will ensure that the best possible
performance is obtained from the system; it will also
minimise the risk of false rejections during operation.
Instructions provided by the system manufacturer will
contain more information than this guide can provide;
however, general principles can be applied to most metal
detection systems, and gaining a basic understanding
of such principles will help with equipment selection,
specification and installation.
Basic guidance covers:
How to Lift and Move Detectors
It may be tempting to move the detector by means of the
aperture, which might appear to be an inviting location for
easy lifting; however, the inner surfaces of the aperture are
usually not structural and cannot support the weight of the
entire detector. So in order to avoid damage, lifting slings or
supporting equipment should never be passed through the
detector aperture when transporting or handling.
Equipment Access
Equipment should be positioned so as to give clear access
from all sides, for ease of cleaning, servicing and operation.
In addition, there should be no need for dismantling during
routine operations.
Keep the ‘Metal-Free Zone’ Clear
The detector‘s Metal-Free Zone should be kept clear of all
metal. Moreover, the specific requirements of the metal
detector manufacturer should be observed with regard to
both stationary and moving metal parts. Consideration
of these factors during installation will provide improved,
consistent and reliable metal detector performance.

60
Vibration and Mechanical Shock Table 11a provides a checklist of items for consideration
As far as is practically possible, metal detection systems during system commissioning:
should not be installed in areas that are subjected to, or
near sources of vibration and mechanical shock. Where Checklist
such conditions cannot be avoided, every effort should be
made to minimise these effects. Equipment and support documentation has been
✓
Electromagnetic Interference (EMI) correctly supplied
Radiated electrical noise generated by surrounding electrical
Installed equipment is in a satisfactory condition ✓
installations may adversely affect the performance of
a system, to a point where the system exhibits erratic
operation and can, for example, produce false rejections. Equipment has been satisfactorily installed ✓
This can prove costly and can lead to the loss of operator
confidence. So, wherever possible, systems should not be Operation of equipment has been satisfactorily
installed in close proximity to any devices, such as radio ✓
qualified in the intended installation
transmitters, which may emit Electromagnetic Interference
(EMI). All inverters and variable-speed drives in the Equipment is capable of reliably detecting and
proximity of the detector should be installed in accordance ✓
rejecting at the defined sensitivity standard
with the manufacturer’s instructions.
Where possible, cables from inverters, variable-speed Operators have been trained to a minimum basic
✓
drives, etc. should not be in close proximity to the detector level (operation, care and maintenance)
or the detector cables. In particular, care should be taken
Table 11a
to avoid placing the detector in close proximity to any
equipment that generates Electromagnetic Interference in the
same frequency range as the detector. A trained engineer from the original equipment manufacturer
(or their representative) is the recommended resource for
Clean Power Source
carrying out the required commissioning process. Experience
Power cable noise may arise from any significant changes gained in other installations can enable engineers to identify
in the loading of the electrical mains that feed the system. potential problems early, so that corrective actions can take
Power cable noise may adversely affect the performance of place during the commissioning process.
the system to such an extent that the system exhibits erratic
operation, e.g. false rejections. The optimum power supply for Documented evidence should be generated so as to
a metal detector should be from a source which supplies only demonstrate that all key aspects of the installed metal detector
low-power equipment; furthermore, it should not be connected system have been satisfactorily qualified, prior to operational
to other power sources supplying varying current loads. use. This qualification should be considered specific to the
actual installation location and surrounding environment.
Installation Compliance
Re-qualification of the installation should be considered if
All aspects of metal detector installation should satisfy
there is a significant change in or around the installation,
relevant legislation which is applicable to the country in
or if the equipment is moved to a different location. The
which the equipment is installed.
operational aspects of the metal detector system should be
System Commissioning re-qualified prior to running new or revised products through
Prior to operational use, the installed metal detection the existing installation. Documented evidence should be
system should be commissioned to ensure that: generated to demonstrate that this process has taken place.
• The installation complies with the
manufacturer‘s recommendations 11.2 System Validation, Verification
• The system operates as intended
• All relevant personnel are trained in its safe and proper use and Maintenance
It should be expected that when a metal detector is supplied
it is accompanied with a set of performance validation
documents to support the initial installation and set up.
Such documents or Installation packs (IPac’s) are designed
to ensure that the equipment can be correctly installed
and maintained throughout its service life. This allows the
equipment to operate at optimum performance with the
maximum possible uptime. The preventive maintenance
programme should include regular maintenance and
performance verification checks by a trained person.

61
11
IPac’s can assist in the auditing process by:

• Meeting the limits of your critical control points
• Giving you the tools to monitor critical control
point performance
• Providing a systematic approach to record keeping
and documentation
• Ensuring that your system is working optimally now and
into the future
Typically a verification process should take place once every
6 to 12 months – and ideally, it should be carried out by
a trained engineer in accordance with an agreed service
contract. An experienced engineer can frequently identify
potential equipment and programme problems, and can
suggest solutions, before they become an issue.
A Performance Verification Certificate should be issued
for each piece of equipment audited. Over time this builds
into a complete log of system performance verification
which can be used to prove due diligence and regulatory
compliance in line with the demands of your customers.
The system also assists external auditors in understanding
the safety measures in place and adherence to standards.
This simplifies procedures and ensures you meet or exceed
audit requirements every time. Refer to Chapter 12 for
further information on performance validation, verification
and monitoring.
11.2.1 Belt Maintenance

Certain substances, (e.g. metal fragments, liquids etc.) can
adhere to the conveyor belt – and if detected by the metal
detector, they are likely to cause unexpected detections,
often giving the appearance of erratic or incorrect operation.
To minimise the chance of this occurring:
• Operations that create metal fragments (such as welding,
metal drilling or cutting) should not be carried out in the
vicinity of the conveyor; otherwise, such operations
may cause metal fragments to come into contact with
the conveyor
• Conveyor belts should be cleaned regularly
• If a belt needs to be replaced an anti-static belt should not
be used (Refer to Chapter 4.1.1 for further information).

62
11
Notes

63
12
Performance Validation, Verification and Monitoring

This chapter provides guidance on the essential elements of validation and verification procedure. The definition and
use of the terms can vary from one organization to another. Irrespective of terminology, clear guidance is required to
ensure that validation, verification and monitoring is ongoing and forms part of the local HACCP plan.
12
Performance Validation, 12.1 Validation Procedure
Verification and Monitoring All metal detection equipment must be validated at
12.1 Validation Procedure the time of installation by the manufacturer or their
representative. They should show, through the provision of
12.2 Verification Procedure objective evidences, that the requirements for the specific
12.3 New Installations intended use or application have been met. If substantial
modifications subsequently occur, a revalidation of the
12.4 Predictive Analytics machine should be carried out.
12.5 System Monitoring (Testing)
12.6 Frequency of Testing 12.2 Verification Procedure
12.7 Number of Tests Any metal detection system should be periodically
verified (typically at 6 to 12 month intervals) in order to
12.8 Detector Sensitivity and Reject Device Test Methods
demonstrate due diligence. In addition, verification will
12.9 Product Rejected During Normal Testing ensure that:
12.10 Fail Safe Systems Testing / Conveyor Systems • It continues to operate in accordance with the specified
sensitivity standard
12.11 Performance Verification and Monitoring Routines • It continues to reject contaminated product on detection
12.12 Test Results of contaminants
• All additional warning/signalling devices are effective
(e.g. alarm conditions, reject confirmation)
• Installed fail safe systems are functioning correctly
• All current safety standards are being complied with.
The verification procedure should ensure that the company/
line/product sensitivity standard and metal detection policies
are being complied with. All metal detection equipment must
be independently verified at minimum, on an annual basis.
12.2.1 Verification Audits

When audits of metal detection systems are carried out
by trained service engineers, these audits can provide
an additional, valuable service. Such audits will support
the overall metal detection programme by ensuring
that equipment complies with the manufacturer’s
recommendations and good practice. Experienced metal
detection experts can often spot potential problem areas
(and suggest solutions) before such problems become
apparent to the user.

64
The performance verification should be undertaken by the
equipment manufacturer or their appointed agent who
12.3 New Installations
can demonstrate evidence of competency by providing New metal detector installations shall be validated by the
valid and up to date training certificates relating to the manufacturer and have the verification and monitoring
equipment concerned (make and model). The certificate process in place before production commences. Once
must relate to the individual concerned and not just the complete the records will be kept and the normal verification
organisation they represent. Where this is not possible an and monitoring frequency will be followed.
independent company, who can demonstrate evidence of
their competency, can perform the procedure. 12.4 Predictive Analytics
There is a more robust and valid approach to determining
12.2.3 Built-in Performance Routines that the metal detection system is continuing to detect
A metal detection system that has a built-in performance and reject in accordance with the sensitivity standard; this
verification and monitoring routines can aid the discipline and approach is to undertake continuous checks for changes
record-generation of testing procedures. Such routines can in the key operating parameters of the metal detector. If the
automatically request a test at an agreed pre-set time interval. key operating parameters can be continually monitored
for stability using condition monitoring, there is scope for
The approved test operative should enter a personal access
reducing the frequency of testing beyond the existing time
number into the detector to allow the test to be completed
period with the attractive benefit of increasing the users
with the correct test samples. Failure to test the equipment
Original Equipment Effectiveness (OEE).
at the agreed time interval could cause a range of different
outcomes. Hard-copy documentation (which proves that Some commercially available metal detectors offer
testing has been carried out) can be provided through a continuous monitoring features (see section 2.1.7).
local printer; alternatively, it can be downloaded to a central However, when considering their use, it is important to
PC, using a detector with network connectivity capabilities. ensure that the system will automatically alert users when
there has been an unexpected change in the monitored
12.2.4 Documentary Evidence parameter. This should prompt a verification test and a
stop alarm if there is an unacceptable change. However if
Copies of training certificates/evidence of competency for he system remains in specification until the auto alert is
all external personnel must be attached to all Performance activated the benefits can be considerable to the user.
Validation Certificates issued for each metal detector checked.
12.2.5 Annual Verification 12.5 System Monitoring (Testing)

In ensuring that the metal detection system continues
The annual verification must be far more in depth than
to detect and reject in accordance with the documented
simply repeating the scheduled daily validation tests that
standard, the purpose of the test is to ensure that there has
are carried out by the individual sites. These verification
been no significant change in the detector’s performance
checks must be in line with general HACCP
level since the last successful test. These changes could
based requirements, fully documented and should include
occur as a result alterations to:
as a minimum the following:
• Machine settings
• Metal detector manufacturers initial build parameters –
• Product signal
not accessible to the user
• Metal detectors functionality
• Electrical and mechanical installation checks
• System functionality checks including adherence to the The selection of correct metal contaminant types for
specified critical limits testing is important; this is because the significance of
• Product related information checks a change in machine settings, product signal or metal
• Fail safe functionality checks detection functionality can vary, depending upon the type
• Customers test piece verification checks of metal contamination.
• Verification that line personnel are trained and For example, it is allowable that the phase of the product
knowledgeable in undertaking the SOP regarding the may change to the extent that stainless steel operating
local verification and monitoring tests sensitivity is maintained, whilst ferrous sensitivity is lost.
If, however, testing is only conducted with stainless steel,
A summary of the verification tests must be completed and
there will be no indication of loss of operating sensitivity to
an indication of the performance since the last test and any
ferrous contamination.
potential degradation in the previous year and the following
year should be commented upon. Guidance is sometimes given on non-ferrous and stainless
steel ball sizes that can be detected, based upon actual
In addition to the above it is required that the test engineer ferrous ball size detected. For example:
also verifies how the system is being verified and monitored • 1.5 x ferrous ball size = detectable stainless
by auditing a member of the production personnel (at steel ball size (at 300 kHz)
random) with respect to carrying out the regular metal
detector tests as detailed in the manufacturing sites SOP for • 1.2 x ferrous ball size = detectable stainless
monitoring CCP’s. steel ball size (at 800 kHz)

65
12
However, such generalisations should be treated with 12.5.1 Test Samples Types to be Used
caution. The actual ratios are very much application
specific and even when application information is available, Test samples can be used on their own (without product) in
necessary assumptions and approximations may limit the order to qualify the operation of a metal detection system.
accuracy of the information provided. Ideally though, to monitoring the actual performance during
production, they should be placed within the product or
In reality, the relationship between one contaminant type should be securely attached onto packed product.
and another is complex, and is based on:
• Test ball sizes Test samples comprise a precision ball-bearing that is
• Material encapsulated within a non-metallic/non-conductive carrier.
• Metal detector operating frequency Various ball-bearing materials are available to represent
• Metal detector phase setting (i.e., whether it is a potential sources of contamination.
conductive or non-conductive product) It is recommended that test samples are purchased from
For the reasons outlined above, best practice is to perform metal detector suppliers who manufacture such samples in
monitor tests using all three contaminant types, i.e. ferrous accordance with a certified quality system e.g. ISO9001:2000
metal, non-ferrous metal and stainless steel (assuming that certified for the provision of test samples.
they are all potential sources of contamination). Test samples should be certified and permanently marked
However, if this is not considered practical, a compromise with the ball size, material and batch-specific reference
would be to focus on testing for ferrous contamination on number. This information allows for traceability back to the
dry product applications (including non-ferrous at higher original precision ball-bearing producer’s manufacturing lot.
frequencies) and testing for stainless steel contamination on The certification should also state:
wet/conductive products. • Reference number
• Nominal spherical ball diameter
The reason for this recommended procedure is that these • Material
tests are more likely to highlight changes in phase and • Manufacturing standard to which the test sample complies
sensitivity affecting detection capability. The degree of
compromise ultimately depends on: Some of the most common test sample types available are
• The stability of the metal detector listed in Table 12a and are shown in Figure 12.1
• The consistency of the product
• The level of control over operator intervention i.e. Generic Carrier Type Typical Application
setting changes
Test Card Conveyor lines with discrete
Regardless of any guidance provided, there is no substitute for
packed products
both ’in-plant’ knowledge, and carrying out tests on the actual
product. From risk assessments already conducted, (see
Test Stick Conveyor lines with discrete
Chapter 9) the types of potential metal contaminants within
packaged or bulk products
the manufacturing facility should be known – and using this
knowledge, the following factors should be determined:
Test Tablet Pharmaceutical and
• Which contaminant types are the hardest to detect? nutraceutical applications
• Where is the worst-case detection location for each
given contaminant? Test Ball Gravity-feed inspection of
This information will help to produce the most effective test powders and granules
method for any given application.
Test Rod Gravity-feed inspection of powders
Ultimately, when testing for contaminant types, the and granules and inspection of
minimum requirements for such tests should be that liquids, pastes and slurries (where
they satisfy any external customer code, retailer code or test sample retrieval is not practical)
company specific policy/test-requirements.
Table 12a: Common test sample types available

66
12
12.5.3 Effective Use of Test Packs

Test packs are commonly used on packed product
inspection lines (Figure 12.2) – and the following
requirements should be defined and contained within the
test procedure:
• The method of checking that the packs are free from
contamination before inserting/attaching test samples
• The method of making up the test pack, including the
position/location of the test sample within/on the test pack
• The frequency at which the test packs should be made
up, so as to reflect the nature, durability and shelf-life
of the product. Ideally, test packs should be freshly
made up for each production batch, since the ageing
of the product may affect the sensitivity, and will not be
representative of the actual product being manufactured
• The method for labelling of the test packs (e.g. marking
Figure 12.1
with coloured tape), so that they are not accidentally put
into the supply chain
Some metal detector suppliers can provide application-
specific test samples – and these suppliers will be able to Non-food test packs can be used for testing metal detector
offer guidance on the most appropriate type of test sample fail safe systems; however, they should be representative of
carrier to be used for a given application. the size, shape and weight of the food products travelling
down the line.
The monitoring procedure should fully define the test
samples to be used, including the actual material type (e.g.
ferrous, non-ferrous, brass, aluminium and 316 stainless
steel) and not just a generic name (e.g. stainless steel).
This is because of the differences that will exist in magnetic
permeability and electrical conductivity (see section 3.2).
Test samples should always be visually inspected before
use, to ensure that they remain fit for purpose. If there is any
doubt as to the integrity of any given test sample, it should
be replaced.
Figure 12.2
12.5.2 Positioning of Test Samples

The preferred location of the test sample is the position in
which it is least likely to be detected, which could be either
the front, middle or back of the pack and this would have
been identified during the commissioning phase. So it is
important that the positioning of the test samples is clearly
defined, both within procedures and within the company’s
Sensitivity Standards.
As well as positioning he test sample at the front, middle
or back of the pack it should always pass as close as
possible to the geometric centre of the aperture – i.e. in the
least sensitive position. In most instances, this will be a far
greater challenge for the metal detection system than the
finite worst-case location within the product/test pack or on
the product/test pack.
The position of test samples is important for the correct
testing of the reject device and its timing functionality.
Further guidance on the positioning of test samples is
provided within each of the test methods, as covered in
section 12.8 of this guide.

67
12
12.6 Frequency of Testing Customer, Retailer and Consumer Brand Codes of Practice
Customer, retailer and consumer brand codes of practice
Procedures should clearly state when scheduled testing may well specify a frequency of tests that may need to be
should be performed within the manufacturing cycle. adhered to.
Consideration should be given to implementing verification
testing at the following stages: Margin of Detection
• At the start and finish of daily production/shift When there is a good margin of detection (see Chapter 11),
• At regular intervals during the production run and it is certain that the system will fail safe in the event
(as necessary) of a problem, there is scope for reducing the frequency of
• At changes in production batches performance tests. This is based upon the fact that even
• At changes in machine settings if there are small changes in phase and sensitivity, the
• After downtime for repairs metal detector will still comfortably detect the specified test
samples. Such decisions should only be taken when the
Considerations for each of the above stages are
margin of detection can be quantified and when the risks
defined in sections 12.6.1 – 12.6.5.
are considered acceptable.
12.6.1 Start and Finish of Daily In practice, the sensitivity standard may apply to many
different detectors – and these detectors may vary in
Production/Shift accordance with the manufacturer, their type, age, reliability,
Consideration should be given to conducting testing at and other factors. Therefore, the margin of safety may not be
the start and end of the daily production/shift. This will the same for all detectors and products.
ensure that the metal detection system detects and rejects Fail Safe System Design
in accordance with the sensitivity standard; it will also
Robust fail safe system design and access control can be
ensure that any additional warning systems are functioning
used to good effect in reducing the likelihood of a failed test
correctly, e.g. Reject Bin Full indicator.
and in also reducing the frequency of testing. For example,
In addition, where fail safe features have been included as if production-line operators are restricted in making setting
part of the system specification, they need to be verified at changes (e.g. lowering sensitivity by means of access
the start of each shift. If a failure is observed, this should be control), the potential for a test failure is reduced. Likewise,
corrected before commencement of the daily production/shift. if the metal detection system automatically requests a
test each time there is a product change, this will limit the
12.6.2 Regular Intervals During possibility of a product being run on the incorrect product
memory selection.
Production Run
Automatic System Testing
The frequency of testing during a production run needs to
be defined within the procedure, and will ultimately depend Automatic testing systems may be used in addition to
upon the probability and consequences of a failed test. The manual test where physical contaminants make testing
following factors should be taken into account: difficult, i.e. throat, gravity fall and pipeline applications.
However, they should not be used as a total replacement for
• Quarantine period manual testing.
• Customer, retailer and consumer brand codes of practice
(if applicable) Whilst automatic testing systems may have the potential
• Margin of detection to detect any reduction in sensitivity of the metal detector,
• Fail Safe system design they cannot identify any problems with the reject
• Automatic system monitoring mechanism or timing.
Quarantine Period Automatic testing cannot be operated continuously

since the detection of metal in every pack on which it
The quarantine period relates to the time the product
runs is compromised, the signal from the test ball interferes
remains on site after the last successful test and if needed,
with or could in some cases cancel the signal from any
can be prevented from being shipped.
real contaminant.
The testing period should always be shorter than the
Such devices have not been widely adopted within the food
quarantine period, because if there is a test failure, the
industry however the benefits can be attractive as they can
product manufactured since the last successful test will still
support the customer’s drive to support OEE.
be on the company premises. It can then be easily identified
and isolated pending further action (Refer to Chapter 13 for
further information).

68
12
12.6.3 Production Changes However, where good detection capability has been
established during commissioning, acceptable practice
Whenever there is a change in product type running through would be: one test per test sample material type
the metal detection system, consideration should be given and position.
to performing a verification test to confirm detection and
rejection in accordance with the sensitivity standard. This is The number of tests to be performed for each test sample
most important when the change of product type requires material type ultimately depends upon the level of statistical
a selection of a different product memory within the metal significance required within the producer organisation.
detection system. The number of tests to be performed also depends on
the level of statistical significance needed to fulfil any
external requirements.
12.6.4 Setting Changes
Some Metal Detection manufacturers offer multi pass testing
Whenever there is a change in the metal detection software within their automated test routines.
system settings, a verification test should be performed
to confirm detection and rejection in accordance with the
sensitivity standard. 12.8 Detector Sensitivity and Reject
Device Test Methods
12.6.5 After Down-time for Repair
Test procedures should include precise details of the methods
If maintenance work or repairs have been carried out on to be used. These methods will vary, depending on the metal
the production line during down-time, the metal detection detection system design and the actual application.
system and reject mechanism should be re-verified at the
recommencement of production. As well as ensuring that the metal detector is performing to
the required sensitivity standard, it is important to test that the
Good practice is to use a built-in performance verification reject device is functioning correctly; this is to ensure that it is
routine (see section 12.13) when the following three still capable of rejecting the detected contaminated product.
circumstances occur:
For example, it is common for conveyor speeds in plants to
• When a performance test needs to be carried out, and…
be changed for a variety of reasons. If this occurs, and the
• Before normal operation can be resumed, and…
reject timing is not suitably adjusted, it may be possible to
• When there is a ‘power-down/power-up’ situation
reject the wrong product. Similarly, the air supply to an air-
blast reject device could be easily disconnected, resulting
12.7 Number of Tests in failure to reject contaminated product. Consequently, it
is more efficient to devise a test method that applies to the
The number of tests to be performed should be based
metal detection system (detector and reject device) at the
on the level of confidence established during the original
same time.
commissioning activity, when the capability of the metal
detection system will have been originally established. As proof of the test’s success, all the test packs/test
samples should be detected and rejected to the correct
If there was good, repeatable detection capability, then
reject location.
this sense of confidence should be carried through to
production verification testing, i.e. if there is a good Should any part of the test fail, there should be isolation
margin of detection on a single test, what is to be gained of product that has been manufactured since the last
by conducting further tests? satisfactory test. This product should then be re-screened
using a functioning detector (Refer to Chapter 13 for
Alternatively, if a test sample is only just detected during a
further information).
test, then repeatability may be questioned. Further testing
may give greater confidence – though if testing is conducted 12.8.1 Testing Conveyors with Discrete
three times, and the results obtained are one marginal detect
and two good passes, what is the statistical significance of Products or Packs
this in relation to a high-volume production line? The test should be carried out so that, as a minimum,
Statistically, further marginal passes or even a mis- there is a test sample on the leading edge of one of the test
detection could be expected; so the system probably does packs, test sample in the middle of one of the test packs
not have sufficient detection capability in the first instance, and a test sample on the trailing edge of another test pack.
and consideration should be given to increasing the Test packs should be passed down the production line
frequency of testing. through the metal detector one after another.
For production test purposes, the maximum practical level (N.B. The above testing assumes that the system is as per
would be: three tests per test sample material type and position. the original validated system).

69
12
This test method will generally represent the worst-case The spacing between the packs should be the normal
scenario. It will verify whether the metal detection system distance between products which travel down the line – and
can detect and reject contaminated product, regardless of: the packs should be placed so as to break the photo-gate
• The position of the contamination in the product sensor beam (i.e. the pack sensor beam), if one is fitted.
• The systems ability to detect consecutive contaminants For unpacked discrete product, the test sample should
be placed into (or securely on top of) the product in the
If testing is carried out with three different contamination specified locations.
material types:
For a belt stop alarm system, each individual pack should
• Sample 1 should be placed on the leading edge
be passed down the line. The test pack should be detected,
• Sample 2 should be placed on the trailing edge
and the conveyor should have stopped for the test to have
• Sample 3 should be placed in the middle
been successful. It should only be possible to re-start the
If testing is only being carried out with one material type, system using a key held by a nominated person.
(e.g. ferrous metal only, as per a Ferrous-in-Foil detector),
The test sequence should be repeated for the specified
then two packs should be used:
number of tests. As discussed previously, the types of
• Test piece one should be placed on the leading edge of the contamination to be tested (and the number of tests to be
first test pack performed) depend upon a variety of factors. Ultimately, the
• Test piece two should be placed on the trailing edge of types of contamination to be tested, and the number of
the second test pack tests to be performed, depend on the level of risk that the
Precautions should be taken to ensure that any non-rejected company is prepared to take.
test packs or test samples do not become lost in the
product flow. 12.8.2 Testing Conveyors with
For smaller-length or triangular-shaped packs (e.g. Bulk Product
sandwiches), the positioning of the test sample pieces
at leading and trailing edges may not be practical. In Precautions should be taken to ensure that test samples will
these cases, the test pieces should be placed in the most not be lost if they are not detected or rejected, especially
convenient position that allows them to pass as close as if the product is being fed directly into another processing
possible through the centre of the metal detector aperture. machine after the metal detection system.
With the conveyor set at the normal production-line running The specified test samples should be evenly spaced,
speed, all test packs should be placed onto the production and placed into the product flow in the centre of the belt.
line, as shown in Figure 12.3 (assumes testing of ferrous The test sequence should be repeated for the specified
metal, non-ferrous metal, aluminium and stainless steel). number of tests.
Metal Detector
Metal Detector Packs
Packs equally spaced
equally spaced 12.8.3 Memory Testing
Some code owners and retailers request this type of
testing is undertaken. The memory test comprises of 3
contaminated packs and 2 non-contaminated test packs
as illustrated in the lower of the two images of Figure
12.3. The theory being that the 3 contaminated packs are
rejected and the 2 non-contaminated packs are accepted.
However in many cases due to line speed and the
Direction of travel
Direction of travel proximity of packs the detector cannot distinguish between
consecutive contaminated and non-contaminated packs.
In addition, if amplitude detection is being operated, the
non-contaminated packs will be rejected. In this case the
code owner may require a written report from the equipment
Metal Detector
Metal Detector Packs
Packs equally spaced
equally spaced
manufacturer that the equipment is working as intended.
Direction
Direction of travel
travel Test
Figure 12.3

70
12
12.8.4 Testing of Free Fall Vertical Product Flow

Product Flow
Metal Detector
Metal Detector
Packaging Applications
Ideally, test samples should be placed independently in the
product flow, and the reject device should be observed to
see if it successfully catches or removes the contamination. Test Piece and
Carrier Introduced Between
It is important to check that the specified test sample type is Pipe and Metal Detector
capable of being recovered if it is not detected or if it is not
Figure 12.5
successfully rejected. If this is the case, the specified test
sample can be inserted into the product flow to verify the
correct operation of the reject device, e.g.: 12.8.6 Testing of Gravity-Feed
• Double pack made Inspection Systems (Bulk Powders
• An audible and/or visual alarm activated
• Line stops and Granular Products)
If this check procedure is not possible, the system will have Ideally, test samples should be placed independently in the
to be tested by inserting a test sample as close as possible product flow at the point were product begins to fall and the
to the centre of the aperture. Then the response from both the reject device should be observed to see if it successfully
metal detector and the reject mechanism should be observed. diverts the test sample to the reject position.
In such instances, the sensitivity gradient of the metal detector If the equipment has been correctly specified, a test sample
should be known and accounted for; this is because the access port and safety catch grids should be located
test will be carried out at a more sensitive part of the metal beyond the reject device. The purpose of the sample access
detector than that part of the metal detector which contains the port and safety catch grids is to catch a test sample if the
product (worst-case sensitivity is the centre of the aperture). detector fails to detect – or to catch a test sample if the
reject device fails to operate correctly.
The test should be repeated for the specified number of
times and for each metal contamination type. If the above devices have been included, the specified test
sample can be inserted into the product flow. Once the test
sample has been inserted, its diversion to the reject location
12.8.5 Testing of Pipeline Applications can be checked. The safety catch grids should be removed
(Liquids, Slurries and Pastes) after testing so as not to restrict the flow of the product.
Ideally, test samples should be placed independently in the If there is no test port, an access point will need to be
product flow, and the reject device should then successfully identified above the metal detector; this access point will be
divert the test sample to the reject position. for the insertion of the test sample.
If the equipment has been correctly specified, there should The location of the access point should be as close as
be a test sample access port and a means of catching the possible to the location at which the product begins to fall;
test sample if it is not being rejected. If this is the case, the this will ensure that the speed of the test sample will be the
specified test sample should be inserted into the product same speed as that of the product.
flow; it should also be confirmed that the test sample is If the test sample is not rejected, a method of ensuring its
diverted to the reject location (Figure 12.4). recovery will also be required.
Product Flow
Product Flow
The test should be repeated for the specified number of
Metal Detector MetalTest Piece
Detector times and for each metal contamination type.
Input Port
The latest developments on gravity fed systems can
now include an automatic testing system which enables
performance monitoring tests to be conducted regularly
Test Piece
Test Piece inin without interrupting production. This is achieved through the
Reject Valve Fail Valve Product Flow
Flow
Product use of an integrated system that circulates and retrieves a
Figure 12.4 test sample through the aperture.
If this is not possible, the system must be tested by

inserting a test sample rod/wand between the pipeline and
the detector aperture. The test sample rod/wand should be
located as close as possible to the centre of the aperture. At
the same time, the response from the metal detector and the
reject mechanism should be observed (Figure 12.5). The
test should be repeated for the specified number of times,
and for each metal contamination type.

71
12
12.9 Product Rejected During Pack In-feed Sensor

Essential for the optimum timing and operation of the reject
Normal Testing device. Ensures the correct contaminated pack or packs are
If in sound condition and with the test piece retrieved, removed from the line regardless of the size and position of
product rejected during normal test procedures can be put the contaminant in question.
back in the product flow. Key Switch Reset
This product should be placed on that part of the production All of the fail safe system devices which result in the
line that comes before the metal detection system, so that it conveyor stopping should be linked to a key-operated reset
can be re-inspected by the metal detector. switch linked to the restart button. Only authorised key
Once placed on the production line, product rejected during holders should be permitted to restart the system after faults
normal test procedures should be regarded as standard are identified and subsequently rectified.
product that needs to be inspected in the normal manner. Automatic Contaminated Pack Reject Mechanism
A choice of rejection mechanisms are available dependent
12.10 Fail Safe Systems Testing / on line speed, pack speed, pack weight, pack dimensions
and the nature of the packaging material. Reject devices are
Conveyor Systems normally pneumatically operated and controlled via input
A test method should be established for each fail safe from both the metal detector and the pack in-feed sensor.
system that has been built into the metal detection system. Reject Confirmation Sensor
The following are examples of some common fail safe Some metal detection system fault conditions can allow
devices which may be incorporated into the metal detection metal contaminated product to pass through the system
system design (see Figure 12.6). These fail safe devices can without being rejected. To minimise the risk of this, a reject
also be built into associated test methods. confirmation system should be utilised.
Air Blasts or Punch/Pusher With Reject Confirmation Reject Check Sensor
Testing should be carried out by passing a test pack down The addition of a reject check sensor provides real-time
the line while temporarily interrupting the electrical supply monitoring of the pack in-feed sensor. This in turn monitors
to the reject device solenoid. The result should be that the the performance of the reject check sensor itself. Therefore,
reject mechanism does not operate and that the conveyor the sensors constantly monitor each other. If failure of either
belt then stops. sensor occurs, the system alarms within 3 packs of the failure
During the initial-set up of this test, it is important to note allowing the necessary corrective action to be undertaken.
the location at which the test pack comes to rest. If this Reject Bin With Bin-Locked Sensor
does not occur on the metal detection system belt, the
Metal detection systems that include an automatic reject
relevant downstream conveyor should be linked to the reject
device should include a lockable reject bin. In some
confirmation system’s stop circuit.
scenarios, the reject bin can inadvertently be left unlocked
The purpose of this process is to ensure that any metal- leading to a risk of contaminated product being taken from
contaminated packs are easily retrievable for investigation the bin and placed back on the production line after the
after a production-run system failure that causes a belt stop. metal detector.
‘Reject Bin Full‘ Indicator (See chapter 17.0 for a more detailed description of
This should be checked by breaking the beam for the fail safe features)
required length of time and observing that the belt stops.
Condition Monitoring Technology
Provides advanced warning of adverse trends which could
lead to the potential for downtime.
High-Security Operator Access and Event Logging
Access to all metal detection system controls are password
protected via a highly secure, dual level user name and
individual password login function. Data captured and displayed
on-screen includes date, time and name of individual login-in.
Reject Bin Door Status Control
Ensures the reject bin ‘lock’ and ‘unlock’ functions are
controlled through the metal detector operator interface via
a password protected login access.

72
12
Automatic Contaminated
Pack Reject Mechanism
Air Failure Switch
Drive Roller
Rotary Encoder
Fixed Position
Pack In-feed Sensor Tunnel Guard
Reject Check Sensor
Reject Confirmation
Sensor
Key Switch Reset
Reject Confirmation
Test Facility Reject Bin Door
Status Monitor
Reject Bin Full Sensor
with Test Facility
Figure 12.6
12.11 Performance Verification and Should any test (or part of a verification test) fail, the cause
should be immediately investigated and rectified before
Monitoring Routines production re-commences.
A metal detection system that has built-in performance Product manufactured since the last satisfactory test should
verification and monitoring routines can aid the discipline and be regarded as suspect and treated accordingly (see
record-generation of testing procedures. Such routines can Chapter 13). Details of the fault (and subsequent corrective
automatically request a test at an agreed pre-set time interval. action) should be recorded as part of the test record.
The approved test operative should enter a personal access The accurate recording of test results is extremely
number into the detector to allow the test to be completed with important; in the event of a customer complaint or audit, a
the correct test samples. Failure to test the equipment at the manufacturer may need to rely on these records to prove
agreed time interval could cause a range of different outcomes. that procedures were correctly followed and that the metal
detection systems were functioning correctly to the agreed
Hard-copy documentation (which proves that testing has
sensitivity standard. (For an example of a typical record
been carried out) can be provided through a local printer;
sheet, USB and digital report sheet please refer to Figures
alternatively, it can be downloaded to a central PC, using a
12.7, 12.8 and 12.9). Historically all test reports were
detector with network connectivity capabilities.
manually recorded but modern solutions/systems offer data
collection methods (see chapter 15.0 for more information).
12.12 Test Results
All test results should be documented including any fail safe
tests in order to demonstrate that all requirements of the test
monitoring procedure have been executed. These records
should include:
• Metal detection system unique identification reference
e.g. serial number, CCP number
• Product being produced
• Date and time of test
• Test samples used
• Name of the person who conducted the test
• Test result for both detection and rejection
• Test result for any fail safe devices
• Fault details and corrective action taken (where applicable)
• Production line reference

73
12
Metal Detection System Manual Monitoring Record Sheet

System Identification Product
Date Time Test Sample Detection Test Rejection Test Fail safe Test Corrective Action Tested By
Figure 12.7: Example of a typical monitoring record sheet collected manually.
*********************************
PERFORMANCE CHECK START
*********************************
Date :03/08/2008 Date & Time stamp
Time
as required to prove
:09:11:17
due diligence
Line ID :0002
Log of authorised
Account Username
:JOHN SMITH – QA personnel with password
CURRENT SETTINGS
protected access
Name :500g Mozzarella

Sensitivity :896/653
Phase :0.71 Named product details and
Reject On :300ms
detector setting information
Detect To Reject
:450ms
TEST RESULTS
Material :FERROUS Figure 12.9: Example of a typical data management record digitally
Size :1.0m m
collected via ERP software such as ProdX.
Detection :YES
Signal Strength (%
Test Results, including
:231 material type, detection
of Trigger)
result, rejection result, and
Reject Relay :YES
test result
Result :TEST PASSED
Material :STAINLESS STEEL
Figure 12.8: Example of a typical data record sheet collected via USB

74
12
Notes

75
13
Dealing With Suspect and Rejected Product

If a metal detection system fails a scheduled test, the product that has passed through the system since the last test
should be considered suspect. If product is rejected by the metal detection system during routine operations, it should
be considered as contaminated until proven otherwise. In both instances, a clearly defined process is required for
dealing with product safety concerns - from the point of identification through to root cause
investigation and final resolution.
13
Dealing With Suspect and This chapter aims to provide practical guidance when
dealing with product specifically rejected by metal detection
Rejected Product systems. Intentionally, this chapter does not cover more
13.1 More Than Just a Metal Detector general aspects of dealing with rejected product e.g.
identification and traceability, final product disposal,
13.2 Authorisation and Rejects
product recall etc.
13.3 Action Required If a Scheduled Test Fails
13.4 Treatment of Rejected Product 13.1 More Than Just a
13.5 Corrective & Preventive Action Metal Detector
13.6 Metal Detection System Fault Condition Although installing a metal detection system can reduce
the risk, the opportunity for metal contamination reaching
the end consumer is still too high in some cases. Research
points to procedural and system failure rather than metal
detector failure as the main cause. The “due diligence”
enhancement detailed further in Chapter 17 improves the
level of control and management of the inspection system.
This ensures the critical control point (CCP) delivers
not only high performance metal detection but also, an
increased level of fail safe functionality that improves the
capability and performance of the complete system.

76
13.2 Authorisation and Rejects The cause of the failure should be determined, and:
• If it is established that the failure occurred as a result
Only authorised and trained personnel should be allowed to of tampering or because of a change in production
have access to rejected product – and only such personnel conditions, procedures should be established to prevent
should be able to undertake subsequent evaluation and a recurrence of the failure. If the metal detector system
investigations. In addition, proper controls should ensure can be adjusted so as to resume correct operation, this
there is no risk of mixing rejected product with good should be done and noted in test records.
product. To maintain this level of control during production • If it is established that the cause of the failure is due
some additional features should form part of the overall to a system fault, it should be repaired before re-
metal detection conveyorised system as standard. commencing production.
• A Reject Bin Door Access Control. This would ensure In both the above cases, the metal detection system should
that reject bin ‘lock’ and ‘unlock’ functions are controlled be re-tested prior to commencement of production.
through the metal detector operator interface and have a
password protected login access. This is preferable to a The suspect product should be re-inspected through a
key operated access which could be subject to abuse or known working metal detection system. This should have
lack of control. the same operating sensitivity standard as the original
• R
eject Bin Door Status Monitor. Metal detection metal detector system used on the production line. Discrete
systems that include an automatic reject device should and packaged product should be passed through the metal
include a lockable reject bin. In some scenarios, the detection system in the original product orientation.
reject bin can inadvertently be left unlocked leading to Product that passes re-inspection procedures can be
a risk of contaminated product being taken from the bin considered as being acceptable. Any product rejected by re-
and placed back on the production line after the metal inspection procedures should be considered as contaminated,
detector. Advanced metal detectors should include a bin and ought to be subject to further investigation.
locked feature that allows entry into the reject bin for a
pre-set time only. If the bin is left unlocked and the pre- If the product was originally frozen or has undergone a
set open-time is exceeded, a signal is generated raising change in temperature, it may not be possible to replicate
an alarm and shutting down the conveyor system. the original test conditions. In such instances, procedures
• R
eject Confirmation Sensor. Some metal detection will need to define how re-testing and investigation is to
system fault conditions can allow metal contaminated proceed (e.g. frozen state to be maintained etc.).
product to pass through the system without being
rejected. To minimise the risk of this, a reject confirmation
system should be utilised. This takes the form of a
13.4 Treatment of Rejected Product
sensor situated in or across the mouth of the reject bin. Any product rejected during normal production
Upon detection of metal, the system can be configured operation should be regarded as contaminated and
to expect a further signal from the reject confirmation subject to investigation.
sensor that a pack or number of packs has been The evaluation of rejected product should take place as
rejected. If confirmation is not received, the system will soon as possible. Ideally the evaluation should occur within
alarm and stop the conveyor. one hour of rejection, but certainly within that production
• Reject Check Sensor. The addition of a reject check shift – and certainly before the product batch leaves
sensor provides real-time monitoring of the pack in- site. Investigation immediately upon rejection would be
feed sensor. This in turn monitors the performance of considered best practice.
the reject check sensor itself. Therefore, the sensors
constantly monitor each other. If failure of either sensor Identification of contamination by visual inspection is
occurs, the system alarms allowing the necessary unlikely to be successful. Ideally, search for contamination
corrective action to be undertaken. The health of these should be carried out using the metal detection system
sensors, reject failure is averted. The reject check sensor which initially rejected the product. If this is not possible,
also acts as a back-up check for the performance of the an off-line metal detector system with the same (or higher)
primary reject confirmation system. operating sensitivity should be used.
Discrete and packed product should be passed through the
13.3 Action Required if a metal detection system in the same orientation as when it
was originally inspected. It should then be passed through
Scheduled Test Fails the detector a further three times in different orientations; this
If a metal detection system fails to detect or reject a test helps to ensure that the check is as thorough as possible.
sample during a periodic scheduled test, production should Best practice would be to dispose of any product that has
be stopped. been originally rejected by a metal detection system after
Product that has been manufactured since the last re-testing, regardless of whether it is rejected again.
successful test should be regarded as suspect product. It
should then be identified accordingly and isolated from the
rest of the production process, whilst awaiting re-inspection.

77
13
However, this is not always economically viable, especially Procedures should clearly define under what circumstances
if a producer is incurring high levels of rejected product, due production should be shut down. This decision should be
to excessive false rejects. In such instances, the producer based on:
must ensure that all reasonable measures have been taken • The frequency of findings
to ensure that the product complies with the stated sensitivity • The nature of the hazard
standard or more advanced metal detection solutions should • The metal type
be sought that do no suffer false rejections. • The size of the contaminant
If the product is rejected at any stage during investigation, it An example of a major retailer requirement is highlighted
is essential that contamination is found and identified. in Table 13a:
Any packing materials should be removed; only then should
the rejected product be re-passed through the metal detection
system. If the product is still rejected and the contamination Production should be stopped in the
is not found, the product should be divided into smaller and following circumstances:
smaller portions until the contamination is found.
If more than one metal contaminant is detected on any one
After removing the contamination, the remaining product production line within any 24 hour period.
should be re-tested to ensure that there is no further
contamination present. If more than one metal contaminant of the same type is
Finding and identifying individual pieces of metal detected in the factory within any 24 hour period
contamination in the rejected product is important because:
Table 13a
• If the source can be identified, steps can be taken to
prevent contamination recurrence i.e. contracting raw
ingredients suppliers. More modern metal detectors can be programmed to
• Correctly identified contamination can give early automatically do this.
indication of the break-up of a piece of machinery The results of any investigations should be fully documented
• If line operators can see the results, it will help build for future reference and ongoing analysis. These results
confidence in the metal detection system should comprise:
• Details of contamination found
13.5 Corrective and • Source of contamination
• Actions taken for future contamination prevention
Preventive Action
Procedures should clearly define the corrective and 13.6 Metal Detection System
preventive actions necessary if metal contamination
is confirmed. Procedures should also set out who is Fault Condition
responsible for determining the significance of the Sometimes, activation of a fault during a normal production
contamination, as well as who has the authority to hold process results in a ‘Stop Production’ situation. When
product and assign disposal. this occurs, the necessary corrective action should be
If metal contamination is confirmed, an immediate risk undertaken and the system should be re-tested.
analysis should be undertaken to determine its significance; Once the fault has been rectified and the system has been
the risk analysis should also assess the possibility of re-tested, all product on (or in) the stopped process flow
further product contamination. should be collected and re-passed through the metal
Any contamination that is found should be shown to line detection system. This includes any downstream systems
personnel; this will help to build up confidence in the metal (where relevant).
detection system. The same contaminants should then be
kept and stored for future reference.
Locating and retaining particles of contamination has an
added advantage: if, for example, a screen or blade is
known to have broken and fallen into product, the detected
pieces can be collected and the component can be
reassembled to ensure that all fragments are recovered.

78
13
Notes

79
14
Data Analysis and Programme Improvement

The effectiveness of a metal detection programme can only be determined by efficient collection of data, as well as
effective trend analysis. Using this information over a period of time will help to determine the effectiveness of the
metal detection programme; most important, though, it will be the first step in financially quantifying the savings or
increased profit generated.
14
Data Analysis and This chapter highlights some of the typical data sources
that should be analysed when reviewing the operational
Programme Improvement effectiveness of a metal detection programme; some of the
14.1 Data Analysis potential rewards of such a process are also highlighted.
14.2 The Cost of False Rejects
14.3 Programme Improvement
14.1 Data Analysis
Data can be collected, analysed and used in many
different ways. The most effective method of undertaking
such collation and analysis will vary from organisation to
organisation; it will also be dependent on the needs and
capabilities of the business.
However, it is vitally important that there is integrity in the
source data; furthermore, it is crucial that the analysis is
clear in its conclusions, so that it can achieve maximum
buy-in throughout the organisation.
Once the data has been analysed and resultant actions
have been decided upon, this information should be
communicated to those responsible for providing the
original source data; this will help to ensure that data flow
is sustained.
If it is seen that data is not being used to good effect, then
its value will be questioned within the organisation; this
could result in reduced discipline and decreased efficiency
in data collection and recording.
Wherever possible, a cost element should be included in
data that has been collected for analysis. This will help to
accelerate the process when improvement initiatives are
being decided upon and prioritised.

80
14.2 The Cost of False Rejects Customer
Customer Complaints Trend
Complaints Trend
The cost of false rejects isn’t just the cost of the product
but should include all associated costs to conduct and
20
report on the reasons for the rejected product. Many new
legislations and standards include this as a requirement
and therefore cost to support them should be included. The 15
above can lead to the MD system operating at a reduced
sensitivity allowing more metal to be undetected, which in 10
turn would see an increase in customer complaints. It is
imperative that the latest metal detectors are used to reduce
5
false rejections.
The presence of a cost element in analysis data will also
provide justification for the additional capital expenditure Jan Feb Mar Apr May Jun Jul Aug Sep
Jan Feb Mar Apr May Jun Jul Aug Sep
required to remedy faults; e.g. customer complaints cost
Figure 14.1
£6,000 in a given period, whilst false rejects cost £14,000
loss of production in the same period.
14.3.2 Food Safety and
14.3 Programme Improvement Management System Audits
The following are just a few examples of the types of analysis Usually conducted by a company‘s internal Quality
that can prove beneficial in the review and improvement of Department, as well as external regulatory bodies and
a metal detection programme. The same principles can be customers, these audits offer an independent view of the
applied to a variety of data sources. effectiveness of the metal detection programme.
Feedback is a valuable source of information, whether
14.3.1 Customer Complaints received as official non-conformance or as an observed
Each and every metal contamination-related customer improvement opportunity. Ongoing analysis of audit findings
complaint should be investigated to determine its cause. can give additional assurance of effective operation or can
The programme documentation, together with the records identify system weaknesses that need to be improved.
that have been generated, will greatly assist in the
investigation. In addition, they may even prove useful as 14.3.3 Detection Events
evidence in defence against an unjustified complaint.
Detection events are caused by actual metal contamination
The investigation should: or ‘false rejects‘, resulting from interference within the
• Identify the cause of the fault metal detector itself. This could be caused by factors such
• Identify any ineffective monitoring of the Critical Control as electrical noise; ‘false rejects‘ could also be caused
Points (CCPs) by external interference from the installation or from the
• Highlight any new and unidentified Critical Control Points surrounding environment, e.g. earth loop, infringement of
• Establish if the metal particle detected is smaller than the metal-free zone, etc.
the operating sensitivity performance of the metal
detection system Detection event information should be regularly collated
and monitored on a trend chart in order to identify
Corrective and preventive action should be taken on common causes.
the basis of these findings, and future metal detection
programmes should be improved accordingly. Analysis of contamination type and frequency of events,
line by line, or machine by machine, can identify particular
The number of complaints and assigned causes should be sources of concern, such as the quality of suppliers who
monitored over time to make sure that improvements are provide raw material ingredients. Other sources of concern
being made (Figure 14.1); any underlying common causes can include:
should also be identified and eliminated. Such actions can • Inefficient production staff or production methods
drive step-improvements in the reduction of complaints, • Incorrectly set metal detector
with the ultimate he aim of reducing them to zero. • Excessive product variation or metal detectors ability to
handle this.
• Incorrectly designed metal detection system
• Vibration
• Electrical and airborne interference
• Inadequate maintenance routines (Figure 14.2).
There should be a clear distinction between normal
production reject events and reject events that occur when
carrying out routine tests.

81
14
Metal
Metal Items Detected
Items Detected
14.3.6 General
There are numerous other sources of data that can be
analysed to good effect. The key is to focus on the areas
20
20 that can generate the greatest return, in terms of increased
profitability and reduced risk.
15
15
Ongoing analysis of the programme data can identify
underlying common causes that, in isolation, do not
10
10 appear to have great significance. However, when these
common causes are considered in terms of their frequency
5
of occurrence, they can become an incentive to take actions
necessary to prevent occurrence in the future.
Wk1 Wk2 Wk3

Wk1 Wk2 Wk3 Wk4
Wk4 Wk5
Wk5 Wk6
Wk6 Wk7
Wk7 Wk8
Wk8 Wk9
Wk9
Figure 14.2
14.3.4 Performance Monitoring Tests

The results of performance monitoring tests should be
Incorporated as an ongoing process. If a high frequency of
testing is being conducted (e.g. every 30 minutes), and over
time the analysis shows that the tests are always positive,
due consideration could be given to reducing the frequency of
testing. This should take into account such factors as fail safe
system design, access control, data / usage monitoring and
margin of detection (as discussed in section 12.6.1).
There are also large OEE benefits to be gained from
reducing the frequency of testing especially where the
application includes either a stop alarm reject mechanism
i.e. Throat MD’s and those metal detectors that reject large
volumes of product during the test i.e. wide band chocolate
enrober / moulding applications.
Caution should always be exercised to ensure that any external
standards or codes of practice in place are not contravened
and that the risks involved are known and acceptable.
Analysis of ‘false rejects‘ can prove beneficial in identifying
poor installations and equipment that have become unreliable;
this kind of analysis can also identify systems that can no
longer cope with current operational sensitivity standards.
Such data could be used as justification to upgrade to a more
modern and capable metal detection system.
14.3.5 Maintenance Records

If the analysis of preventive maintenance records and
incident reports shows that a particular piece of equipment
rarely needs any maintenance, there may be sufficient
justification to reduce the frequency of maintenance on the
plan, providing this is not contrary to the manufacturers’
recommendations or risk assessment. Alternatively, analysis
may show that maintenance is often required – and that the
frequency needs to be increased.

82
14
Notes

83
15
Data, Connectivity and Improving Performance

The collection of data (both for traceability and performance validation), regulatory compliance and
meeting due diligence obligations are essential requirements for a modern business.
Gathering data efficiently and effectively from a production processes can still prove to be a challenge.
Advanced metal detection systems can utilise a wide range of collection methods that ensure information
is available to meet the changing demands of a business or its customers.
15
Data, Connectivity and While simple plug-in hand held printers have become
somewhat outdated, USB data collection methods
Improving Performance are now the minimum standard with more advanced
15.1 Understanding the Importance of Connectivity electronic connectivity solutions also available to suit
your requirements. In today’s highly accountable business
15.2 Connectivity Media world, manufacturers need to be able to access real-time
production data from process machinery and operators
15.3 SCADA Systems
at shop floor level. If they can do so, this data becomes
15.4 OPC Technology invaluable when made available across enterprise-wide
management systems, remote departments and multiple
15.5 Data Management Software production sites. By making such information so widely
accessible, all those involved in the day-to-day running of
15.6 Improving Production Performance
an organisation are never more than a keystroke away from
15.7 PackML and OEE Calculations the mission-critical information that can govern success.
15.1 Understanding the Importance

of Connectivity
Collecting data supports your business in meeting Hazard
Analysis and Critical Control Points (HACCP) requirements
and the broader needs of external GFSI food safety regulations
and standards (as discussed earlier in Chapter 8).
When installing factory management systems, many
benefits can result from integrating metal detection
equipment into these systems. A well designed factory
management system can include facilities for:
• Remote Management
- e.g. changing product information, re-setting
counters, etc.
• Remote Monitoring
- Monitoring of process events, e.g. reject data,
performance tests and pack counts
- Monitoring of running conditions, faults and warnings
- Communication of alerts and warnings
- Exporting of alerts and warnings to other devices, e.g.
email, pagers, PDAs etc.

84
• Data Collection and Recording 15.3 SCADA Systems
- Recording of performance data, test routines, etc.
- Providing data for product traceability Supervisory Control And Data Acquisition (SCADA) factory
- Providing proof of due diligence and compliance with management systems are becoming common place in many
industry regulations manufacturing environments. These highly customisable,
sophisticated systems can be used to provide data from
multiple processes at a single interface, either by direct
15.2 Connectivity Media communication with individual pieces of process equipment
– or via communication utilising OPC server technology.
Data from metal detection systems can be captured and
stored via a variety of technologies, two of the more popular
methods being: 15.4 OPC Technology
In most manufacturing plants, it is often preferable to
15.2.1 Serial Communications gather data from multiple processes and applications –
The traditional method of plugging a data cable into a serial and then view that data on the same computer screen
port (RS232 or RS422) and passing data through the cable or interface. However, this can be problematic, since the
to a PC or other collection device (Figure 15.1). A more various individual items of process equipment are likely to
basic entry level method of data collection, it is relatively communicate using different languages.
simple option to implement, however the collection of the OPC DA (and more recently UA) is one of the world’s fastest
actual data relies on individuals to support the process, growing standards for the exchange of process control data.
which could have additional cost and security implications.
The introduction of OPC DA (and UA) server technology
enables communications from multiple pieces of processing
Metal
Metal Detector
Detector
and packaging equipment in different software formats
to be distilled into a single, uniform common language.
This data can be transferred to a host of SCADA based
systems and factory management software solutions to
give total visibility of operational information. Advanced
RS232 RS422
RS232 / RS422 Computer With
Computer With metal detection systems can be equipped with Ethernet
Cable
Cable ‘Client’ Software
‘Client’ Software connectivity enabling the detector to be integrated
Figure 15.1 into factory communication data networks* providing
comprehensive collection of critical process information.
15.2.2 Ethernet Communications The OPC technology business benefits include:
Modern manufacturing plants frequently incorporate • Compatibility with SCADA, ERP and MES systems
Ethernet networks for the transfer and exchange of process • Simplified communications system design
and manufacturing data. Metal detection systems can be • Standard technology used across multiple
connected to such networks utilising a Fieldbus Interface production processes
Module (FIM) installed with the metal detection solution. • Facilitation of a standard and stable solution
This allows connection to a range of industry standard • Reduced dependency on multi-vendor solutions
protocols (EtherNet IP, Modbus TCP and Profinet IO) • A cost-effective solution (system integration, support
enabled devices such as Programmable Logic Controllers and training)
(PLCs) and Manufacturing Execution Systems (MES). Metal * This assumes that the appropriate application is also
detection data can then be viewed on secure networked installed onto the OPC DA (or UA) host PC/Server.
PCs or other devices (Figure 15.2) located anywhere in the
factory or company itself.
Networked PCs
Networked PCs
Ethernet
Ethernet Network
Network
Metal
Metal Detectors
Detectors
Figure 15.2

85
15
15.5 Data Management Software Advanced metal detection systems that are compatible
with statistical quality control (SQC) software such as
In a manufacturing facility where the connectivity those mentioned above can become even more valuable
infrastructure only allows for a pre-packaged software within a food or pharmaceutical environment. Powerful
solution to be deployed, companies can still benefit from statistical quality control software programmes can report
improved levels of data collection and traceability. on all aspects of packaging and filling quality control and
Proprietary software such as ProdX and FreeWeigh. can help reduce downtime and form part of a continuous
Net software solutions can be readily deployed on a process improvement framework.
standard IT infrastructure and are designed to provide real
time monitoring and reporting of all product inspection 15.6 Improving Production
equipment they are linked to. This type of software supports:
• The collection and storing of quality data and process Performance
visibility reports from your inspection systems. Data Advanced metal detection systems can now be supplied
can be analysed and presented in a range of standard Packaging Machinery Language (PackML) enabled and
report templates in order to support compliance needs. include on-board Overall Equipment Effectiveness (OEE)
Information can also be exported in CSV or XML formats for reporting. These new features support production efficiency
interpretation in other external data management systems. improvements by providing OEE data either at the device
• The collection of Performance Verification Routine or through various communications protocols into a
(PVR) data to meet compliance needs. The manual Management Execution System (MES).
collection of data from CCP’s can now become a thing of
the past. This type of software can collect and collate the
data from performance test routines and provide records
15.6.1 PackML
to prove due diligence has been exercised. PackML is structured using three categories of information;
• Real time rejects monitoring. Specific monitoring States, Modes, and PackTags. States are the most
screens within such software would allow you to track fundamental conditions seen in a production line. They give
all products which have been rejected from each device, the line-control Programmable Logic Controller (PLC) and
production line, area, or site. Accurate accounting of all other equipment in the line, knowledge of the condition of
non-conforming products is an important element in a the metal detection system. Modes are common forms of
robust quality programme. operational activity. PackTags provide the definition that
• System security and user management are a key allows modes, states and other administrative data (such
ingredients in ensuring a safe and secure system. as efficiency calculations) to be communicated to external
Such systems often provide a convenient and flexible devices or systems.
method of user administration. Alternatively, the use of a The PackML state model approach not only simplifies
Windows Active Directory Service feature and leave user machine-to-machine integration but enables communication
administration to the IT Department. of relevant data from production to the office. PackTags
• The change or adjustment of product inspection are named data elements used for open architecture
equipment settings remotely. Equipment settings can interoperable data exchange in packaging machinery
be stored for future use enabling set-up times to be from the “shop-floor to the top-floor”. This interoperability
significantly reduced. It is also possible to allocate the between packaging machinery and upper level systems
settings from one inspection device to another within offered by the PackML state model provides an easy way
your plant to further increase set-up efficiency. to integrate a packaging line and deliver OEE data from
• A
dvanced warning of line threatening problems on individual machines into business information systems.
your production line. The software should monitor the The data can be quite granular, rendering the production
status of each piece of product inspection equipment and process highly transparent, which is why end users are
display data in an easy to interpret way e.g. With colour increasingly interested in it. PackML builds off a proven
coded icons. Early warnings give you the opportunity to industry standard and the end user benefits include:
take evasive action to avoid costly downtime.
• Monitoring of events as they happen to enhance • A consistent look and feel for the operator and technician
productivity. Staying ahead of your inspection processes • A foundation for vertical and horizontal integration
by monitoring activities, warnings, and alarms that occur • Standard information in/out of any TR88.00.02 (formerly
on each device helps pro-actively maximise uptime. PackML v3.0) capable packaging machine
• Packaging line plug-and-play functionality
• More consistent end user specifications
• Faster software integration time
• Ultimately, efficiency in reusable hardware and
software components.

86
15
15.6.2 Overall Equipment Effectiveness 15.7 PackML and OEE Calculations

OEE is a recognised industry standard method for measuring PackML development focuses on measuring production line
and quantifying the performance of production line equipment. OEE and identifies a number of standard machine states
The ratio of actual output divided by maximum capable output that contribute to the OEE measurement. This machine
(shown as a percentage) is expressed through three primary data is combined with data from other equipment on the
factors: Availability, Performance and Quality. production line to yield the overall line OEE.
Table 15b gives examples of how the day to day use of
product inspection devices can impact a production line
Actual Good Output
= OEE % OEE calculation.
Maximum Capable Output In order to achieve continuous improvement, production line
managers have to assess OEE from every angle. A metal
detection solution can help them. Sophisticated on-board
Every piece of equipment on a production line can and analytical tools give them real-time data on production line
does impact the overall OEE percentage and metal detection OEE and metal detection system OEE. The data can be easily
systems are no different. extracted as advances make it available in various formats,
Three primary factors – Availability, Performance and and because the complete system is easily connected to a
Quality – are considered. production line network managed by a programmable logic
controller or SCADA/MES system.
• % Availability – actual uptime / planned uptime
• % Performance – actual throughput / planned throughput At any moment in time, managers can see how their line is
• % Quality – actual good product / total product produced performing. They have the data at their fingertips that invites
Table 15a gives examples of how the day to day use of a them to assess it deeper – to see where they can make the
metal detection solution can impact a production line’s critical changes that raise their OEE ever closer to a world-
OEE calculation. class target.
How metal detection Impact on Impact on Impact on Typical metal detection system
systems can affect OEE Availability % Performance % Quality % feature sets to minimise impact
Production line downtime Yes Yes No Reliable construction, Auto Balance,

due to detector system failure Enhanced Noise and Vibration
or fault Immunity and Condition Monitoring
reduces potential detector downtime
Detector system stopped due Yes Yes No Intuitive product clustering eliminates
to product changeover product changeover time
False rejects due to incorrect No No Yes Accurate Auto set-up and Optimised
set-up set-up virtually eliminates false rejects
Lost production time and Yes Yes No Predictive Analytics* and enhanced
products destroyed during PVR can support a reduction in the
PV testing frequency of PV testing
Packs rejected due to metal No No Yes Ultra high, multi and variable frequency
contamination operation delivers the highest, yet most
reliable levels of sensitivity
* Predictive Analytics features monitor the impact of any changes in a metal detector’s sensitivity levels. An early warning
is usually given before such performance levels fall below a defined factory specification. In conjunction with Condition
Monitoring features, which monitor the metal detector’s critical parameters, these features can be used to reduce the amount
of system testing. Reducing the frequency of testing can increase operator efficiency and production line capacity.
Table 15a: How metal detection systems can affect OEE

87
15
Product Inspection (PI) Impact on OEE Impact on Impact on Impact on

Availability Performance Quality
Packages rejected due to contamination
Packages rejected for specification violation
Successive package faults
PI device stopped for product change over
Production time and product loss due to PVR* testing
False rejects due to improper settings
Production Lines stop due to PI device failure or fault
*PVR, performance validation routine, is a software feature in our metal detector and X-ray machines
where the user can set up a time interval to alarm an operator when it is time to run a performance test.
Table 15b: Product inspection impact on OEE

88
15
Notes

89
16
Calculating the Total Cost of Ownership (TCO)

for In-line Metal Detection Equipment
Sound investments are the foundations of a company’s successful future. It takes careful planning and implementation
to ensure that those investments reap significant rewards – and it is important to make decisions with respect to these
investments on the basis of well-informed strategy, rather than on impulse and assumptions.
16
Calculating the Total Cost of Ownership Based on these principles, the process of assessing the
Total Cost of Ownership (TCO) was developed over 20
for In-line Metal Detection Equipment years ago in the United States, and has since been used by
16.1 Defining Reasons for Investment Decisions many companies worldwide. Consideration of TCO helps to
understand more clearly the true costs of owning items, over
16.2 Consideration of Operating Goals and above their purchase price. In addition, TCO helps when
comparing suppliers’ different offers – plus it can provide
16.3 The Importance of Correct Calculations
substantiation for a final investment decision.
16.4 Basing Investment on a Solid Business Model It is important that the TCO assessment methodology is
16.5 Ascertaining the Costs always oriented to the individual circumstances of the
company and the industry within which it operates. This
16.6 Consideration of TCO Over Several Years also applies to product inspection equipment – specifically,
metal detection equipment.
16.7 Benefiting from Potential Savings
This chapter addresses the following matters with
16.8 M
etal Detection – a Considered Investment respect to TCO:
Can Be Very Worthwhile • Defining fundamental reasons for investment decisions
• Understanding the advantages of a TCO assessment
16.9 The Need for Support from the Supplier
• Identification of critical dynamic product inspection
equipment costs
• How to save money by specifying the correct in-line
metal detection equipment
• Calculating the Return On Investment (ROI) time of
an investment
• Gaining the support of the metal detection
equipment supplier

90
16.1 Defining Reasons for If possible, all the costs connected with the new purchase
should be taken into account – and TCO can help
Investment Decisions companies to identify such costs, all the way from the
purchasing process through to final disposal.
Any investment in plant and equipment needs to be made
on the basis of clear and fundamental principles. In the food TCO assessment as the basis for management and
and pharmaceutical industries, metal detection systems project decisions
are accepted as an effective and invaluable component in
An investment made for a specific department within a
ensuring that only premium-quality products are offered
company can sometimes have an effect on different parts of
to clients and end customers; however, the installation of
the organisation. Furthermore, certain investment decisions
metal detection equipment, and its associated costs, do not
can compete with different financial priorities, so other projects
automatically guarantee that better products are made.
must be postponed. Consequently, conflicts can arise between
A clear understanding of operating targets and requirements business management and project management.
makes a vital contribution to the basic principles that
However, thanks to TCO considerations, accurate budgeting
underpin investment decisions. Closely connected to
– over the entire service life of an investment –can make it
these principles is the knowledge obtained through TCO
clear (particularly to senior management) which financial
assessment – in other words, the Total Cost of Ownership
expectations will be satisfied by the investment. Decisions
of the investment over its service life, together with the
then become easier and conflict between departments can
expected income to be derived from the investment.
be prevented.
TCO assessment as a protection against hidden costs
16.2 Consideration of
The selection process demands transparency with respect
Operating Goals to costs. A major reason for introducing TCO assessments
Operating goals should be clearly defined before introducing is to create a better basis for choosing a supplier. An
a product inspection programme. Operating goals form investment commitment that is based exclusively on a good
the basis for taking first steps towards investment, such as purchase price can turn out to be financially disastrous over
making clearly formulated enquiries to potential suppliers. the course of time.
Typical operating goals include: Hidden costs can only be recognised in ongoing operations,
and can result in an apparently low-cost item causing
• Introduction of a new system major problems. If the design makes modifications difficult
• Improvement of production time by a given percentage, and costly, or if frequent maintenance intervals require
so as to improve brand protection frequent stoppages, the costs can rise significantly. With
• Compliance with national, international and/or global TCO assessments, hidden cost drivers can often make the
standards e.g. the Global Food Safety Initiative (GFSI) difference even before reaching an investment decision.
• 100% of the products are to be inspected in future When all the costs of an investment are taken into account,
• Reduction of waste rate due to over/under filling of a this will mitigate against hidden costs and will help to
given percentage protect against any negative effects they may have.
• Reduction of false rejects by a given percentage
• Reduction of quality assurance costs by a
given percentage 16.4 Basing Investment on
• Increasing on-line performance (accuracy and/
or sensitivity)
a Solid Business Model
Industries and companies are distinguished from each
Each target should be measured with the aid of traceable
other in many ways, but one rule applies to all businesses:
data, and each goal should be presented in terms of its
a solid business model should be created before making
financial implications for the company.
an investment. The TCO calculation supplies solid evidence
for the Total Cost of Ownership for management and
16.3 The Importance of decision-makers.
Correct Calculations It is appropriate to take into account the particular
circumstances of each industrial sector and company.
Decision-makers are expected to track investments and
estimate them precisely. Acquisition costs do not usually When tailored to product inspection technology, it is vital (in
present a problem; in fact planning them within the the TCO consideration) to highlight the savings that result
context of an investment decision is usually simple and from an investment. TCO is also important when assessing
straightforward. However, calculations need to include the various costs connected with the system and process.
critical costs for the entire course of usage, which requires Only such a procedure leads to an accurate overview of the
a systematic approach to evaluation. Total Cost of Ownership – and thus to a clear assessment of
the Return On Investment time.
Cost components can easily exceed the purchase price.
Machines must be maintained, supplied with power or
additional materials, and employees must be trained – often
for a machine lifetime that can easily exceed a decade.

91
16
Calculating the Total Cost of Ownership for In-line Metal Detection Equipment
16.5 Ascertaining the Costs 16.6 Consideration of TCO

TCO reveals the costs of an investment over its entire life Over Several Years
cycle, so all expenses connected directly or indirectly with
the investment are added into the calculation. When examining a TCO analysis, the first 12 months, and
subsequent years, are critical. Product inspection TCO
Direct costs distinguishes between the costs of the initial investment (the
Direct costs are normally easy to determine, since they first year) and the costs in subsequent years. The first year
are the follow-up costs that can be directly attributed to an after purchase is the most cost-intensive, with purchase
investment, such as: price, installation, training, spare parts packages and
integration into the production line all having a significant
• Procurement costs
impact. In some cases, external consulting costs or the
• Software update costs
disposal of obsolete machinery must also be considered.
• Machine operation costs (energy, compressed air, etc.)
• Wear and tear parts costs Subsequent years have their own cost aspects; in addition
• Training costs to operating and maintenance costs, unplanned downtimes
• Service contracts, maintenance and calibration costs and costly guarantee/warranty extensions can be expensive.
• Labour cost (line operator, quality engineer, operator If official verification certification is required, costs for
and testing) conformity assessments and official tests need to be taken
into account. It should be borne in mind that the service life
Indirect costs
of a machine now can be ten years or even longer.
Indirect costs are more difficult to determine, since they
cannot be attributed precisely to one investment, and Costs due to the implementation of a product
usually arise if productivity is impeded in connection with inspection programme
the investment. For example: When considering initial investment costs, the following
• Failure times arise from lack of maintenance, repairs etc., should be borne in mind:
and the system is not available • Purchase price: a basis is provided by the quotations
• Incorrect machine settings lead to poor performance, from various manufacturers who were invited to bid.
or production stops • Installation/commissioning (initial operation): the
• When faults occur, colleagues are brought in from other relevant quotations indicate the external costs for support
departments in order to resolve the problem by service providers, consultants or fitters. Internal costs
• False reject costs are determined through in-house hourly rates or charge
• Management of false reject costs (FSMA) rates. The critical figure is the total time required, starting
• Production down time due to testing from the stoppage of the production line for the purpose
• Wasted product due to testing of installation (i.e. integration of the equipment solution)
• Costs associated with product changeover through to the resumption of production.
Examination of the TCO assessment offers an excellent • Validation documents: costs for validation and
basis for decisions made in favour of an investment, and certification (e.g. according to the Global Food Standard
in favour of a specific supplier. Whilst such decisions are can be supplied by the manufacturer concerned.
based on quantifiable factors, it must be borne in mind that • Costs for official verifications: a competent supplier will
other, more subtle, non-monetary factors may also play a indicate all the costs for necessary official verifications,
role in decision-making. These factors can include: from support through to official testing.
• Training with the supplier - or on-site with the system:
• The reputation of an equipment supplier
the costs are quantifiable, because training is offered
• Service offerings
directly by the equipment supplier.
• Quality of service
• Purchasing costs for spare parts: a competent
• Availability of consumables and spare parts
supplier can make appropriate statements regarding
As it is often difficult to quantify indirect costs, estimates spare parts that may be required in the first year and
should be used on occasions when a calculated value is in subsequent years.
not available. It is, however very important that these costs • Service offerings: a number of suppliers offer service
are taken into account – and their identification may show agreements which include various services, such as
areas within the company where other improvements can inspections, maintenance visits and spare parts. The
be made to reduce such expenses. various offers for service reaction time should also be
considered, such as inclusive services, price discounts
Table 16a at the end of this chapter outlines key cost
for spare parts, and remote diagnostics options. Remote
considerations for calculating the TCO of a typical installation.
diagnostics/remote maintenance reduces costs because
aberrations can be detected earlier. As a result, any
intervention by technicians can be prepared for, allowing
problems to be rectified more quickly.

92
16
• Integration into the production line: the expense of Important items in service agreements worthy of particular
integration can vary according to the circumstances, and attention include:
consideration must be given to whether the equipment • Whether the agreement includes all the necessary visits
is being introduced for the first time or whether existing and services
equipment is being replaced or expanded. Producers can • Whether the agreement include a single lump-sum payment
be helpful within the parameters of their capability, and (regardless of how much service work becomes necessary)
will be able to highlight potential for optimisation. • Whether the agreement also includes the cost of spare
• Disposal of old equipment: on request, a supplier can parts plus wear and tear parts
take responsibility and provide a firm price. • Whether the agreement also includes all travel costs and
When viewing costs over subsequent years, the following the technicians’ hourly rates
must be taken into consideration: Table 16b at the end of this chapter provides a useful
• Operating costs: the cost of energy and additional table you can use to capture the ongoing costs as part of
materials may vary widely. The supplier should have the a TCO calculation.
corresponding technical information to hand.
• Maintenance costs: The supplier can specify
maintenance intervals and expenditures. A competent
16.7 Benefiting From
supplier should be able to quote average values, as Potential Savings
related to repairs.
• Unplanned downtimes: an overview of the past (and Whenever an investment is made, the cost is always of
the calculations that were previously made) is the most crucial importance to management and decision-makers.
helpful guide. In many case, these can be the single However, the consideration of savings (especially over
biggest costs incurred by the user, especially when the the entire lifetime of equipment), can be decisive when
line cannot run without the inspection equipment on line considering an investment in a new system. It is worthwhile
and functioning in accordance with specifications. including the different savings potentials and allowing them
• Guarantee/warranty extension: the supplier will provide to influence the TCO calculation.
the corresponding quotations and prices.
• Software/hardware updates: the supplier will provide 16.8 Metal Detection – a Considered
information concerning frequency and costs.
• Personnel costs: time required for the creation/set- Investment Can Be Very Worthwhile
up of new products and switching of product set-ups Selecting the correct make and type of metal detector
(changeover). There are major differences between the can result in considerable cost-savings .Production
solutions of various suppliers under these circumstances. practice shows that the greatest savings are achieved
An estimate should be made of either how often a by the reduction of waste (false rejects), the reduction of
completely new product needs to be set up on the operational costs (product changeovers and performance
production line or how often the product set-up must testing) and by increasing uptime through improved OEE
be changed over. and speed of service provision/first time fix rates.
• Test costs: these include not only the personnel costs
associated with the time spent to undertake the test and Cost savings:
the cost of product destroyed due to testing. The annualised cost-saving from one make/model to
• Unplanned downtime another can be as much as $35,000.00.
• Planned downtime (the actual testing)
The savings potential:
• P roductivity inefficiency as result of lost production
due to the line stopping or slowing as the scheduled • Using a metal detector that is more reliable and less
performance test is carried out. likely to suffer electronic drift
• Using a metal detector that is less likely to suffer from
Normally, every equipment supplier is convinced that their external influences that can result in the rejection of
own system is the fastest and easiest to set up. However, good product
suppliers need to show manufacturers exactly which • Using a manufacturer who can demonstrate an uptime
working steps are necessary, and then manufacturers percentage of 0.01 better than their nearest competitor
can decide for themselves whether the operation is time- • Using a manufacturer whose speed of response for
consuming and requires considerable personnel input, or servicing, and whose first-time fix rate are at least two
whether it saves time and costs. hours faster than the next best supplier (the greater the
time difference, the greater the cost savings)
• Using a metal detector that allows the frequency of
testing to be twice that of an alternate make/model
• Using a metal detector that can be set to prevent/
minimise the need for operator intervention each time
product changeover is initiated.

93
16
Practical example: Line Speed 100ppm, 20 hrs/day During the implementation of a product inspection
production runs over 250 days per year, with a typical programme, savings can be achieved by:
production operative cost to be £40.00 per hour. 1. Reduction of scrap
A) A typical metal detector has a false reject rate of one Accurate and reliable metal detection equipment ensures
pack per 5,000 produced (99.98% efficient). This the implementation of statutory regulations, and thereby
equates to 6,000 packs per year that are wrongly prevents expensive scrap. The financial benefit can be
rejected. Assume a product/pack cost of £1.00; this estimated by comparing the ‘before’ and ‘after’ rates.
equals an annual cost of £6,000.00. Compare this to a
metal detector that has a false reject rate of one pack per 2. Reduction of rework
10,000 packs produced (99.99% efficient). Savings add Additional work that arises from rejected products can be
up to £3,000.00 per year. calculated from additional personnel costs.
B) A metal detector with a 99.8% uptime rate (0.2%
3. Reduction in the cost of working time
downtime rate) will be down for 10 hours per year. If the
lost production costs are £2,000.00 per hour, the down- The supplier provides information on the time for product
time cost is £20,000. Alternatively, if the uptime of an setup/changeover (refitting) and on cleaning times.
alternative metal detector is proven to be 99.9% (0.1% 4. Reduction of ‘wasted’ material
downtime rate) the downtime costs are £10,000.00 per
The costs for over rejection upon detection can be
year; alternatively, the saving is £10,000 per year. If
calculated based on a simple before and after calculation.
organisation B has just a two-hour better response/first-
time fix rate than organisation A, an extra £4,000.00 5. Prevention of product returns
is saved. Modern metal detection programmes inspect 100 percent
C) The single biggest cost associated with metal detection of the products manufactured. Deviations that run contrary
equipment is that associated with the mandatory to official regulations or industry standards are detected
requirement to carry our frequent performance as early as possible and avoided. Potential savings are
verification (PV) testing. Assume that it takes 10 minutes calculated by comparison with previous production and
each time a PV test is carried out (walking, talking, the cost of product returns.
testing and documenting) and a test is done every two
hours. This equates to an annual cost of £16,666.66. 6. Protection of the brand and customer relations
Compare this to a metal detector fitted with due diligence Non-material values, such as brand and consumer loyalty,
fail safe technology that makes acceptable the risk can be difficult to estimate. However, they form the basis
associated with a frequency reduction from two-hourly for reinforcing a repeat sale process, as well as attracting
to four-hourly. The saving is £8,333.33 per year, which new customers.
does not take into account the cost of wasted product. In
some cases, this can be as much as (or more than) the 7. Reduction of the expense involved in audits
cost of the time associated with the testing. (e.g. for IFS, BRC and others)
D) Many metal detectors inspect more than one type The preparation of equipment tests and audits (and their
of product on the same production line, making it subsequent documentation) can be time-consuming and
necessary for the operator to switch from one setting to costly. The equipment-supplier should be asked for a
another each time a product changeover occurs. Take documentation scheme that records all the relevant tests
a production line with three different products, with and audits, whilst also keeping this documentation up to
each product run once per 20-hour shift. Assume it date. Using this procedure, the proper operation and use
takes the operator five minutes to change over the metal of equipment, both internally and for the requirements of
detector settings (walking, talking, changeover and external auditors, can be easily documented at any time.
documentation); the cost is £2499.99 per year. More Table 16c at the end of this chapter provides a framework
modern metal detectors allow products to be clustered for calculating the possible savings associated with a
on a single setting to allow operator-free running (within typical metal detection system installation.
the specification requirements laid down by QA). The
effect of this is to totally remove this cost from the
equation. Total cost savings, as detailed above, are
£27,833.32 or £139,166.60 over a five-year period .

94
16
16.9 The Need for Support

from the Supplier
In the course of carrying out a TCO calculation and
considering the potential saving resulting from the
investment, there is a requirement for a wide range of data
and figures that are relevant to the useful life of the system,
from purchasing to disposal.
Competent suppliers therefore represent an important
source of information for the values that will be entered
into the calculation. Suppliers of machines and equipment
should be willing to help and provide relevant information.
In particular, this includes indications of operating and
maintenance costs and of unplanned downtimes. In
addition, active support at the investment planning stage
can be an important criterion when selecting the supplier.

95
16
Calculating the Total Cost of Ownership (TCO) for a Metal Detection inspection system
Manufacturers should enter their values:
Costs of initial investment Year 0 Year 1 Year 2 Year 3 Year 4 Year 5
Purchase price – – – – –
Installation/commissioning
– – – – –
(initial operation)
Validation documents – – – – –
Official annual verification

– – – – –
costs (If applicable)
Training with the supplier or

– – – – –
on site with the system
Procurement costs for

– – – – –
spare parts packages
Service contract – – – – –
Integration into the

– – – – –
production line
Disposal of old equipment – – – – –
Other – – – – –
– – – – –
– – – – –
Total – – – – –
Table 16a: Overview of the initial costs of a typical metal detection system installation for calculating TCO

96
16
Subsequent years Year 0 Year 1 Year 2 Year 3 Year 4 Year 5

(generally up to 5 years)
Operating costs –
Maintenance costs –
Unplanned downtimes –
Guarantee/warranty
–
extension
Annual verification costs

–
(If applicable)
Software/hardware
–
updates and support
Personnel costs –
Service contract –
Mandatory schedule user

–
performance testing costs
Other –
Total –
Table 16b: Ongoing costs of a metal detection system installation for calculating TCO

97
16
Savings Year 0 Year 1 Year 2 Year 3 Year 4 Year 5
Reduction of waste
Reduction of rework
Reduction of work required
Reduction of
product ‘give away’
Avoidance of returns
Protection of brand
reputation and customer
relationship
Reduction of expense
involved in audits
Reduction in lost downtime
Reduction in the costs

of mandatory testing
Other
Total
Table 16c: Overview of the possible savings of a typical installation:

98
16
Notes

99
17
The Principles of Due Diligence For Quality Control and

Legal Defence
Increasingly, food manufacturers must take more precautions to ensure that their products are safe,
free from contamination and unlikely to harm the end consumer in any way.
17
The Principles of Due Diligence For As highlighted in Chapters 8 and 9, HACCP (Hazard
Analysis and Critical Control Points) leads the way in
Quality Control and Legal Defence providing a framework within which food manufacturers
17.1 Duty of Care can work, whilst the Global Food Safety Initiative (GFSI)
manages and controls the bodies that can provide
17.2 Due Diligence certification and accreditation.
17.3 Hazards Analysis Critical Control Points The organisations and bodies that provide GFSI recognised
schemes include:
17.4 Instances of Metal Contamination
• The British Retail Consortium (BRC)
17.5 A Metal Detection System: Concerns and Solutions • The International Food Standard (IFS)
• FSSC 22000
17.6 Components of a Fail Safe Metal Detection System • SQF
• Dutch HACCP
Other schemes exist, but those listed above probably equate
to over 90% of current adopted standards.
Consideration also needs to be given to local regulations
that may affect food manufacturers selling into a particular
market, such as the FDA’s Food Safety Modernization Act
(FSMA) in the USA and the Brazilian Health Surveillance
Agency (ANVISA) in Brazil for example.
In these litigious times, lawyers and consumers will seize
any opportunity to take legal action against manufacturers
if they find something wrong with a product they have
purchased. For these reasons, food manufacturers who
supply retailers will understand the need to ensure that
product quality is of the highest level, so it is in the best
interests of manufacturers to ensure that systems and
procedures are in place to minimise the risk of litigation. In
the event of such an instance, they should also have the
necessary documentary evidence to prove they have been
duly diligent in the manufacturing process.

100
17.1 Duty of Care 17.4 Instances of
In law, every individual has a Duty of Care which requires Metal Contamination
adherence to a standard of reasonable care while performing
any acts that could foreseably harm others. The Standard of The manufacturing environment (and general food
Care is the degree of watchfulness, attentiveness, prudence processing) can create the risk of metal contamination. A
and caution of an individual who is under a Duty of Care. metal detector often acts as a Critical Control Point to mitigate
this risk. A number of additional elements should be included
In the food industry, the Standard of Care is determined by in the process in order to safeguard customer welfare and
the standard that would be exercised by a reasonably prudent provide the basis for a robust due diligence defence.
manufacturer of a product. Failure to meet the standard could
be regarded as negligence, and any resulting damages may Furthermore, a suitable metal detection system will allow
be claimed in a lawsuit by the injured party. manufacturers to deliver the very best level of consumer
and brand protection. All conveyor systems used to inspect
products should be specifically designed to do just that, and
17.2 Due Diligence not simply provide proof that metal detection equipment is
on the line and functioning.
The Due Diligence defence is available to manufacturers
accused of a breach of food safety regulations. Essentially,
the defence is that the ‘accused’ took all reasonable 17.5 A Metal Detection System:
practicable steps to avoid the breach. It is a sufficient
defence for the person charged to prove that: Concerns and Solutions
• All reasonable precautions were taken The opportunities for metal to find its way into a food
• They exercised all due diligence to avoid the occurrence, product are numerous, because the majority of equipment
whether personally or through any person under their used in food processing plants is made of metal. For
control. ‘Taking all reasonable precautions’ includes example, cutting blades, grinders, mixers, transport
setting up systems of control which are appropriate to the conveyors and packaging machinery are all predominantly
risk. What is ‘reasonable’ is determined by the size and metal-based, as are hand tools, machinery structures and
resources of the business. ‘Exercising all due diligence’ support frameworks.
involves having procedures in place which review and
It is conceivable that some of these items could shed a
audit the system to ensure that it is operating effectively.
small piece of metal into the manufacturing process under
Whether or not a defence will be successful depends on
normal working conditions, without the equipment failing;
the circumstances surrounding each case.
however, a metal detector downstream of all processes
ensures that the resulting food product has been checked
17.3 Hazards Analysis for the inclusion of metal.
Critical Control Points Metal detectors are a common sight in most modern food
manufacturing plants, and the technology employed is
In food production, most manufacturers utilise a HACCP considered to be highly reliable. However, the incidence
based system as a framework to identify where hazards of metal reaching the end consumer remains high. More
might occur. The HACCP structure is then used to put alarming is the fact that, upon investigation, the metal causing
into place procedures to minimise the risk of the hazard complaints is invariably detectable by on-line equipment.
occurring in the first instance.
This suggests that operational procedures in place in the
The HACCP process strictly monitors and controls each manufacturing or inspection process may be at fault – and
manufacturing step, so as to reduce the probability of simply installing a metal detection system will not eradicate
hazards occurring. HACCP is based on seven core principles: the incidence of metal reaching the end user.
• Conduct a food safety Hazard Analysis
A broad approach to Quality Management must be
• Identify the Critical Control Points (CCPs) (point at which
employed and, as many metal detectors are defined as
a hazard is optimally controlled)
Critical Control Points (CCPs), it seems common sense that
• Establish critical limits for each CCP
this CCP is managed accordingly.
• Establish CCP monitoring requirements
• Establish corrective actions when monitoring indicates A metal detection system fitted with a suitable reject
that a particular CCP is not under control mechanism and lockable reject bin will go a long way
• Establish record-keeping procedures towards providing a solution; however, system and
• Establish procedures to verify system is working procedural failure can have a serious impact on the overall
as intended effectiveness of the system employed.
All contaminated food packages should be rejected efficiently
from the process or packing line (and remain rejected),
and the highest levels of compliance with the necessary
standards should be met. To help achieve these aims, the
table 17a identifies relevant concerns and solutions.

101
17
The Principles of Due Diligence For Quality Control and Legal Defence
Concern Solution
How can I ensure that metal contamination is detected in Install a metal detector that is capable of detecting all metal
accordance with the highest levels of performance? types and understand its ability to detect non-spherical
contaminants, such as wire and thin slivers of metal.
Metal detection system failure leads to costly downtime. Undertake a preventative maintenance programme on the
How do I maximise uptime? conveyor system, whilst ensuring the metal detector has a
built-in condition monitoring system which can give an early
warning of potential downtime.
How do I ensure the metal detector is set correctly and Ensure that the metal detector has an accurate auto
that false rejections do not occur? set-up feature that displays the margin of safety between
the background product signal and the metal detector’s
trigger point.
If metal is detected, how can the contaminated pack be rejected Utilise an automatic pack reject mechanism that has been
from the process without causing production stoppages? designed specifically for the application in question.
How can I ensure that consecutive contaminated packs are Ensure that the reject mechanism is working in conjunction
rejected? And how do I guarantee that the correct pack is with a pack sensor, which controls the operation of the reject
rejected, irrespective of the position of the metal within the pack? mechanism and the metal detector.
How can I ensure the pack sensor is working? Fit a reject check sensor that permanently monitors the
pack sensor
How do I ensure that I have a sufficient supply of compressed Fit an air reservoir to the system or fit an air failure switch to
air to deal with multiple reject events? the conveyor’s pneumatic feed.
How do I ensure that the reject mechanism is functioning The timing of the photo-gated reject mechanism (as described
correctly when the conveyor system runs from a variable above) must be controlled via a belt speed encoder to ensure
speed drive? accurate rejection, irrespective of belt speed.
How can I ensure contaminated product is not removed from Install a system tunnel guard from the detector to a point past
the line after detection but before rejection? the reject device.
Where should the contaminated packs be collected when they Inside a secure reject collection area with controlled access
are rejected? via key or password to enhance product safety a bin door lock
monitor can be included. .
How can I ensure that the contaminated pack has been rejected Install a reject confirmation system linked to both the pack
from the process or packing line? in-feed sensor and metal detector.
What if the reject bin gets full of contaminated product and there Install a Bin Full sensor at 80%-full level, with an alarm if
is no more room to accommodate further rejected product? the situation becomes critical.
Table 17a: Concerns and solutions

102
17
17.6 Components of a Fail Safe Many types of reject mechanisms are available, and most
are pneumatically-operated (such as air-blast mechanisms,
Metal Detection System pushers, sweep arms, etc). Such pneumatically-operated
reject systems may be fitted with an air failure switch, which
17.6.1 Metal Detector will raise an alarm if air pressure falls below a critical point
that could prevent efficient rejection. Air reservoirs can also be
A satisfactory metal detector should meet required detection fitted, to increase the overall fail safe nature of pneumatically
standards. It must be set up to operate within the sensitivity operated reject systems.
guidelines detailed in either the manufacturer’s code of
practice or in accordance with the requirements of third-
party customers such as retailers.
17.6.3 Pack Sensor and
The general rule governing metal detection performance is
Conveyor Belt Speed Encoder
that as the size of the aperture decreases, the performance These work in conjunction with the reject device and metal
improves. So, in general, the aperture size chosen should be detector. Together, they determine the exact position of
based on the maximum size of the product being inspected. a contaminated pack on the conveyor belt, so that the
pack is removed successfully from the line. The pack
When comparing the sensitivities of different metal
sensor identifies the presence of each pack at known fixed
detectors, these should not be made simply on their ability
distances from the metal detector and the reject mechanism.
to detect spheres of metal; comparison should also be
made with their ability to detect non-spherical types of Using the metal detector’s built-in timer without the use of
contamination, such as wire and fine slivers of metal. an additional pack sensor is not recommended. Failure to
use a photocell is, potentially, the single biggest reason for
Fault monitoring is standard with many metal detection
contaminated products still reaching the end consumer. This
Detector Heads, and if a fault should occur, the metal
is because the timing of the reject mechanism can vary,
detector will alert the user to the problem and shut down
depending on the position of the metal within the product
the system. The disadvantage of fault monitoring is that the
and the actual size of the contaminant.
system could be out of use until the fault is fixed.
This can make the timing of the operation of the reject
More advanced detectors utilise Condition Monitoring
system prone to variation – and potentially, it could fail to
technology, which is consistent with HACCP monitoring
accurately reject the correct contaminated product. Together,
requirements. Condition Monitoring checks that the critical
the combination of the external pack sensor and the built-in
elements of the metal detector are working, and measures
reject timer ensure far greater levels of successful rejection.
any changes that could lead to a reduction in performance
or to a detector fault. If using a conveyor system that utilises a variable speed
drive, a belt-speed encoder should be used in conjunction
Before these changes become critical, an early warning
with the pack in-feed sensor to control the operation of the
system brings the changes to the attention of the user. This
reject mechanism.
allows maintenance to take place, avoiding the potentially
high costs of lost production through line downtime. This ensures that the time between metal being detected and
Planned corrective actions can take place when the system the reject mechanism operating is calculated accurately,
is scheduled to be off-line. enabling the reject mechanism to identify the contaminated
pack, irrespective of line speed. This is also a requirement if
17.6.2 Automatic Pack Reject Mechanism the line is prone to frequent stopping and starting.
Where possible, the system should include an automatic

product reject mechanism, which is activated when the
17.6.4 Lockable Reject Collection Bin,
metal detector has identified metal contamination. Its Reject Confirmation Sensor, and
purpose is to remove the contaminated pack(s) from the Bin Full Sensor
production line prior to despatch.
The purpose of the reject collection bin is to provide
The rejection mechanism should be designed specifically for temporary storage of rejected (i.e. contaminated) packs.
the products being inspected, so it will depend on application The bin must be lockable, to make sure that contaminated
requirements. The rejection mechanism should take into packs cannot be removed and re-introduced further down
account line and pack speed, pack weight, pack shape and the production line, downstream of the inspection system.
dimensions, as well as the nature of the packaging material. The key for the lock should never be left in situ, and should
This not only ensures maximum rejection capability, but also be held by a senior/authorised staff member.
takes away reliance on line operators – which can often be
the biggest area of system failure. It is recommended that
the use of a ‘stop-alarm and manual rejection’ type system
should only be specified in extreme circumstances.

103
17
Removal of the key from the locked bin eliminates the The occurrence of these events assumes that the reject
possibility of non-authorised personnel gaining access confirmation system has not failed as well. However,
to contaminated product – which is consistent with Due waiting for the system to fail is contrary to good working
Diligence and HACCP principles. Advanced metal detectors practice, and would result in the quarantine and re-
can be configured to activate a timer when the reject bin door inspection of all product inspected since the last
is opened; in addition, they can automatically shut down the successful performance verification test.
system if the bin is inadvertently left open for more than a
The addition of the reject check sensor provides real-time
pre-set time. Systems can also be supplied that replace the
monitoring of the pack in-feed sensor and vice-versa. If
need for a physical key with an unlocking password.
failure were to be identified by either sensor, the system
These security processes further enhance the security and would issue an alarm, allowing the necessary corrective
integrity of the reject bin, since only authorised personnel action to be undertaken.
can gain access. A reject confirmation sensor should be
As well as providing a health check of the in-feed pack
situated in or across the mouth of the reject bin – and once
sensor, the reject check sensor also acts as a back-up to
metal has been detected, the system can be configured to
the reject confirmation system; this dramatically increases
expect a further signal (from the reject confirmation sensor)
the overall fail safe nature of the entire system. There are
that a pack has entered the reject bin.
some unlikely scenarios in which the reject confirmation
If no such signal is received, a system alarm is raised and system has been satisfied, yet the contaminated product
the conveyor is stopped. The reject confirmation system can still be allowed to travel down the production line.
must be intelligent enough to handle multiple detection
For instance, it is possible that the contaminated product
events, whether they are:
pack could somehow bounce out of the reject bin after it has
• Detection events caused by multiple packs containing been confirmed as being successfully rejected. If this were
metal, or to occur, the reject check sensor would act as a back-up to
• Multiple detection events caused by one or more large the reject confirmation system, because it would expect the
pieces of metal contaminated pack to have been rejected. The reject check
A ‘Bin Full’ sensor removes the risk that a contaminated sensor would then issue an alarm when identifying a pack
pack might not be removed from the conveyor because where a gap should be. Subsequently, a fault condition would
the reject bin is full of rejected product. Once the level bin be created, and the conveyor would stop.
approaches its capacity (recommended to be set at 80%
full), an alarm can be activated (or the conveyor can be 17.6.6 Key-operated Reset Switch
configured to stop) so that the bin can be opened and the
All the fail safe elements that stop the conveyor should
reject packs can be removed for disposal. This avoids the
be linked to a key-operated reset switch, rather than a
risk of a failed rejection due to the reject bin being full.
push-button. Only authorised and nominated key-holders
A tunnel guard or enclosure should be fitted to the out-feed should be allowed to re-start the system after the fault
side of the system. As a minimum, this should extend from or condition has been rectified. The key should never be
the out-feed side of the metal detector to a point beyond left with the system, and it should only be held by the
the end of the reject bin. The purpose of this guard is to authorised key-holder.
prevent unauthorised removal of contaminated products
from the system; this prevents contaminated products from
being accidentally re-introduced to the system after the
17.6.7 Warning Beacon Stack
point of rejection. A warning beacon stack attached to the metal detection
system can signal warning faults. It is usually a high-
visibility colour-coded fault beacon, enabling rapid
17.6.5 Reject Check Sensor identification and rectification of the problem.
For the reject mechanism to perform accurately, the metal
This will help to ensure that downtime is kept to a minimum.
detector and pack in-feed sensor both need to function
Audible alarms can also be configured to activate when the
100% of the time. If the metal detector should fail, the built-
warning beacon operates. It is recommended that if any of
in fault monitoring system would stop the conveyor.
these fault conditions occur during normal manufacturing,
If the pack in-feed sensor should fail, the reject mechanism the process should cease immediately until:
would be inoperable. Because no reject confirmation • The fault condition is rectified
signal would have been received, the reject confirmation • The system has been validated by the appropriate system
sensor would identify this failure the next time that metal test procedure
was detected. • The system has been documented as fully functioning by
the appropriate system test procedure

104
17
17.6.8 Access Log and

High-Security Log-in Facility
Sophisticated metal detection systems can help the user to
comply with standards; it can also provide a valuable audit
trail. This is achieved by issuing unique single-user pass
codes, and by making these pass codes language-specific.
This process ensures that each user carries an appropriate
level of personal responsibility for his/her actions.
A system of this type is normally sufficient to prevent
misuse, and supports the needs for regular inspections, so
as to provide the basis of a Due Diligence defence. In such
systems, an automatic log is produced, recording all log-
ins made at the metal detector, as well as detailing the date,
time and name of the person logging on.
By recording this information, and by instituting system
access only through individual password control, this
demonstrates compliance with standards and HACCP
record-keeping requirements, and forms a robust basis
for a Due Diligence defence.
17.6.9 Management Responsibility

Since many metal detectors are considered to be CCPs, it
is a management responsibility to ensure that all personnel
treat these Critical Control Points accordingly. Operators
must be aware that their actions are essential to the
operation of the control point – so any misdemeanour’s
should be subject to disciplinary action.

105
18
Understanding Challenging Applications for

Improved Metal Detection
It is well documented In the food industries, many products that are inspected for metal contamination exhibit a
phenomenon known as ‘Product Effect’. This effect means that the type of product being inspected can, itself, hinder
the ability of inspection technologies to identify that particular contaminant.
18
Understanding Challenging Applications Other inspection technologies may refer to this phenomenon
differently – but for metal detection, products are categorised
for Improved Metal Detection as being either ‘wet’ or ‘dry’. This chapter explains what is
18.1 Product Effect Explained meant by these terms, and also describes why it is important
to consider product effect when choosing the right metal
18.2 Factors that Influence Product Effect detector to use.
18.3 What is a ‘Dry’ Product? To understand product effect, it is important to recall how
a metal detector works, details of which are contained in
18.4 Dealing with Product Effect Chapter 1. Familiarisation with this chapter will make it
18.5 Summary clear why products have product effect and why product
effect matters.
18.1 Product Effect Explained

Metals are not the only materials that can conduct electricity
and generate magnetic fields. Many food products have the
same ability – though to a lesser extent than metals.
For example, salt water (saline) is a relatively good
conductor of electricity, but has very low permeability,
compared to iron. If a saline-rich product is subjected to a
magnetic field, eddy currents will form in the product and
(like metals) will produce a magnetic field (see Figure 18.1).
Rx2
Rx2 Tx
Tx Rx1
Rx1
Apposing
Apposing Eddy
Eddy
Magnetic
Magnetic Currents
Currents
Fields
Fields
Metal
MetalContaminant
Contaminant
Figure 18.1

106
The low permeability of saline solution will tend to weaken 18.2.2 Temperature
the magnetic field, but will not cancel it out. This magnetic
field is relatively small (compared to an equivalent-volume The temperature of the product being inspected has a
of iron), and will have very little effect on the magnetic field major impact on its ability to affect the magnetic field in
in the aperture. the detector.
However, metal detectors are precision instruments which Changing the product temperature alters the product’s
detect very small metal contaminants in much larger conductivity, so it alters the formation of eddy currents and
products. If a saline-rich product were big enough, it would the resultant magnetic field.
create a magnetic field large enough to disturb the magnetic When frozen products are exposed to higher temperatures,
field in the same way as a small metal contaminant. condensation forms on the outside of the product, changing
In real-life applications, the volume of a typical metal its influence on the magnetic field. If the product is allowed
contaminant is many hundreds (or even thousands) of to warm up even further, it will start to thaw.
times smaller than the products that are being inspected – Thawing allows the moisture in the product to conduct eddy
and it is this volume difference that causes the product to currents better than when frozen. Both surface condensation
affect the magnetic field in the aperture, in the same manner and thawing can have a major impact on the ability of a
as a metal contaminant. metal detector to maintain its sensitivity.
When a product can affect a metal detector to the same However, the greatest impact of changing temperature is
extent as a metal contaminant, it has ‘product effect’. not the strength of the magnetic field generated, but the
Many products with high moisture content exhibit a resultant change in the characteristics (phase angle) of the
naturally high product effect – and these are generally voltage induced into the receive coils.
referred to as ‘wet’ products.
Relatively small temperature changes (i.e. 5°C) will
Items such as fresh meat, chicken and fish have high significantly change the product’s signal in the metal detector.
moisture content, and are relatively good conductors.
These products are more likely to produce a signal in 18.2.3 Size and Shape of the Products
the metal detector, in the same way that small metal
Typical packaged products have a uniform shape that
contaminants would. This product effect makes it more
gives a consistent product signal; this is easily interpreted
difficult for the detector to distinguish between the product
by the detector. Other products, such as whole chickens,
and the metal contaminant.
vary significantly in size, shape and weight. In general,
The size of the contaminant, which can be detected in a bigger chicken will give a bigger product signal than a
product effect applications, is, therefore, larger than in ‘dry’ smaller chicken.
applications. The detector sensitivity also varies with the
type of metal contaminant. 18.2.4 The Position and Orientation
of the Product Through the Detector
18.2 Factors that Influence
If the size and shape of a chicken influences the magnetic
Product Effect field, it is easy to understand why its orientation through
Many factors affect the characteristics of a product, and the detector can have a similar effect. If a chicken is passed
these variations are difficult to control on a production line. through the detector with its short edge leading (i.e. head-
To compensate for such variations, metal detector sensitivity first), its signal appears much smaller to the metal detector
is often reduced, so that the majority of products can pass than when it is passed through the detector sideways.
through the detector without false triggering; however, On the production line, it can be difficult to control the
overall detector sensitivity is reduced. orientation of products such as fresh chickens. It is not
uncommon for multiple chickens to pass through the
18.2.1 Moisture Content Variation aperture at the same time. This dramatically increases the
variation in the product signal.
Moisture and salt content can change significantly from
pack to pack; in beef, for example, the moisture and salt The product position on the conveyor is also a variable in
content are dependent on the cut of beef that is being the detectability of metal contaminants. The magnetic field
inspected. Moisture and salt content will also vary from one in the metal detector aperture is weakest in the centre of
animal to the next. the aperture.
If the product is marinated, the amount of marinade (and its Therefore, a product that passes through the centre of the
consistency) is not always constant, and these factors can aperture is exposed to a weaker magnetic field than a
very quickly change the conductivity and permeability of product passed through the detector close to the sides of
the product. the aperture, (where the magnetic field is at its strongest).
So the impact of the product on the magnetic field is
dependent on its position in the aperture.

107
18
Understanding Challenging Applications for Improved Metal Detection
18.2.5 Consistency and 18.3 What is a ‘Dry’ Product?

Density of the Product In the balance coil detector described above, various
Different materials exhibit different levels of permeability products can be passed through the detector without any
and conductivity – and their impact on the magnetic field significant change in the magnetic field. For example, if a
in the aperture will also vary significantly. The ratio of the bag of dry flour (with very low conductivity and permeability)
ingredients (or the bone content in meats), will all have an were passed through the detector, it would not have a
impact on its ability to be inspected. significant impact on the metal detector’s balance state.
This inconsistency in the product being inspected is one of These products are referred as ‘dry’ products. However, the
the biggest challenges when inspecting products such as term ‘dry’ can sometimes be a little confusing, since some
ready meals. A tray with mashed potatoes, sausages and dry products do have significant product effect, whilst some
gravy has a product signal which varies significantly, with ‘wet’ products appear as if they were dry products, when
variations in both the quantity and ratio of the ingredients being inspected.
which causes the size of the signal to vary. Products such as fresh meat have high levels of product
effect, though frozen meat appears more like a dry product.
18.2.6 Packaging Material The main reason for this is that the conductivity of frozen
water drops to almost zero, preventing the formation of eddy
A variety of packaging materials is used in the food currents and their resultant magnetic fields.
industries today. Many of these packaging materials have
very little effect on the metal detector’s sensitivity.
18.4 Dealing with Product Effect
However, packaging materials such as metallised film can
Despite the challenges presented by product effect, a modern
have a significant impact on achievable sensitivity levels.
and well-designed metal detector (using the technology
Metallised film is a material made by coating materials available) should still be able to perform inspection functions
(such as polypropylene or polyethylene terephthalate) with a to a very high standard, assuming that the metal detector is
thin layer of metal – typically, aluminium. specified to suit the application.
The thin layer of aluminium (~0.5um) on the film’s surface
will have a similar magnetic permeability to free air, but will 18.4.1 Low Versus High Frequencies
also have relatively good conductivity levels, depending on Metal detectors can be designed to operate at a variety
the thickness and uniformity of the aluminium layer. of frequencies, from as low as 25 kHz up to 1 MHz. If a
The relatively high conductivity of the metallised film will metal detector operates at 1 MHz, the magnetic field in the
allow the formation of eddy currents in the thin layer of metal detector aperture changes polarity 1,000,000 times
aluminium, generating a magnetic field that will affect the per second.
magnetic field in the metal detector aperture. Eddy currents are formed in an alternating magnetic field, and
Metallised film packaging can make it more difficult to the amount of current generated is depended on the magnetic
detect metal contaminants inside the product. In certain field strength and frequency. The eddy currents generated in a
cases, it is generally recommended that products are product will be much higher at 1 MHz than at 25 kHz.
inspected before they are packaged into metallised film. At low frequencies, such as 25 kHz, the magnetic field
The widely adopted practice of using throat metal detectors generated in the aperture is relatively strong, but the
in the snack food industry is a good example of appropriate corresponding eddy currents and magnetic field generated
inspection techniques when inspecting products in in the product are relatively low. The magnetic field in the
metallised film packaging. product has very little effect on the very strong magnetic field
in the aperture, and so the product signal is relatively low.
If this form of inspection is not possible, it is still possible to
inspect the finished pack and using the latest MSF technology However, the signal from the metal contaminant is also
metal detection extremely high levels of sensitivity are still relatively low – and at these low frequencies, the electric
achievable. In some cases (where the metallized film is thin) and electromagnetic noise becomes a dominant factor in
the level of sensitivity achieved are virtually the same as if the sensitivity that the metal detector can achieve.
inspecting a poly wrapped “dry” product. At very high frequencies, the maximum power that can be
switched into the transmit coil is limited, so the magnetic field
in the aperture is much lower than at the lower frequencies.
However, the amplitude of the eddy currents in the product
and the contaminant are higher at high frequencies,
allowing a larger magnetic field to form around the product.

108
18
The resultant product signal is relatively high, compared to 18.4.3 Multi-Simultaneous-Frequency

the magnetic field in the aperture – and this makes it hard
to detect the metal contaminant. At very high frequencies (1 Metal Detectors
MHz), the metal detector is easily saturated by high product The most sensitive metal detectors on the market today
effect applications, and the sensitivity is compromised. operate at more than one frequency simultaneously, known
In traditional metal detector technology, there is always a as Multi-Simultaneous-Frequency (MSF). These detectors
trade-off between the product effect, maximum operating address the problem of product effect in a new and
frequency and metal detector sensitivity. The old rule was: innovative manner.
the higher the product effect (i.e. the wetter the product), the The new MSF metal detectors use various combinations
lower the optimal frequency and the lower the sensitivity of of high and low frequencies simultaneously. The most
the detector. sophisticated detectors use built-in Product Signal
However with the advent of MSF technology this rule no Suppression technology, with two stages of discrimination:
longer applies, as the metal detector will now operate using frequency and phase.
2 or more frequencies at the same time with at least one of This cancels the information from combinations of high
these frequencies operating in the high frequency mode. and low frequencies. The result is effective removal of
In comparison, the lower the product effect, the higher the the product signal, allowing for much smaller metal
operating frequency and detection sensitivity is. Dry products contaminants to be detected.
with low product effect have very little impact on the metal This technology can also effectively deal with product
detector at high frequencies, so the detector can easily detect variations. Once a detector has been set up to inspect a
very small metal contaminants at these frequencies. particular product, product signal suppression technology is
applied to each product that passes through the detector.
18.4.2 Phase Discrimination On-board detector electronics adjust for variations in product
effect for each product inspected, dramatically increasing its
Operating frequency alone is not sufficient to deal with the
ability to detect small metal contaminants consistently.
effect from a product, so additional techniques are required
to allow detection of very small metal contaminants. It is so effective at adjusting for these variations that it can
even compensate for several of the same products passing
The most common technique used in metal detectors
through the metal detector at the same time.
that operate using 1 discrete frequency is called
‘phase discrimination’. The improvement in detector performance – from the
traditional single-frequency metal detectors to multi-
This technique separates or ‘discriminates’ between signals,
simultaneous-frequency detectors – is as much as 50%
meaning that it can dramatically reduce the product signal,
in product effect or metallised film applications.
while amplifying the metal signal. It does this by ‘learning’
the characteristics or ‘phase’ of the product signal, and In more challenging applications, operating to a factory
placing a ‘discrimination envelope’ around the product signal. detection standard can prove difficult for any one (or all)
of the product effect factors listed in this chapter.
This effectively ignores the product signal inside the
envelope detecting signals outside the envelope. Attempts to meet a factory standard normally involve
increasing sensitivity levels to a point where the metal
Other common names for phase discrimination are
detector set-up becomes unstable and the system rejects
‘phase filter’ and ‘phase control’. In early analogue metal
good product. This is known as a ‘false reject’ or a
detectors, phase discrimination was implemented using a
‘false positive’.
relatively simple technique that provided a basic level of
sensitivity performance. Operating with a high false reject rate (FRR) can be very
costly for a business, to the extent where a trade-off or
As industrial metal detectors evolved and became more
concession to the factory detection standard must be given.
digitally based, phase discrimination became more
sophisticated, and sensitivity performance improved. The advent of MSF and Product Signal Suppression
technology now gives manufacturers greater confidence
The most sophisticated metal detectors on the market
that they can meet (or exceed) a factory detection standard
today have a dedicated Digital Signal Processor (DSP) to
without the worry of generating costly false rejects.
handle phase discrimination and other advanced signal
processing techniques. This allows the detection of even
smaller metal contaminates.

109
18
18.5. Summary A good metal detector supplier should offer a wide range of
metal detectors for every application, so when choosing a
Every product inspected by a metal detector has an metal detector, the supplier should be chosen with equal care.
influence on the ability of the metal detector to find a
metal contaminant in the product. This influence is called A competent metal detector supplier should be able to offer:
its ‘product effect’. Products that have high product effect • A complete range of products operating across a range of
are usually referred to as ‘wet’ products or ‘high product technologies with the best sensitivity
effect applications’. • Good-quality, highly stable metal detectors with excellent
protection from electromagnetic and other interference
The factors that influence the inspection of wet products or • A reliable service support network with worldwide
products with high product effect are: coverage, so that support is available wherever a metal
• The moisture content in the product detector is installed
• The temperature of the product • An understanding of what is needed to develop a
• Size and shape of the product contaminant detection programme that meets the strictest
• Position and orientation of the product through the detector food safety standards
• Consistency or density of the product • A metal detector that is easy to set up and use
• Packaging material • Future-proof metal detector solutions which can grow
• Frequency at which the product is inspected with a manufacturer’s business
Products with very low product effect are often referred to To make it easier for users to pick the right metal detectors for
as ‘dry’ products. These products can be inspected at very their product, suppliers should offer various ranges of products
high frequencies and sensitivity levels. to suit each requirement. Single frequency tuned metal
Products with high product effect are much more detectors should be selected for the inspection of dry products,
challenging, and the processes of choosing a metal detector whilst multi-simultaneous-frequency detectors are appropriate
plus a metal detector supplier are much more complex. for all product effect and metallised film applications.
For such applications, manufacturers should consult with

an expert supplier who has the technology and service
support to eliminate these concerns.
Technology Dry Product Sensitivity Challenging Sensitivity Future Proofing
Single frequency metal detector Average Good Poor

(Low to mid-range tuned frequencies)
Single frequency metal detector Excellent Average Poor

(High-range tuned frequencies)
Fixed 2 of 3 frequency metal detector Average Good Good

(Low to mid-range tuned frequency)
Fixed 2 of 3 frequency metal detector Excellent Good Very Good

(Low to high-range tuned frequency)
Variable frequency metal detector Average Good Excellent

(Untuned – large frequency range)
Dual simultaneous frequency metal detector Good Very Good Excellent

(Discrete operation)
Multi simultaneous frequency metal detector Good Excellent Excellent

(Simultaneous subtraction technique)
Table 18a: Metal detector frequency technology and relative inspection capabilities

110
18
Notes

111
19
Explosion Protection for Metal Detectors

Ensuring Safety for Conveyorised and Gravity Fall Applications
19
Explosion Protection for Metal Detectors 19.1 Better Safe than Sorry
Protection and safety are crucial aspects of industrial
19.1 Better Safe than Sorry manufacturing. They not only protect people’s lives and
prevent injuries, but also avoid financial loss. We often hear
19.2 Explosion-proof Protection about accidents in production companies causing damage
that could easily have been avoided. Time and again,
19.3 Playing it Safe significant injury to persons or damage to property results
19.4 On the Safe Side from a failure to heed safety measures, improper use of
machinery or simply carelessness.
19.5 An All-round Safe Bet
As modern industry has developed, machinery and
19.6 Safety First production plants have been designed and constructed to
be increasingly safe. Rules and regulations have always
19.7 Thought of Everything? Let‘s be on the Safe Side played a part in this and continue to do so, as they indicate
framework conditions for manufacturers and operators
19.8 References which help ensure safety and protection in manufacturing.
Given the complexity of the industrial landscape and its
increasingly international nature, there are a huge number
of transnational regulations dealing with safety, which
particularly affect companies and manufacturers operating
on a global scale. These regulations might have similar
principles underpinning them, but they differ in terms of how
they are executed. In all cases, however, they have to be
observed by every party affected by them.
Flammable substances and potentially explosive
atmospheres in production represent a significant source of
danger in modern manufacturing facilities. Many branches
of industry use flammable substances such as gases,
vapours, mists or dust, which can quickly form an explosive
mixture when they come into contact with oxygen: this is
a problem that affects much more than just the chemical
and petrochemical industries. Even in seemingly non-
hazardous production facilities, such as those used in

112
food production, there is a risk of explosion: flour dust, for However, this does not necessarily mean that an explosion
example, can ignite when making bread. And when you will occur whenever these three components come together.
use toiletries or cosmetic products, do you ever stop to There are other physical properties and requirements that
think that filling aerosol cans demands particular protective have to be taken into account (Figure 19.2). For example,
measures for employees and machinery to eliminate the risk the concentration of the mixture plays an important role. If
of explosions? it is too high (rich) or too low (lean), no combustion or only
stationary combustion will take place. The range between
In production areas at risk of explosion, ‘better safe than
the points at which the mixture is too lean and too rich
sorry’ is a useful maxim to work by. Observing regulations,
defines what is known as the explosive range, in which an
putting them into practice and maintaining an awareness of
explosive reaction will take place. Not only do flammable
the prevailing physical conditions all play an important part
gases and vapours have different upper and lower explosive
in establishing safety and protection in dangerous working
limits, but mixtures of different flammable substances have
areas. Using metal detectors in these hazardous areas
their own limits too. These limits are also affected by the
requires the manufacturer to have in-depth knowledge of
ambient pressure and temperature as well as the proportion
the physical circumstances as well as extensive expertise
of oxygen in the air.
that enables them to comply with the various international
regulations. Meanwhile, operators must be aware of the
location and equipment requirements of these regulations
Air
and conditions, to make sure they are implemented correctly. 100 Vol % 0 Vol %
concentration
This chapter provides an overview of explosion protection in
relation to metal detectors with the aim of ensuring as high
a level of safety as possible in manufacturing environments.
After all, it’s better to be safe than sorry. Mixture too lean Explosive range Mixture too lean
19.2 Explosion-proof Protection Partial combustion,

No combustion
Physical principles and definitions. The established physical no explosion
principles that apply here are based on physical laws which
have, for the most part, been proven and defined through Lower Explosive limit Upper
experiments. An explosion is defined as a chemical reaction
or a physical process in which temperature and pressure Concentration
increase significantly in a short space of time. This leads to 0 Vol % of flammable 100 Vol
a sudden volume expansion of gases, which releases huge substance
amounts of energy in a small space. An explosion always
has the same three ingredients (Figure 19.1): Figure 19.2: Illustration of explosive limits
1. A flammable substance in the required proportion
and concentration When it comes to igniting potentially explosive
2. Sufficient oxygen from the air atmospheres, it is the ignition source that is the decisive
3. An (effective) ignition source factor, and this can take a variety of forms. The ignition
sources most commonly found in industry are:
• Hot surfaces
• Flames and hot gases
Ign
• Mechanically generated sparks

it io
en
Explosion
yg
ns
• Electrically generated sparks

Ox
ou
rce
• Stray electric currents

• Static electricity
• Lightning strike
Fuel • Electromagnetic waves
Figure 19.1: The explosion triangle • Ionising radiation
• Ultrasound
• Adiabatic compression/shock waves
• Exothermic reactions

113
19
19.2.1 Explosion Protection – Step by Step 19.2.4 Stage three:

The explosion triangle provides a starting point for Tertiary Explosion Protection
considering how an explosion could be prevented. It If is not possible to prevent potentially explosive
states that there must be oxygen in the air, a flammable atmospheres from forming, or it is only possible to an
substance and an ignition source in order to trigger an unsatisfactory degree; and it is also impossible to eliminate
explosion. This means we must find a way of removing one ignition, then it is necessary to introduce appropriate
or more of the elements from the fire triangle. For example, measures for reducing the impact of an explosion to a non-
we could remove or replace the flammable substance. hazardous level. These might include:
Alternatively, we could place the equipment in a housing
which provides the flammable substance with a flame proof • Encapsulation in a flame proof or shock pressure-
enclosure. Another option would be to remove the oxygen resistant enclosure, which protects the external
by replacing it with an inert gas, i.e. one which exhibits environment in the event of an explosion. The advantage
very little reactivity, such as nitrogen. There is also a very of this is that standard industrial equipment can be used,
commonly used method which involves reducing the energy and only the housing has to be adapted.
in the equipment to such an extent that it can no longer • Using pressure-relief equipment: The energy of the flame
be a source of ignition. Taking this approach results in a is released in a controlled way through an opening to
systematic explosion protection procedure consisting of prevent ignition of the general area at risk of explosion
three stages which build on one another. outside the housing.
• Using special extinguishing equipment to
suppress explosions.
19.2.2 Stage one:
These three stages of explosion protection measures form an
Primary Explosion Protection integrated system which also specifies the order in which the
Primary explosion protection covers all measures which measures should be applied (see Figure 19.3).
prevent the formation of a potentially explosive atmosphere.
These might include:
• Using substitutes for flammable substances Integrated explosion protection
• Using inert gas
• Using gas detectors
• Preventing the formation of explosive atmospheres in 1.
hazardous areas, e.g. using ventilation Preventing formation of hazardous
potentially explosive atmospheres
19.2.3 Stage two:
Secondary Explosion Protection
2.
If it is not possible to stop potentially explosive atmospheres Preventing ignition of hazardous
forming, or it is only possible to an unsatisfactory degree, potentially explosive atmospheres
this stage involves implementing appropriate measures
designed to prevent ignition of the potentially explosive
atmosphere. These might include:
3.
• Using appropriate materials, such as stainless steel, to Reducing impact of an explosion
prevent sparks from forming to a non-hazardous level
• Preventing static electricity, e.g. by means of proper
earthing and conductive materials Figure 19.3
• Avoiding hot surfaces

114
19
Guidelines Standard
CENE EC (Europe) ATEX
CEC (Canada) CSA
NEC (USA) FM
NEC (USA UL
GOST (Russia) GOST R-Ex
AQSIQ (China) NEPSI
MHLW (Japan) TIIS
IEC (Australia) IEC-Ex
Figure 19.4: Standards and regulations differ around the world
19.3 Playing it Safe 19.3.1 The Jungle of Relationships

Looking at the world from the perspective of explosion Between Standards and Regulations
protection, we can state first of all that every location is, The main driving force behind for standardisation over the
by its very nature, governed by the same laws of physics years has been the international standards organisation IEC.
and chemistry. Secondly, we have developed technical The IEC (International Electrotechnical Commission) is the
procedures and protective measures that are broadly similar leading organisation for international standards governing
across the board, albeit containing some differences to a electrical, electronic and related technologies. Its objective
greater or lesser extent. is to harmonise the many regulations and standards around
Standards and mandatory regulations facilitate the free the globe and to remove trade barriers affecting products
movement of goods by identifying a uniformly recognised related to one another.
framework: This covers everything from product certification To this end, they have developed the
requirements to protective measures for employees who IECEx System.
work with the products. Against this background of the This defines a precise process for testing and
free global movement of goods, recognised authorities certifying new products and appoints authorised bodies to
work to develop uniform standards on both a national and carry out these tasks. Added to this, the certification system
international scale. However, historical and country-specific has been extended to cover maintenance and repairs for
developments have meant that many areas – including explosion-proof equipment. A third branch of the IECEx
explosion protection – do not yet have a global standard in System defines the expertise required by specialists working
place. But this is now set to change. in areas at risk of explosion. The definitive standard is IEC
For manufacturers of equipment used in areas at risk 60079 and its subdivisions: this relates to the requirements
of explosion, this means that they must comply with for areas at risk of explosion. As there is no globally
different regional or country-specific regulations in cases recognised standardisation at present, however, national
where products are to be supplied to the region or country regulations may deviate from these standards. In view of
concerned. Manufacturing companies who operate this, in each case it is necessary to check the extent to
worldwide are confronted with different protective measures which IEC standards can be applied in the individual
and operator obligations in each case, meanwhile, and they countries concerned.
are well advised to know the relevant rules.

115
19
European Contracts 19.3.2 What the EU Directive 99/92/EC

(ATEX 137) contains
Article 95 Article 137
This directive is aimed at operators in a facility that is at
risk of explosion, and specifies their tasks. In accordance
with the directive, the explosion risk of a facility must be
Directive for manufacturer Directive for operator assessed before it is commissioned. To do this, the facility
is divided into classified zones – areas in which potentially
explosive atmospheres may occur. All measures taken to
protect employees must be listed and documented in a
ATEX 95 ATEX 137 special explosion protection document.
EU
Dir.94/9/EG Dir.99/92/EG
19.3.3 What the EU Directive 94/9/EC
(ATEX 95) contains
ExVO D BetrSichVO
This directive establishes basic safety requirements for
equipment and protective systems used in areas at risk of
explosion. For example, equipment is divided into categories
which reflect its safety levels. This makes it clear which
zones the equipment can be used in without being a potential
ignition source. Appropriate protective measures put in place
Figure 19.5: Implementation of European directives into national by the manufacturer, known as types of protection, have the
regulations using Germany as an example
task of ensuring this safety is maintained.
What are the current legal foundations in Europe? In the The directive also regulates the requirements which a
1970’s, the Council of the European Community laid the piece of equipment or system must meet in order to be
foundation for the free movement of explosion-proof electrical eligible for European trade. This means that it stipulates
and non-electrical equipment within the European Union. how a Conformity Assessment Procedure (for equipment
Corresponding European standards developed by CENELEC approval) should be structured, based on the equipment
(the committee responsible for European standardisation category. In some cases, the manufacturer has the option
in the area of electrical engineering) and CEN (responsible of performing and documenting the procedure themselves;
for standardisation in all other technical areas) replaced normally, however, a notified body (such as BSI – British
the national standards which had been valid until then. The Standards Institution – etc. in the U.K. or TÜV, Dekra, PTB
harmonisation of regulations continued; for example, the etc. in Germany) must carry out an EC-type examination
international standardisation organisation IEC adopted the and certification. Additionally, equipment cannot be put on
European standards for electrical equipment developed by the market without operating instructions or without CE and
CENELEC, the European standardisation organisation. Ex marking. CE and Ex marking confirms that the equipment
complies with all relevant EU directives.
The 1990’s saw the introduction of two directives
which achieved full harmonisation and also took all Although the basic principles of explosion protection may be
types of equipment into consideration. Directive the same worldwide, different systems with deviations from
94/9/EC (ATEX 95) regulates requirements for the the IEC’s specifications have also developed.
properties of explosion-proof equipment and
protective systems, while Directive 99/92/EC (ATEX 137)
provides specifications for operation in areas at risk of
explosion as well as measures aimed at ensuring the safety
of those working in such areas. These two directives were
then implemented on a national level in each country.
Directive 94/9/EC (ATEX 95) had to be adopted into national
law without deviation. In Germany, this was achieved by
means of the Explosion Protection Regulation (ExVO).
Directive 99/92/EC (ATEX 137), on the other hand, had the
option of expansion when it was adopted into national law.
This resulted in the introduction of the Ordinance on
Industrial Safety and Health.

116
19
Div 1, Zone 1
Div 2, Zone 2 Div 1, Zone 0
Div 1, Zone 0 or 1
Figure 19.6
Classification of environment at risk of explosion for materials in class I

(gases, vapours and liquids)
Flammable material is Flammable material is Flammable material is only

always present, or is often present during present during normal operation
present for long periods normal operation
NEC zone system,

Zone 0 Zone 1 Zone 2
Article 505
NEC zone system,

Division 1 Division 2
Article 500
Table 19a
19.3.4 USA/Canada/Russia/China Russia. In 2001, a series of new standards was

introduced in Russia as a step towards
USA/Canada. In the USA, all regulations relating to harmonisation with the IEC’s international
manufacturing facilities at risk of explosion are standards. They define not only requirements for
found in the National Electrical Code (NEC) constructing appropriate equipment, but also the
Handbook; in Canada, they are found in the classification of areas at risk of explosion and the levels of
Canadian Electrical Code (CEC). They are protection required in those areas. It is still clear, however,
comparable with the ATEX regulations in Europe, although they that the national regulations deviate in many ways from the
refer to a series of additional standards from other institutions. international standards.
The ATEX directives are not recognised in the USA.
China. All equipment used in areas at risk of
While the regulations in these countries reveal differences in explosion must be tested and approved by a
comparison with the IEC system, they also show that there national testing and certification company (e.g.
have been attempts at harmonisation. For example, areas NEPSI). This is a statutory requirement in the
at risk of explosion within a company were traditionally People’s Republic of China. National standards form the
classified into divisions rather than zones. However, this basis of these certification procedures. A process of
changed in the 90’s when the IEC zone concept was harmonisation with the national standards of the IEC and
introduced in Canada and brought in to supplement the the American NEC has been in progress for some time;
existing system in the USA. This means that the USA now however, it is still far from complete.
has both a zone and a division system.
Conclusion: Many standards that are applied worldwide
In the USA, organisations such as Underwriters Laboratories are based on other standards, but the key fact is that these
Inc. (UL) or Factory Mutual Research Corporation (FM) set standards are similar all over the world. Despite this, there is still
the standards. In Canada, meanwhile, it is the Canadian no uniform global standard. Therefore, products which are sold
Standards Association (CSA) as well as standardisation in different countries also have different certifications for different
organisations and certified bodies, which are comparable environments at risk of explosion. Furthermore, the symbols on
with the PTB in Europe. They have their own testing the respective labels differ from one another. Many countries in
methods for proving conformity with NEC guidelines and Southeast Asia and Latin America have no local standards of
they issue their own approval documents. their own and therefore accept ATEX or FM approval.

117
19
19.4 On the Safe Side The system underpinning this requires the company to carry
out an analysis of the environment and classify it so that
Technical principles the appropriate equipment can be used there. For example,
Manufacturing companies with environments in which a key directive for German companies (and, therefore, EU
potentially explosive atmospheres may occur are obliged to members) is EU Directive 99/92/EC (ATEX 137), which
use equipment and materials which can be safely operated is implemented at a national level in the Ordinance on
in these environments. The relevant directive and standard Industrial Safety and Health. In accordance with this, a
indicate which criteria apply in each case. “competent person” assesses the areas at risk of explosion
and classifies them into defined zones. To allow them to
carry out this classification, this person must have extensive
Zone classification
knowledge of the flammable substances, processes and
equipment involved. The system based on the regulations
Zone An area in which a hazardous, referred to here consists of three different zones.
0 potentially explosive atmosphere in
There are also the manufacturers who provide
the form of a mixture of air and
equipment and materials for the three defined zones with
flammable gases, vapours or mists
corresponding protective devices. The specifications
is always or frequently present,
for Europe are outlined in Directive 94/9/EC (ATEX 95),
or present for long periods.
which the relevant countries were required to adopt into
national law without deviation. In Germany, for example,
Zone An area in which a hazardous,
this was carried out in the form of the Explosion Protection
1 potentially explosive atmosphere in the
Gas Regulation. It defines the applicable equipment category,
form of a mixture of air and flammable
(G) which specifies exactly which piece of equipment may be
gases, vapours or mists can sometimes
used in which defined zone. This information is guaranteed
occur during normal operation.
by prescribed certification processes and conformity
assessments, which must be documented accordingly.
Zone An area in which a hazardous,
2 potentially explosive atmosphere in the
form of a mixture of air and flammable 19.4.1 What are the Technical Principles
gases, vapours or mists does not That Apply Today
normally occur, or only occurs briefly,
during normal operation. From a historical perspective, explosion prevention was
first required in the field of mining (firedamp), but it was
Zone An area in which a hazardous, growing industrialisation that increased the need to regulate
20 potentially explosive atmosphere in explosion protection. As well as the chemical industry,
the form of a cloud of flammable production areas with a high potential for explosion were
dust contained in the air is always increasingly added to the list alongside traditional areas of
or frequently present, or present for industry such as mills. These included the textile industry,
long periods. the food industry and the woodworking industry.
It is likely that this historical development is the reason for
Zone An area in which a hazardous, today’s two-tiered classification system. Both EU Directive
21 potentially explosive atmosphere in 99/92/EC (ATEX 137) and the IEC directives distinguish
Dust
the form of a cloud of flammable dust between two main groups of equipment:
(D)
contained in the air can sometimes
Group I: Equipment used in underground mining
occur during normal operation.
Group II: Equipment used in surface industries
Zone An area in which a hazardous, (other industries)
22 potentially explosive atmosphere in
the form of a cloud of flammable dust Note: As metal detectors are not used in underground
contained in the air does not normally mining, this chapter does not provide an explanation of
occur, or only occurs briefly, during protective measures in this area; it focuses exclusively
normal operation. on group II.
Table 19b: An overview of zone classifications with definitions
The directives and their implementation make another basic
according to the Ordinance on Industrial Safety and Health.
distinction by differentiating between the various flammable
materials in their classifications. They generally differentiate
between the following:
• Gas (G)
• Dust (D)

118
19
Definition of equipment categories and Equipment Protection Levels (EPL)
Classification of equipment for use in surface areas at risk Classification of equipment for use in surface areas at risk
of explosion within Europe into three equipment categories of explosion into three protection levels for areas at risk of
for areas at risk of explosion due to gas (G) or for areas with explosion due to gas (G) or for areas with flammable dust (D)
flammable dust (D) in accordance with EU Directive 94/9/EC in accordance with international directive IEC 60079-0
(ATEX 95)
Category 1: 1G or 1D EPL Ga or Da
Very high level of safety. Safe even when rare equipment faults Equipment with "very high" protection level for use in areas
occur. Two independent explosion protection measures; even at risk of explosion where there is no ignition risk during
safe when two faults occur independently of one another. normal operation, or in the case of predictable or rare
faults/malfunctions.
Category 2: 2G or 2D EPL Gb or Db
High level of safety. Safe even in the case of equipment faults Equipment with "high" protection level for use in areas at risk
which occur frequently or which are usually to be expected. of explosion where there is no ignition risk during normal
Even safe when a fault occurs. operation, or in the case of predictable faults/malfunctions.
Category 3: 3G or 3D EPL Gc or Dc
Normal level of safety. Safe during normal operation. Equipment with "extended" protection level for use in areas
at risk of explosion where there is no ignition risk during
normal operation, and which has some additional protective
measures which ensure that there is no ignition risk in the
case of predictable equipment faults.
Table 19c
Terms
Zone 0 / 20 EPL / Ga / Da
Zone 1 / 21 EPL / Gb / Db
Zone 2 / 22 EPL / Gc / Dc
Category 3
Category 2
Category 1
Ex atmosphere > 50% over an 8-hour working day
Ex atmosphere between 1% and 50% over an 8-hour working day
Ex atmosphere < 1% over an 8-hour working day

Figure 19.7: Relationships between zones, categories and EPL

119
19
19.4.2 Zone Classification 19.4.3 More Information About Gas and

As previously explained, areas at risk of explosion are Dust Explosion Groups
classified into zones which reflect the probability of potentially Up to this point, explosion-proof equipment has been divided
explosive atmospheres occurring. The operator can then into equipment group I (= underground, for mine workings
use this classification to select the correct equipment and at risk of firedamp) and equipment group II (= surface
materials. This system is defined in both the IEC set of equipment, for areas at risk of explosion excluding mine
standards 60079 and in European Directive 99/92/EC (ATEX workings). Equipment group II, which is the focus of this
137), as well as in the national implementations derived white paper, draws a further distinction by differentiating
from this. In Germany, the national implementation is the between areas at risk due to gases, vapours and mists,
Ordinance on Industrial Safety and Health. and those at risk due to dust. The information in Figure 19.8
A competent person or body must carry out the zone shows the relationships between the gas/dust explosion
classification process and related explosion risk group and the protection type required in each case.
assessment, and determine what protective measures are
necessary. Equipment to be used must correspond to the 19.4.4 Ignition Temperature and
assigned equipment categories and Equipment Protection
Levels (EPL) as prescribed by the zone classification. Temperature Classes
These equipment requirements are categorised in relation “The lowest temperature of a heated surface at which a
to the European equipment categories on the basis of EU flammable substance in the form of a gas/air or vapour/
Directive 94/9/EC (ATEX 95); the Equipment Protection Level air mixture ignites under defined conditions is the ignition
(EPL), introduced by the IEC, applies on an international temperature.” Temperature classes from T1 to T6 have been
level. Both classifications can be used when marking defined for flammable gases and vapours as a means of
equipment. This classification tells us about the probability ensuring safety and protection. In practice, this means
of ignition, taking into account potentially explosive gas and that the maximum surface temperature of a material must
dust atmospheres. always be lower than the ignition temperature of the gas/air
or vapour/air mixture.
Gas explosion groups Dust explosion groups
IIC – e asily ignitable IIC – conductive dusts

(e.g. hydrogen, acetylene) easily ignitable (≤10³ Ωm)
IIB – ignitable IIB – non-conductive dusts

(e.g. coalgas, ethylene, ethylene glycol) (>10³ Ωm)
IIA – difficult to ignite IIA – flammable fibres

(e.g. acetone, benzene, toluene) (>500µm)
Figure 19.8
Temperature Classes
Ignition temperature of gases and Temperature class Maximum surface temperature of

vapours in °C equipment in °C
>450 T1 450
>300 to 450 T2 300
>200 to 300 T3 200
>135 to 200 T4 135
>100 to 135 T5 100
>85 to 100 T6 85
Figure 19.9: Example classifications of gases and vapours into

explosion groups and temperature classes

120
19
Explosion Temperature Classes

Group
T1 T2 T3 T4 T5 T6
I Not relevant for metal detectors
II A Acetone Ethanol Benzine Acetaldehyde

Ethane i-amyl acetate Diesel fuel Ethyl ether
Ethyl acetate n-butane Aviation fuel
Ammonia n-butyl alcohol Domestic fuel
Benzene (pure) oils
Acetic acid n-hexane
Carbon
monoxide
Carbon dioxide
Methane
Methanol
Propane
Toluene
II B Town gas Ethylene Carbon

(illuminating disulphide
gas, coal gas)
II C Hydrogen Acetylene Hydrogen

disulphide
Table 19d
19.4.5 Types of Protection in the Field Type of protection/ IEC, EN Standard

of Metal Detection Designation
To prevent the ignition of a potentially explosive atmosphere,
one protective measure is to eliminate the ignition source Protection by enclosure "t" IEC 60079-31, EN 60079-
as a possible explosion trigger. This measure forms part 31, IEC 61241-1, EN
of secondary explosion protection (see Page 5) and can 61241-1
be implemented in a large number of ways. Types of
Table 19e: Types of protection for electrical equipment in areas at risk
protection are defined in various international and European of explosion due to dust
standards would be IEC/EN 61241 (dust atmospheres).
The construction and building requirements for the different
types of protection are described in the relevant standards,
and the manufacturer must comply with these during
construction, manufacturing and testing. Each manufacturer
must carry out an ignition risk assessment in accordance
with the latest standards: this is obligatory regardless of the
zone and category that apply.
The types of protection are divided into different levels for
dust atmospheres and also for electrical and mechanical
operating equipment. Different protective measures can be
used to prevent an explosion, and this is reflected in the
different types of protection. Specific letter combinations
denote the various types and levels of protection. As a rule,
however, all the types of protection are equally safe.

121
19
19.4.6 ‘Ex’ Marking 19.6 Safety First

Certain data must be affixed on every piece of equipment Equipment construction and operation in areas
and protective system as a minimum requirement (see also at risk of explosion.
2014/34/EU); this must be clear and indelible:
Everyone involved must work together to ensure safety
• Name and address of manufacturer in areas at risk of explosion. As well as the operator and
• CE marking manufacturer of the equipment, the constructor, relevant
• Series and type designation authority and (where applicable) the notified body also have
• Fabrication number or serial number if applicable certain obligations.
• Year of manufacture
• Special explosion prevention marking along with the Operator: Responsible for the safety of their equipment and
equipment group and category marking for equipment must, therefore, comply with the relevant national laws and
group II, the letter “G” for areas containing potentially standards. Their primary duty is to assess the explosion risk
explosive gas, vapour, mist or air mixtures and/or the in the facility and specify the zones. These form the basis
letter “D” for areas in which dust can form potentially for the protective measures to be taken and for selecting
explosive atmospheres. appropriate equipment. The operator must also ensure that
• If necessary, all information required for safe use must employees comply with the protective measures and, if
also be affixed on the equipment. necessary, must provide appropriate training. Additionally, it
is the operator’s responsibility to ensure that the equipment is
Important: Markings on explosion-proof equipment must correctly installed and checked before being commissioned.
ensure that it can be used correctly according to its safety- Regular maintenance and testing must be carried out to
related properties. ensure safe operation of the equipment. If the equipment
is moved, its new location must be checked to ensure it is
19.5 An All-round Safe Bet appropriate for the equipment dimensions.
Special requirements for metal detectors Constructor: Those responsible for constructing or installing
equipment in a potentially explosive atmosphere must
Manufacturers of explosion-proof metal detectors for gravity be diligent in complying with the relevant requirements.
fall applications have to meet a whole range of requirements Information concerning the subsequent operator as well as
when it comes to implementing the appropriate protective the legal requirements provide a framework for ensuring
measures for the risk classification in question. This this compliance. In the field of metal detectors, it is very
involves not only complying with the legal directives, but often the manufacturer who installs the equipment on
also considering how their manufacturing facilities are site; however, this can also be carried out by an external
organised for the production environment concerned. company or by the operator personally.
Depending on the customer’s requirements, it may be that Manufacturer: It is the manufacturer’s responsibility to
the metal detector is to be used entirely in an environment at ensure that a piece of equipment is eligible for trade as
risk of explosion. Equally, however, it may be that only parts intended. The manufacturer must therefore comply with the
of it are affected. For example, the HMI (Human Machine laws and directives of the country in question and carry
Interface) may have to remain outside the zone while the out the appropriate testing and assessment procedures.
remaining metal detector components are within it. The manufacturer must also provide the relevant equipment
Expert manufacturers are ideally placed to provide the markings and documentation. In Europe, for example,
best solution in each case for meeting the requirements equipment must have the CE and Ex markings as well as
of the particular circumstances, from both a technical and operating instructions. An appropriate quality assurance
financial perspective. The large number of key components system must be in place to ensure that every piece of
in a metal detector means that tried-and-tested protective equipment is produced using tested construction methods.
measures, which can stand up to any challenge, are a Notified body: Notified bodies are neutral and independent
must. In turn, this makes an integrated safety concept for organisations whose main task is to carry out conformity
explosion protection essential for manufacturers. This is the assessments on products intended for free trade movement,
only way of protecting all modules so that they can face the if this is intended for the product in question according to
demands of practical scenarios and regulations. the relevant EU directives. In Germany, these organisations
include the TÜV, Dekra and PTB.

122
19
19.6.1 A Word on Product Approval 19.7 Thought of Everything?

in ‘Ex’ Areas Let‘s be on the Safe Side
Depending on requirements, various approvals may Explosion protection checklist for a metal detector
be required before a piece of equipment can be used in (using example of 2014/34/EU)
an Ex area:
A manufacturer can only decide to use a certain metal
• Certification for manufacturer’s production facilities.
detector in an area at risk of explosion after a thorough
• An EC-type examination certificate for the product within
examination of the requirements imposed by your
EU member states, ATEX approval. For global distribution:
production circumstances. This is not always an easy task,
IECEx approval. This proves that the equipment meets
because some factors are either assumed to be obvious
all safety-related requirements contained in the relevant
or are not known at all. As the customer, the best way
legal regulations.
to help your supplier is to determine the key parameters
• Declaration of conformity. With this, the manufacturer
yourself in advance of obtaining a quote or discussing
declares that the product has been constructed and put
your requirements. Taking this approach should form the
on the market in accordance with the specified legal
groundwork for successful planning. We have summarised
requirements. (Cat. 3G/D)
• Possible additional approvals: Although ATEX approval the most important points in a checklist.
is sufficient for a large number of countries outside
the EU, there are some exceptions. In countries such
as Russia or the USA, for example, separate national
approval is required.
Atmosphere classification (see page 116) Designation
Dust atmosphere D Yes No
Zone classification (see page 117)
Gas Dust Probability of occurrence of Designation

potentially explosive atmosphere
0 20 Constantly, for long periods, Category 1 Yes No

frequently, predominantly
1 21 Sometimes during Category 2 Yes No

normal operation
2 22 Rarely, briefly Category 3 Yes No
(Please attach the safety data sheet if applicable)
Values for your production circumstances
Relative air humidity [ % ] %
Ambient temperature [ °C ] °C
Marking sought in accordance with 2014/34/EU:
Table 19f

123
19
19.8 Reference Sites

ATEX manufacturer’s directive 2014/34/EU
http://ec.europa.eu/enterprise/sectors/mechanical/
documents/ legislation/atex/
CE marking
http://ec.europa.eu/growth/single-market/ce-marking/
International explosion protection
http://www.iecex.com
IEC standards
http://www.iec.ch
Standards
http://www.beuth.de
DIN standards
http://www.global.ihs.com
Safety data sheets for gases and dust
http://www.dguv.de/ifa/de/gestis/stoffdb/index.jsp
ATEX and Explosive Atmospheres
http://www.hse.gov.uk/fireandexplosion/atex.htm

124
19
Notes

125
Notes

126
Product Inspection Solutions
www.mt.com/metaldetection
For more information
Mettler-Toledo GmbH
8606 Greifensee
Switzerland
E-mail: [email protected]
For your local office visit: www.mt.com/contact
Subject to technical changes.

© 06/2016 Mettler-Toledo GmbH.
PI-MD-Guide-EN-GEN

Metal Detection Guide en

Uploaded by

Copyright:

Available Formats

Metal Detection Guide en

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Metal Detection Guide en

Uploaded by

Copyright:

Available Formats

The Metal Detection Guide

Building an Effective Programme

Introduction to Metal Detection 6

Key Design Features 12

Factors Limiting Sensitivity 18

Systems Design & Applications 26

Metal Detection, X-ray Inspection or Both Technologies? 36

Reasons For a Metal Detection Programme 42

Building an Effective Programme 46

Prevention of Metal Contamination 48

Selecting Control Points 52

Installation and Commissioning 60

Performance Validation, Verification and Monitoring 64

Dealing With Suspect and Rejected Product 76

Data Analysis and Programme Improvement 80

Data, Connectivity and Improving Performance 84

Calculating Total Cost of Ownership for In-Line Metal Detection Equipment 90

Understanding Challenging Applications for Improved Metal Detection 106

Explosion Protection for Metal Detectors 112

© 2016 Mettler-Toledo Safeline Ltd

However, merely installing metal detectors as stand-alone

© 2016 Mettler-Toledo Safeline Ltd

Customer/Industry Consortia Industry Standards Regulatory Authorities

Effective Metal Detection Programme

Metal Detection System Selection

Building an Effective Programme

Producer, Retailer and Consumer Benefits

Reasonable Precautions Reduced Failure Costs Protection Against Product Maintain

Due Diligence Retailer Brand Maintain Approved

Table 1a: Effective metal detection programme

© 2016 Mettler-Toledo Safeline Ltd

This guide provides a comprehensive reference source for

© 2016 Mettler-Toledo Safeline Ltd

© 2016 Mettler-Toledo Safeline Ltd

Introduction to Metal Detection

© 2016 Mettler-Toledo Safeline Ltd

© 2016 Mettler-Toledo Safeline Ltd

Non-ferrous metals, such as brass, copper, aluminium and

The control electronics actually split the received signal into

© 2016 Mettler-Toledo Safeline Ltd

1.4.3 Electronic Techniques MFZ

Mechanical construction methods will minimise false

• Metal close to the aperture Figure 1.3

1.4.4 Metal Free Zone (MFZ)

the detector should be surrounded by an area known as the

• The operating sensitivity

© 2016 Mettler-Toledo Safeline Ltd

There are, however, two different methods of interpreting Large particle

© 2016 Mettler-Toledo Safeline Ltd

1.6.4 Inverse Detection 1.7.3 Productivity Improvements

© 2016 Mettler-Toledo Safeline Ltd

Key Design Features

© 2016 Mettler-Toledo Safeline Ltd

© 2016 Mettler-Toledo Safeline Ltd

© 2016 Mettler-Toledo Safeline Ltd

2.3 Conveyor System Design 2.4 Non-Conveyorised

© 2016 Mettler-Toledo Safeline Ltd

2.7 Health and Safety

2.8 Fail Safe System Design

• A reject mechanism which can isolate a plug of product