Error Proofed Production: Five Steps To Achieve Zero Fault Fastening

PocketGuideError 9833 1437 4-02-18 15.
54 Sida 1
Error proofed production

Five steps to achieve zero fault fastening
PocketGuideError 9833 1437 4-02-18 15.54 Sida 2
The later an assembly defect is identified, the more it costs to

correct. And the end result could be even higher costs and the
time-consuming problem of recalls. This has made error
proofing a critical factor in relation to all complex joint
fastening.
Atlas Copco has defined five tightening process control steps

towards zero fault fastening. Which of these steps is suffici-
ent in any given situation is a matter of individual decision.
With this five-stage classification we have tried to provide an

aid for orientation. Of course, there are variations for every
step. But making the move from one step to the next is
distinctive in each case. The steps are thus designed to help
you make decisions on what error proofing level is appropri-
ate for you, and what safety requirements you have to meet.

Five steps to achieve zero fault fastening
Contents
Chapter...........................................................................................................Page
1. Five steps to zero-fault fastening................................................................ 4
1.1 Functionality ......................................................................................... 14
2. Tool management and process auditing .................................................. 16
3. The new way of tightening........................................................................ 20

3.1 Zero fault production/error proofing..................................................... 21
3.2 Continuous process improvement......................................................... 23
3.3 Documentation for traceability...............................................................23
3.4 Efficient maintenance and monitoring ...................................................24
4. Machine capability of tightening tools:

how accurate is exact torque .................................................................... 26
5. The three tightening tool condition classes in

accordance with VDI guideline 2862 ....................................................... 30
6. Error proofing cases .................................................................................. 31

6.1 Preventative maintenance system at Onan Cummins
Power Generation (Tool management) .............................................. 32
6.2 Ford Camaçari: Fast and accurate tool calibration
(Process auditing)............................................................................... 33
6.3 Valve fitting at Hoerbiger (step 2) ..................................................... 34
6.4 Cutting costs and improving quality at
Johnson Controls (step 3) .................................................................. 36
6.5 Wheel fitting for the Dutch Air Force (step 3) .................................. 37
6.6 Tightening tool approves seal tightness at
the same time (step 3) ........................................................................ 39
6.7 The new way of tightening – beer barrels (step 4) ............................ 40
6.8 Ford uses the new way of tightening on its engine lines (step 5) ..... 43
6.9 KISSQ: Zero defect assembly in the BMW Group (step 5).............. 46
ERROR PROOFED PRODUCTION 3

1000% 100% 10%
1. Five steps to zero-fault

production
Today, there is scarcely any discussion on technical issues
relating to tightening that does not also include process quali-
ty. However, everyone understands process control and pro-
cess quality slightly differently. This is why this division into
steps helps provide some long overdue clarity.
When Henry Ford realized assembly line production, process

quality was not taken into consideration. At that time increas-
ing productivity was the main concern, thereby reducing
production costs. Today, however, we want to enhance
quality besides increasing productivity.
It is no accident that the automotive industry (see chapter 5:

VDI 2862) has only three bolt classes: for safety-critical,
function-critical and customer-critical joints. Each of these is
critical and therefore, some car manufacturers still differenti-
ate between tightened joints that are documented (due to pro-
duct liability, among other things) and other joints that are
not.
Process quality also reduces costs. Generally, it is true that

any defect becomes ten times more expensive to rectify for
every stage of production that passes before it is discovered.
Not to mention the badwill that occurs if an assembly fault
is discovered only once the product has made it out to the
customer.
4 ERROR PROOFED PRODUCTION

Step 1. To assure a correct tightening torque

The first step to zero fault production is obtained by using
an assembly tool that delivers a precise and predetermined
torque. However, only the tightening torque is controlled
at this first step, operators and work pieces are not yet
involved in the monitoring process.
Good examples of tools to be used when taking the first

step are screwdrivers and nutrunners with clutches that
shut off the tool when a pre-set torque is reached.
Another example is the shut-off type of hydraulic impulse
tools.
At step 1, process quality is

still dependent on people and
is based only on the "built-in"
torque accuracy of the tighten-
ing tool.

At step 2 the operator is also

"supervised" by the RE control-
ler. No bolts are "forgotten"
because the entire tightening
cycle is monitored. However, at
this stage, the only thing that is
certain is that the tightening
tool has switched off properly.
What happened in the tight-
ened joint itself is still not
known at this step.
Step 2. To assure that all screws are tightened

One of the most common causes of a faulty assembly is the
fact that the operator simply forgets to tighten a screw or
makes a tightening, a so-called re-hit, on an already tightened
screw. The remedy against this possible error is to use what
is known as an RE-controller. It monitors the tightening
cycle and identifies a proper shut-off of a tool. RE-control-
lers are mainly used for pneumatic tools where a combina-
tion of the air pressure at the tool and time are monitored.
This can be done due to the fact that a pressure pulse occurs
Assembly station
Assembly Repair
Signal light
and alarm
Entry Clock
Alert signal
signal
OK signal
RE-tool
RE-Controller

in the tool at the moment of shut-off when the tool comes to

a stop. But monitoring time is also necessary. A re-hit, for
example, would also cause a pressure pulse but this pulse
then comes immediately after the operator has pressed the
trigger. During a proper tightening a certain time elapses
between trigger pressing and shut-off.
The RE-controller is also used for counting the number of

correctly tightened screws. The RE-controller even provides
a warning if the air pressure in the supply network falls
below the level at which the tightening accuracy is guaran-
teed.
At step 3, the tightening tool is

no longer just an assembly
tool, it is also a testing instru-
ment. If the thread or hole is
defective, it activates a signal
and tells the operator if a
washer or seal is missing. At
this step, the OK or NOT OK
signal is used to indicate
whether the tightened joint is
in the green range.
Step 3. To assure that the joint is correct

With steps 1 and 2 on the path to zero fault assembly the tool
and the operator have been taken into consideration.
However, the joint itself can also be a cause of the incorrect
tightening. There can be several reasons for this. Missing
parts like seals or washers will change the characteristics of
the joint. The bolt quality can be wrong. A too low quality
bolt may lead to deformation of the screw into its plastic
range. This will also change the characteristics of the joint.
Damaged threads or debris in the joint also leads to an
improperly tightened joint. A tool would reach its shut-off
torque level long before the joint is properly tightened.
The way to detect these types of faulty joints is to monitor

the tightening angle during the tightening process. This is the
third step to zero fault assembly. Missing washers, faulty
threads, debris in the joint, too short or wrong quality bolts
mean that the tightening angle will be outside its tolerance
range.

Clamping
angle
acceptance
It is necessary to use assembly tools with angle sensing func-
window
tions to reach step three. The tool does not have to have a
Torque
Angle step 1
acceptance window
proper torque transducer built in. A good example is the
Final
Tensor DS/DL tool. Its controller calculates the torque with
an algorithm. This means that the tool cannot be used for
safety critical joints where reporting of a traceably calibrated
Angle
torque value is required. But the tool has an angle sensing
function. Its controller can be programmed to give a toler-
ance width within which the tightening angle has to lie in
Speed order to obtain an OK signal. Faulty joints would lead to a
NOT OK signal.
Angle monitoring: The tighten-
ing angle is supervised in the
first and the second stage of
The Tensor DS/DL controller also has a screw counting
the tightening to detect incor- function taking care of the requirements of step two
rect joints. described earlier.

At step 4, all tightening data is

also documented and/or for-
warded for the final documen-
tation and can be called up at
the workstation at any time in
order to run error analyses, etc.
Step 4. To assure that safety critical joints are

tightened properly
A safety critical tightening is a tightening that, if it was not
correctly performed, can result in a risk for the user of the
end product. In the case of tightenings on a vehicle the drive-
r’s and passenger’s personal safety being at risk. In order to
limit or avoid recalls it is essential for manufacturers to
prove that critical fasteners are tightened correctly.
This proof requires advanced controlled industrial tools such

as the Tensor S/ST and QMX-series and tightening control
systems such as the Power Focus 3000/3100 and
PowerMACS. In combination these systems provide:
1. Torque control using a transducer that is traceably

calibrated.
2. The ability to store the torque results in the controller
unit itself on a local level.
3. The ability to send the torque results to higher level
systems, such as ToolsNet, for long term storage.
4. Continuous monitoring of the whole tightening process
by using not only torque transducers and angle encoders
but also other parameters such as current limits.
5. The possibility to use Statistical Process Control to
detect and correct deviations to the result even before
a tightening outside of the limits is produced.
Fastening seatbelts is a typical

safety critical application.

If there are indications that a recall is related to the fastening

process, it is easy to search the database, view the tightening
data/trace and pinpoint vehicles with the suspected problem.
It is vital that the spindles are calibrated before delivery and

they must be recalibrated at regular intervals. The calibration
results have to be stored for traceability. This is a service that
can be provided by the Atlas Copco service organisation.
Power Focus 3000/3100 and PowerMACS controllers are

designed to communicate via Ethernet, serial or Fieldbus
technology. If Ethernet is used it is possible to set up a cell
for a intra-controller communication with up to 20 control-
lers in a very cost efficient and user-friendly way.
An assembly cell consists of one or several components with

several joints that can be tightened by one tool using the pro-
gramming alternatives in the controllers. To help the operator
tighten the bolts in the correct order and to not forget any
The Power Focus controllers bolts the Power Focus units can be programmed so that an
can link up to 20 mutually all-OK signal is sent when all joints in the cell have been
synchronized tightening tools
to one another in one assem-
properly tightened.
bly cell. One controller then
acts as a master and issues a
signal to the line when all
tightening cycles have been
processed correctly.
Zero-fault production does not Sixteen con-rod bolts in an 8-cylinder Ford engine are tightened in four
mean that no mistakes are steps in this station by this fully automatic multiple with the torque/angle
made during assembly. controlled target values as 40 Nm and 90°.
However, it does mean that no Assembly tools of the PowerMACS generation stand out due to their
faults are allowed to leave the space-saving stand-alone control modules. These controllers have their
plant. And in the long run this own assigned Ethernet IP addresses, which enables full synchronization
is achievable only at step 5 . and real-time communication options for guaranteeing the process.
10 E R R O R P R O O F E D P R O D U C T I O N
At step 5, the tightening tool is

not only an assembly device
but also an information medi-
um and integral component of
the business system: it carries
out tests and informs the cen-
tral management system as to
whether the correct component
is being fitted. Furthermore it
guides the operator where the
bolt has to be tightened, auto-
matically selects pre-program-
med tightening parameters and
determines which socket is to
be used, etc.
Step 5. To assure zero fault fastening

Having reached step four in the advance to zero fault produc-
tion still leaves room for mistakes. With step five two further
elements are introduced for a fault-free production. One ele-
ment is the introduction of part identification, the other is
reject management. With step 5 the tool controllers are not
only networked – they are also connected to the factory net-
work. Information about components is sent over the factory
network. By identifying the components that are to be
assembled, relevant information is transferred to the tool
controller via the network. This safeguards both that the
correct component is being assembled and that corresponding
tightening parameters are chosen.

F
T
W
In
E
Volkswagen's glass plant in
Dresden, Germany, is laid out
for 800 staff and 150 vehicles
a day. Almost half of the 55
000 m2 production area over
three levels is floored using
wood from Canadian syca-
more and smoked oak.
Beneath this flooring are the
electric cables, which inducti-
vely power all electric tighte-
ning tools (exclusively BTV
battery-operated tools and
Tensor tools from Atlas
Copco Tools and Assembly
Systems).
Phaeton production is almost

purely manual work: automa-
tion stands at 5%. There are S
only three robots at the plant: Id
for fitting the spare wheel,
tightening of the wheels
(with the aid of the automatic
PowerMACS system) and for
inserting and bonding in the
front and rear windscreens.
The glass plant in Dresden was the birthplace of the new way of
tightening, with the first standard network solution for bolt assembly in
accordance with step 5.
The tool carriers with battery-tools and Tensor S tools are powered
inductively via cables under the wooden flooring.
Despite all the functionality described above, cross threads,

operator mistakes etc. can still occur. When a NOT OK sig-
nal is sent it is not always the case that the tightening error
can be corrected at the actual assembly station.
This can be remedied by so called reject management. Via
the factory network information about the faulty joint is sent
to the re-work part of the assembly line. When the compo-
nent (motor, gearbox, whole car etc.) approaches the re-work
station of the assembly line it is automatically diverted to it.
There the faulty joint is identified, corresponding tightening
parameters are set and the results are stored so that a release
document that shows that all bolts have been fastened accor-
ding to specification and the correct components have been
used, can be made.
Factory Overview
ToolsTalk Production
WinTC ToolsNet Control
Internet Server DB System and
Explorer Step 6 Quality
System
Step 4a Step 4b
Tightening data OK/NOK Step 5
OK/NOK
TCP/IP
Ethernet
Step 3
Tightening
System
PLC
Master Line Control
Step 1
Identification Slave
Step 3
Tightening
system
NOK
Re-work

1.1 Functionality Step 1 Step 2
Air-powered mechanical As for step 1, but with

screwdriver or impulse pressure-controlled and
type screwdriver with time-controlled shut-off
automatic shut-off
Torque control Torque setting via Torque setting via

shut-off mechanism shut-off mechanism
Torque monitoring/control
Transducer redundancy
Angle monitoring/control
Joint monitoring Monitoring of shut-off over time
Tightening steps 1 1
Yield point control
Process time monitoring ●
Multiple parameter sets
Batch count ●
Station monitoring
Tool alarms
Tightening data collection
Reject management
Operator feedback ●
Operator guidence
Statistical alarms
Statistical reports
Redundant tightening data collection
Service indicator f (time, cycle)
Service indicator f (usage)
Network capability/Internet
A: Safety-critical *)
B: Function-critical *)
C: Customer-critical *) ● ●
Investment level 100 % 200%
* Joint classes in accordance with VDI guideline 2862 for the automotive industry.
Step 3 Step 4 Step 4 Step 5
Tensor DS/DL electric Tensor S/ST elec- QMX spindles with

screwdriver tric screwdriver multi-channel Zero fault fasten-
with Power Focus PowerMACS ing, network con-
3000/31000 tightening system nected
”DigiTork”, algorithm Direct torque Direct torque
This highest
consisting of motor measurement via measurement via
error proofing
current, frequency, voltage transducer transducer
step uses the
and temperature
technology of
● ● ● step 4. It inte-
Current monitoring Multiple transducers grates the
● ● tightening
In 36° steps ● ● process into the
2 8 Unlimited network for
● production
● ● ● management
10 up to 250 Unlimited system and
● ● ● quality
● ● assurance
● ● ● system.
● ●
In station In station
● ● ●
● ●
● ●
● ●
● ●
● ●
Tensor ST
● ●
● ●
● ● ●
● ● ●
250–400% 600–1000% 600–1000 % per channel

2. Tool management and

process auditing
How often tools need to be torque audited is depending on a

lot of factors, for example the frequency of use and the type
of joint. Some companies does torque audits once a year,
others do it hourly. The frequency is reliant on the company
policy and the application. What is important is that the com-
pany has a routine to check their tools so they have control
over their tightening process.
Without process auditing and tool management, let alone a

correctly set and/or calibrated tool, no process protection can
be achieved. This emerges at ISO audits and must be fulfilled
by any company which holds ISO 9000 accreditation.
The demands can be found in quality standard ISO 9000,

paragraph 4.11.2. This contains the following:
• There must be a QA system that reliably ensures that all

tightening tools keep to the prescribed torques.
• There must be a calibration system for the tightening tools
and transducers in order to monitor the torque.
• There must be a system, which ensures that the tightening
processes remain within monitored limits.
• There must be a process auditing and tool management
system, which monitors the status of the tightening tool.
To meet the ISO 9000 requirements, each assembly operation

has to be equipped with the following:
• Torque transducers (in order to measure the torque at the

tightening tool).
• Torque measuring instruments with the option of statistical
evaluation (in order to analyze the information acquired
using the transducers).
• Test joints (to permit the simulation of tightened joints).
• A process auditing and tool management system (to permit
the significant details of tools in operation to be followed).
The Notebook-sized ACTA 3000 However, as a rule process auditing and tool management
forms the backbone of the QA takes place manually, even today in the IT age, with test cert-
management system for
assembly tools.
ificates completed by hand. Given that assembly systems are
today networked, with details for each tightening tool – such
as machine capability (see chapter 4) this process should no ACTA torque testing for
measuring the applied torque.
longer involve the use of pencils and notepads.
Moreover, manually entering the measurement and tool data

gathered into databases is not only time-consuming, but
above all it is prone to errors. Furthermore, this kind of QA
management of all tightening tools in an assembly operation
accredited to an ISO standard is barely affordable. Computer
solutions for process auditing, tool management and for guar-
anteeing calibration intervals are therefore necessary.
Inline measurement transducer

with the ACTA 3000 for dynamic
torque checking without inter-
rupting work.

A database offering measurement and analysis

functions
The ACTA 3000 system is a QA package for paperless pro-
cess auditing and tool management, which meets all the
demands of ISO 9000. It consists of a type of notebook devi-
ce, a database and measuring instrument all in one, various
measurement sensors and test joints for measuring, recording
and analyzing the tightening process, and Atlas Copco PC
software for process auditing and tool management.
However, the ACTA system does not have to be used for
assembly tool calibration management; it can just as well be
used to manage your quality assurance.
This solution offers PC-supported process auditing and tool

management which is compatible with other QA systems
and, even more besides: the ACTA 3000 is more or less a
database that can also be used for measurement and analysis.
The ACTA3000 automatically In this way, the target parameters of a joint can be deter-
communicates with the mined, for example, and tool controller can be programmed
controller for fast and secure
calibration of the tool.
using it.
ToolsTalk ACTA, a PC software (for Windows operating sys-

tems) takes care of the documentation and evaluation of the
data supplied by the ACTA. It is layout-compatible with
other QA systems, supplies trend analyses and statistics and
permits Statistical Process Control (SPC).
The ToolsTalk ACTA database contains all data on the tight-

ening tool and its supplier. It knows the life of the tool (pur-
chase price, all maintenance, repairs, costs incurred, etc.)
from commissioning to decommissioning. Of course, the
database also includes information on the location of the tool,
the place where it is used and the tool conditions.
Torque limits can also be monitored using the ACTA. Above

all, however, the database contains all measurement data ever
recorded for each tool and supplies the documentation. It
ensures that schedules are observed for preventive mainte-
nance and calibration and automatically reminds the user
when the next calibration is due. At the same time the ACTA
3000 is also a measuring instrument and controls the most
diverse functions. Thus it automatically recognizes the mea-
surement transducer used in each case and sets itself automa-
tically to that transducer (calibration filter data, etc.).
ACTA in-line transducer for

subsequent checking of the
torque.
In particular, it makes the measurement process very simple

and convenient. Even tightening curves can be recorded for
tightening condition analysis.
The ACTA system can be used not only to test and set tools
before they are implemented in production, but it can also be
used to dynamically monitor the tightening process during
assembly. Special torque limits are also used to check that
joints are seated correctly.

Standard, plug-and-play net-

work solutions and the new
generation of tightening sys-
tems make it possible to con-
trol the tightening process
remotely. Users monitor the
process in real time on a laptop
or PC via the Internet and carry
out fault diagnosis or re-pro-
gram tools as if they were actu-
ally standing on the line.
3. The new way of tightening
Information is the key to making the right decisions at the

right time. Simple access to reliable information is therefore
extremely important, especially in production facilities that
are expected to deliver large volumes of high quality
products at a high pace. The New Way of Tightening means
utilization of modern network technology to collect and
distribute tightening information in a structured way, with
minimized manual work.
In factories aiming at zero fault production many different

products with corresponding software are used. If the tight-
ening controllers are networked and integrated with the
overall production control system they can be coordinated
with other products and processes to optimize the production
flow. The collected information can be refined and filtered
by software applications to suit different purposes and users.
Significant improvements can be achieved by visualizing the
process and making information accessible to all persons
involved.
Since there are big differences in how products are produced,

all processes do not need to be on the same complexity level.
However, as in the BMW-KISSQ case (Chapter 6.9) and
most other cases of high volume production, the features that
are described in this chapter can be beneficial.
Atlas Copco has defined the four success factors that should
be focused on to integrate quality in the fastening process
and benefit from the possibilities that the new way of
tightening offers:
3.1 Zero fault production/error proofing

The best way to satisfy customers is to deliver fault free
products. The customers expect products that work, in some
cases it might even be dangerous if the product’s quality is
not secured. What is described as step 1, 2 and 3 in Chapter
1, accurate tools, connected to systems that control that the
correct number of tightenings are made on correct compo-
nents, are the foundation for long term zero fault production.
By using new technology it is now possible to implement the
features in step 4 and 5 at a reasonable cost.
Most fastening processes are more or less flexible, that is,

several different operations are carried out in one station.
The operators may need guidance to make sure that the right
operations are executed in the correct order on the correct
product. By using systems, such as Power Focus 3000/3100
and PowerMACS, that provide guidance and automatically
choose the correct settings, based on product identification
number, the number of faults can be minimized.

Tightening Production
Database Control
System
Build Data Build Data
ID ID
Monitoring
Navigation Results Results
Process Improvement
Maintenance
1. 2.
Not OK
3.
2.
OK
OK
4.
1. A product enters a station. 2. Tightenings are performed 3. The product moves forward 4. A product leaves the line.
The ID-number is read and and the results are sent to the on the line. NOT OK results, A release report shows that
sent to the production control tightening database for docu- which have not been corrected zero faults have been achieved
system. The controllers’ build mentation and further distri- in station, are handled in and documented for traceabi-
data are returned and settings bution to PCs on the network rework stations. Corrected lity.
are automatically set to cor- and the overall production rework results are stored in
respond with the product. control system. the tightening database.
When something goes wrong in the fastening process it is

important that it is rectified before that product leaves the
plant. The sooner the fault is fixed the lower the cost. This
means that, if possible, the fault should be fixed in the sta-
tion. If it is not possible to fix a problem in station the tight-
ening system needs to communicate with the overall system.
The overall system makes sure that the fault is fixed by send-
ing the product to a re-work station that automatically gets
the right settings based on the product’s identification num-
ber. When the fault is fixed the results are stored, linked to
the faulty results, so a release report can be made to show
that the product is 100% OK.
The figure above describes briefly the basic steps in a zero

fault production process.
3.2 Continuous process

improvement
No process is too good to be impro-
ved. Structured work with continu-
ous improvement is a must for all
successful industrial companies.
Statistical analysis of tightening

data tells a lot about a process.
By studying data from all tighten-
ing controllers it is possible to see
where the improvement resources
should be focused to be of most use
for the process productivity. The
quality of components and fine-
tuning of new tightening strategies
can also be determined by using
statistical reports and traces. Factory Overview is a program
for visualization, navigation
and real time monitoring of all
Visualization of the whole process at one central place helps tightening processes on one or
to avoid suboptimization. Access to all information from any more assembly lines.
PC connected to the network makes it possible for all kind of
personnel to use the information that concerns their area of
responsibility. By adding portal functionality that makes it
possible to access configuration and monitoring applications
the user-friendliness is increased even more.
3.3 Documentation for traceability

For safety critical applications traceability is a must and
more and more companies see the benefits of using it for
quality critical applications too. Traceability makes it
possible to retrieve old data and show that a tightening was
correctly made or find a few bad tightenings to make it
possible to avoid recalls of a larger number of products than
necessary. By using tools with transducers and angle
encoders values for each tightening can be stored for future
reference. If tools for quality critical tightenings, without
transducers, are connected to the database, OK or NOT OK
results can be stored.

To get complete traceability the product’s/component’s iden-

tification number and the tool’s serial number should be
stored together with the results. The tools must be calibrated
regularly and the calibrations should be documented to make
it possible to verify that the tools that have been used gave
the correct values for each tightening. Another important fac-
tor to achieved traceability is to make sure that no data is
lost. This can be done by using systems that store the results
locally and have individual numbers on each result. If the
network connection goes down it is then possible for the
collection application such as ToolsNet 3000, to come back
when the connection is back to collect the missing results
into the database.
3.4 Efficient maintenance and monitoring

Correct maintenance at the correct time reduces the total life
cycle cost and improves the equipment’s precision.
Maintenance can be planned based on time or the number of
tightenings a tool has made. Software applications can help
maintenance personal structure their work and give signals
when it is time to service a tool. These signals can be sent
via network direct from the tightening controller or from the
maintenance software. Atlas Copco’s newest electrical tools
have an intelligent service indicator that calculates the wear
based on several different parameters and gives a signal
when it is time for service.
Signals about service and other types of events can be moni-

tored in a centralized or decentralized way in real-time. Atlas
Copco software Event Monitor provides this functionality.
Filters can be used so that the right persons get the informa-
tion of interest to them visually presented on their PC screen.
By having access to all controllers and all related software
functionality it is possible to react quickly to potential
problems and work in a structured way with maintenance.
The new way of tightening:

already implemented in these
vehicles.
Some exampels of vehicles that are assembled according to ”The new way of tightening”.

4. Capability
This chapter contains short information of statistics and how

this is used when checking tools and the production.
The accuracy of a tool tells us something about the perfor-

mance, but this is not enough. The important aspect is how
the tool performs in an application, on the production line.
So, somehow we have to relate the accuracy of the tool to the
application. Every joint has a target value, but also some
tolerance that is acceptable. By relating the mean and the
standard deviation to the target value and the tolerance limits
of an application we can tell how a tool is performing where
it really matters, in its application. This is possible thanks to
different Capability Indices.
Cp
The first, and most commonly used capability index, is called
Cp. The formula for the Cp is:
Cp = Tolerance interval = HI – LO
6σ 6σ
If you look at the formula, you can see that it simply relates
the tolerance interval (HI-LO), to the process natural varia-
tion! If we have a tool with a big spread, and an application
When calculating Cp, the tolerance
interval is related to the 6σ. with very high demands (narrow tolerance limits), we get a
low Cp value. Conversely, if we have a tool with very small
spread (small σ), but very wide tolerance limits, we get a
high Cp. Of course this is what we want, because the smaller
the variation in relation to the tolerance limits, the lower the
risk of tightenings outside the tolerances. The Cp require-
ments vary. The most common is that Cp has to be greater
than 1.33. This indicates that 6 times the standard deviation
covers no more than 75% of the tolerance interval.
But is this enough for us to tell if the tool is good or bad for a
specific application? Do we need something more? Yes. The
Cp does not consider whether the mean of the distribution is
close to the target value or not. This index does not guarantee
that the distribution lies in the middle of the tolerance inter-
val. In the picture below you can see the same tool on the
same application, but before and after torque adjustment. In
both cases we would have the same Cp. If we are off target,
it is possible that the tightenings are outside one of the tole-
rance limits, even if the scatter is small in relation to the tole-
rance interval (high Cp). So we need something more that
also relates the distribution to the target value.
High Cp does not guarantee that we are close to the target value.
Cpk
The Cpk also relates the mean of the distribution to the target
value of the application. The way to do this is to divide the
distribution and the application into two different parts and
make one calculation for each side. The formula looks like
this:
Cpk = min [(HI – AVE) / 3σ , (AVE – LO) / 3σ]
First we relate the difference between the upper tolerance

limit and the average to half the natural variation (3σ). Then
we make another calculation, relating the difference between
the average and the lower tolerance limit to 3σ. We now have
two potentially different values, and the LOWER of the two
is the Cpk. If you think this is difficult, just take a few minu- When calculating Cpk also the
tes to think about this. If the average is higher than the target target value is considered.
value, then the difference between the upper tolerance limit
and the average is smaller than the difference between the
average and the lower tolerance limit. If this is the case, the

“upper calculation” will give us the Cpk, because we are

closer to the upper tolerance limit.
What happens to the Cpk if we are right on target? Well, in

this case we are as close to the upper tolerance limit as to the
lower, and both calculations will give us the same result.
In this case, we can also see that the Cpk has the same value
as the Cp.
Now we have introduced the Cp and the Cpk. By stu-

Bad Cp Good dying the formulas it is easy to see that Cp only rela-
Process not capable Process capable tes the tolerance interval to the process 6σ. Cpk also
Bad
Change tool or
adjust for good
but average needs
to be adjusted. considers the target value. We want both Cp and Cpk
accuracy. to be higher than 1.33. If our average is right on tar-
Cpk
Not possible. Process capable get, the Cp and Cpk are the same. The more off target
Good and well adjusted.
we are, the bigger the difference between Cp and Cpk.
Obviously Cpk can never be higher than Cp.
The relation between Cp When is a process capable?

and Cpk.
The question of “how good is capable?” has still not been
definitively answered. Since Cp was first used, a Cp value of
1.33 has become the most commonly acceptable criterion as
a lower boundary. The Cpk requirements vary. The most
common is that Cpk has to be greater than 1.33. A process
that has a Cpk lower than 1.00 is never capable.
It is very important that you understand why we use both the

Cp and the Cpk. If we only use the Cp, we do not know
whether we are on target or not. If we only use the Cpk, we
cannot know whether a good or bad Cpk value is because of
the centering of the process or because of the spread. So we
have to use both. Together they can give us a very good indi-
cation of how well a specific tool is performing in a specific
application. They are also the perfect way to compare diffe-
rent tools.
Look at the following dartboards:
Dartboard 1:
High Cp and low Cpk.
Dartboard 2:
Low Cp and low Cpk.
Dartboard 3:
High Cp and high Cpk
The first dartboard shows a poorly centered process, but with

a low spread (high accuracy). In this case the Cp is high and
the Cpk low. On the second dartboard, the darts are spread
randomly around the bull’s eye, but the spread is quite large
related to the tolerances. Cp is probably not so good, but if
the “mean value” is on target, the Cpk has the same value as
the Cp. The third dartboard shows a well centered process,
with high accuracy. This means that both the Cp and Cpk are
high; the process is capable.
An example:
A joint should be tightened at 70 Nm ± 10 %. A tool is tested
and we get an average of 71 Nm and a σ of 1.2 Nm.
Cp = (77-63) / 6*1.2 = 1.95

Cpk = min [ (77-71) / (3*1.2) , (71-63) / (3*1.2) ] =
min [ 1.67, 2.22 ] = 1.67
Both the Cp and Cpk values are greater than 1.33 and the
process is capable and does not need to be adjusted.
Machine capability indices

As you now know, Cp and Cpk are process capability indi-
ces. Everything that affects the process affects these indices.
But if we take away all variation affecting the assembly pro-
cess, except the variation in the tool itself, we get what are
called Machine Capability indices. These must be measured
under very controlled circumstances, preferably in a tool
crib. The tests should be carried out on the same joint and by
the same operator (or even better, by placing the tool in a
fixture in order to get rid of all the operator influence). The
calculations are the same for Cm as for Cp, and the same for
Cmk as for Cpk.
So remember, Cp and Cpk determine whether the process is

capable. The Cm and Cmk determine whether the machine
(tool) is capable.
What else is there to think about?

When you analyze the capability of a tool, the sample size is
of great importance in order to obtain reliable mean and stan-
dard deviation calculations. A sample size of at least 25 tight-
enings is strongly recommended.

5. The three tightening tool

condition classes in accordance
with VDI guideline 2862
There are many different joint standards defined by compani-
es and organizations. Most of these standards are very similar
to each other. One example of standard is the VDI guideline
2862, “Use of tightening systems in the automotive industry”.
It is a German standard that has been valid since 1 July 1999.
This was compiled by the “Technical rules for the use of tigh-
tening systems in the automotive industry” study group from
the VDI “Production Technology” section.
This guideline divides tightened joints for vehicle manufacture
into three risk classes:
Risk class A
(“Direct or indirect danger to life and limb”) takes effect
“when the failure of this bolt position is highly likely to lead
to safety failure and/or to the destruction of the entire vehicle
and thus direct or indirect danger to life and limb is indica-
ted.” These kind of joints are also called “Safety critical
joints”.
Risk class B
(“Dropout”) is indicated “when the failure of the bolt position
will lead to malfunction of the vehicle.”
Risk class C
(“Annoyance for the customer”) then applies “when the failure
of the bolt position would provoke annoyance among custo-
mers.”
The minimum demands for each risk class can be achieved by

fulfilling the solutions in the 5 steps to zero-fault production
(Chapter 1) according to the table below.
Risk class Solutions as described in chapter 1

A Step 4 or higher
B Step 3 or higher
C Step 2 or higher
6. Error proofing cases
6.1 Onan Cummins Power 6.2 Ford Camaçari, Brazil. 6.3 Hoerbiger, Austria.
Generation, USA.
6.4 Johnson Controls, USA. 6.5 Dutch Air Force, Netherlands. 6.6 Daimler-Chrysler, Germany.
6.7 InterBrew, Belgium. 6.8 Krause Maschinenfabrik, 6.9 BMW Group, Germany.
Germany.

6.1 Preventative maintenance system at

Onan Cummins Power Generation
Onan Cummins Power Generation in Minneapolis USA
manufactures gas and diesel power generation equipment.
In an effort to have control over their tools and tightening,
Onan decided to set up a preventive maintenance system
consisting of an ACTA 3000, ToolsTalk ACTA software and
in-line transducers. Bruce LaMirande, responsible for Torque
and Fastening control, reports on the impact of their new
QAT system from Atlas Copco.
“We have been using this system for two years. One major
advantage is that we can now predict how long a tool will
run before further maintenance is required. Secondly, now
we can check how the tools are being used and tell exactly
when they are due for maintenance. Previously we threw
away tools when they broke down, which was very costly
Static torque check with ACTA and time consuming. Now we are using tools longer and
3000 and MRTT at Onan realizing tremendous cost savings.
Cummins Power Generation.
“We now have complete control of our tools and tightening.

Previously, we sent tools out for maintenance and recalibra-
tion and got them back some 3 - 4 days later. Now we need at
most 45 minutes to bring a tool in, repair it, do maintenance
and calibrate it before it is out running again. This means that
we have both a more efficient and a faster system.
“Furthermore, by using ToolsTalk ACTA we now have just

one software system that keeps calibration, maintenance etc.
And the complete ACTA system helps us actively trouble-
shoot tool problems before they occur. At every moment we
can tell how the tool is running and performing on the
assembly line.”
6.2 Fast and accurate tool calibration at

Ford Camaçari
Mr. Ricardo Samori is an Industrial Engineer Supervisor at
Ford Camaçari in Brazil. He is responsible for the implemen-
tation of a Quality Assurance Process in line with Ford’s
strict quality procedures. Ricardo says “We were looking for
a level of Quality Assurance that would ensure purpose-
designed customer satisfaction and a level of quality that
would exceed our customer’s demands and expectations.”
This Ford facility opted for a total Atlas Copco solution

including: Tensor DS, ACTA 3000, IRTT torque transducers,
the MRTT Torque Wrench and finally the dedicated softwa-
re, ToolsTalk ACTA. One of the main tasks with this new
system was to calibrate and update the Tensor DS systems
in a fast and accurate way by using the ACTA 3000.
The ACTA 3000 connected to the Tensor unit, compares the

calibration torque in the DS with the values received from
the transducer connected to the ACTA 3000. After the tight-
ening is done, the ACTA 3000 calculates the new torque
tuning value for the Tensor DS and sends it over to the unit.
It also stores the value in the ACTA 3000 tool database. The
Tensor DS is now updated and ready to be used again.
“The ACTA 3000 solution has empowered our Process

Analyses and created a straight communication channel to
improve our overall quality results. Furthermore, ACTA 3000
makes it possible for us to guarantee the process results,
especially when it comes to DS applications, by synchroniza-
tion between the ACTA 3000 and the Tensor DS.”
The ACTA 3000 is used for tool

calibration directly on the
assembly line.

“What we really appreciate is that the ACTA solution includ-

ing the dedicated ToolsTalk ACTA software is so easy to
work with. We save time, have total control over tool
performance criteria and maintenance and rule out data
manipulation. Thanks to the special features like Search
Filters, Tools List, and also the complete information
generated through the graphs, reports etc. we don’t need
additional software to complete the documentation audit.
With the tools and the quality audit system from Atlas Copco
we now feel we have both the best tools and the best quality
control system available” concludes Mr. Samori.
Process 6.3 Valve fitting at Hoerbiger

quality
How do you get process control of two air tightening tools
Step 2
within budget limits when they have to tighten one hard and
one soft joint reliably with precision of ±5 % and ±12 %.
Hoerbiger in Wiener Neustadt was faced with precisely this

question but did not want to pay 7000 euro including tool for
a solution. The budget for the associated workstation had
already been set for half that amount. In addition, joints
requiring documentation, for which a controlled electric tool
would have been needed, were not required. Quality assur-
ance demanded merely reliable process control and recogni-
tion of incorrect tightening. In this case the solution was an
“RE-Tightening controller” developed by Atlas Copco Tools
and Assembly Systems: a small electronic device that was
actually only designed to monitor a single tightening tool
was used for the first time at Hoerbiger to control two tools
The RE-Tightening controller plus operator.
observes the number of
tightenings.
This was necessary for the assembly of a control valve. The
main mounting bolt and cover had to be fitted. First, a bolt
with a synthetic seal is tightened onto the mounting with a
LUM screwdriver. The part is turned round and the cover is
tightened. After this, the operator puts the control valves
assembled in a four-stage cycle into a test station at the
assembly point and loads the next four parts for assembly.
If the operator should by mistake go for the wrong tool, the

operator will notice this because the tightening adaptor will
not fit the mounting bolt that has to be assembled first.
Confusion between tools is thus ruled out. In addition,
between assembly cycles, that is when changing from Pressure and time control
tightening mountings to tightening covers and vice versa, the Torque
operator has to operate a program selector switch to switch

the RE-Tightening controller over to the appropriate tool or
tightening program. At the same time, the selector switch Pressure 2
also controls the direction of the air supply so that air is sup-
plied to the selected tool and the one used previously is cut
off or not under pressure.
Pressure 1 Time
This switch was the real trick in being able to monitor tools Rundown time Clamping
time
with the tightening controller. A simple relay circuit pro-
duced by Hoerbiger itself was used for the purpose. This With the RE-Controller this
tightening process is moni-
gives the controller the time needed to set the appropriate tored with simultaneous pres-
parameters when changing over. sure and time measurements
and two separate pressure/time
windows. This allows operator
Otherwise, the RE-Controller operates in its normal mode: errors to be detected.
for double security, the tightening process is checked with
simultaneous pressure and time measurements and two sepa-
rate pressure/time windows (see diagram). In this way a
particular (programmable) pressure value is assigned to the
assembly time and to the dwell time at the point of discon-
nection.
As a result, operator errors (finger taken off the trigger too

soon) are detected, as are any defects there may be in the
component or bolt since the head support is reached too soon
or the bolt cannot be tightened fully home. In this way, the
controller at Hoerbiger also discovers, for example, whether
the seal under the main mounting bolt has been forgotten.
Above all the mains and flow pressures are checked since the
problem free operation of the disconnection coupling is
dependent on these. If the working pressure is insufficient or
incorrect tightening (NOT OK) occurs, the controller raises
visual and audible alarms and cuts off the tool.
Besides the visual and audible OK and NOT OK checks on

individual connections, “Total OK” processing per individual
cycle and overall cycle is also possible. Furthermore, using
Total-OK and potential free signal inputs and outputs, down-
stream assembly stages can also be positively controlled.
Since the RE-Controller recognizes when the programmed

number of connections per cycle have not been made (cycle

control), the signal raised in NOT OK cases is used to pre-

vent the work piece going to the next stage until the cycle
has been completed and is OK. Conversely, the controller
will not switch over to a new processing cycle if the work
piece is still missing. The controller is incidentally very easy
to program and this can be done within minutes with a sepa-
rate input device no bigger than a TV remote control.
Separating the controller and programming device rule out
unauthorized changes to the control parameters. The only
precondition for using this “low-cost” process control is a
disconnectable tightening tool with an RE (RE = Reporting)
signal output. They may equally well be compressed air tools
or DC tools from the Eliza range.
6.4 Cutting costs and improving quality at

Process Johnson Controls
quality
Step 3
Johnson Controls is a major supplier of integrated seating
and interior systems for the automotive industry. At Lake-
wood in Michigan, USA, the company produces headliners
for large car manufacturers. Prevailing torque, different
screw lengths, and varying joint types are typical challenges
the production engineers face when planning the assembly of
handles, visors and other components to the interior roof.
“This previously led to problems such as cross threads, mis-
sing components and damaged parts,” states manufacturing
engineer Phil Green.
Several Tensor DL screwdriver systems were introduced,

allowing for integration to the line via PLC. Johnson
Controls now has full line control and feedback. Six
conventional tools were replaced with only three Tensor DL
systems per station. These systems are set to tighten 16 Torx
screws at 1.6 Nm with a tolerance of ± 10% of target torque.
Phil Green explains that the new tools have improved quality
The Tensor DL has improved control and taken away the need for final inspections. The
quality and taken away the investment paid off in less than eight weeks by eliminating
need for final inspection at
Johnson Controls.
assembly problems, cutting preventive maintenance service
costs by 70%, and reducing downtime. “The ability to
change rundown speed and torque settings with the tool has
given me the flexibility we never had before,” he says.
“Our experience to date has been excellent with the Tensor
DL and the performance and lack of downtime has just
amazed me.”
6.5 Wheel fitting for the Dutch Air force

“Touch and go” as pilots have it forms a regular part of the Process
training plan of the Royal Netherlands Air force 334 quality
Transport Squadron in Eindhoven in the Netherlands. “On Step 3
their training approaches our transport aircraft hit the run-
way up to ten times an hour and then lift-off again immedi-
ately” says Jake Verhoef. An officer of the U.S. Air force,
he is responsible for maintenance and servicing at the
Eindhoven Air Base. This involves tire changes above all
else since many takeoffs and landings an hour like this are
torture for landing gear and tires. The team in the “Tire
Workshop” (a combination of tire store, workshop and test
stand) has its hands full keeping the squadron’s eleven trans-
port aircraft “rolling”.
Yet their work used to be even harder. Previously, all wheel

bolts were tightened with different impulse tools and then
given a final tightening with a torque wrench. With torques
between 10 and 200 Nm, that was quite a strain on the arms.
What is more, it took much longer than today when two
tools known as “Tensor DS” are sufficient for the same work To improve the process quality
and make the final tightening unnecessary. of fitting tires to the aircraft of
its 334 Transport Squadron,
the Dutch Air force now uses
The base discovered the Tensor DS when the equipment for only two tools for all types of
a new type of tire workshop was being planned. After they tightening jobs. An angle and
a pistol grip nutrunner from
had almost decided to invest in a tightening system as used the Tensor DS Series. The tigh-
in the car industry for particularly critical joints requiring tening control system only
documentation, “Atlas Copco demonstrated a new type of releases the wheel when all
the bolts have been correctly
tightening tool that appeared almost tailor made for our tightened in three stages (mili-
needs”, recalls Willem Roelofsen, Adjutant of the tary regulation).
Maintenance and Engineering Support Office.
It was less expensive but offered the same zero-fault securi-

ty and versatility in the programming of different types of
tightening job. Admittedly, it was not possible to record the
results of tightening using this solution but that was not cri-
tical at all. What was more important was that it was possi-
ble to cover the widest possible torque range without defects
with one tightening tool and while no longer having to do
the final tightening. It was also a relief that this was not a
special solution requiring engineering input but a standard
product.

Two tools for everything

Today, two tools are sufficient for fitting wheels to the five
types of aircraft stationed at Eindhoven – a Tensor Model
DS9 pistol grip driver (for 80 to 200 Nm) and a DS7 angle
tool (10 to 80 Nm) share the tightening jobs on the nose and
main wheels. As each one can store eight freely program-
mable torque parameters, it also replaces 8 conventional
tools. This saves space and money. Using these two tools,
the wheels of all transport aircraft are today fitted with a
torque accuracy that could not be achieved with the former
compressed air tools. Above all, the DS tools offer high pro-
cess-security since they monitor operator and work piece
simultaneously. Thus their control system only releases a
wheel when all the bolts have been tightened accurately and
in the prescribed sequence. If the set torque is not achieved,
perhaps because the thread is damaged, an LED on the tool
shows red. If everything is OK, the green LED lights up.
Should a wheel bolt inadvertently be tightened twice, the

tool control system detects this in exactly the same way as a
forgotten bolt.
Consisting of two rim halves and a (tubeless) outer case and

inflated to 7.5 bar, the aircraft tires must be tightened evenly
and reliably because they heat up seriously in the continuous
“touch and go” maneuvers.
Three-stage tightening
As a result, the twelve bolts on the nose wheel of a C-130-
Hercules are not merely tightened using a set crossover pat-
Tire fitting on the main gear of tern. The maintenance instructions also require that every
a KDC-10 tanker aircraft: the individual joint be tightened in three stages. This evens out
Tensor DS pistol-grip nutrun-
ner acknowledges every OK
locked-in stresses and avoids distortion of the rims. The
tightening operation with a tightening torque required for each stage is set in the elec-
green LED signal (top at the tronic tool beforehand in programming mode and retrieved
end of the tool). If the red LED
lights, the operator must on a case by case basis way via a torque selector switch on
change the tool direction to the Tensor DS tool.
reverse and re-tighten the bolt
afresh. If the tool then shows
red again, the bolt is changed.
Officer Jake Verhoef is delighted:

“In comparison with the previous air tools, bolt assembly is
today not only quieter and more vibration free but, in parti-
cular, free of oil. Above all this Tensor tool meets all the
requirements of the European Joint Aviation Regulations,
which are based on the Federal Aviation Authority (FAA)
rules. What is more, they came out cheaper than the tight-
ening system we had previously budgeted for although
obviously much more expensive than the time honored sock-
et sets. But precision has risen so enormously and fitting
times have fallen to such an extent that our investment will
quickly pay for itself”, reckons Willem Roelofsen.
6.6 Tightening tool approves seal tightness

at the same time
Process
Normally, controlled tightening tools only consider whether
quality
the tightening parameters and tightening cycles have been Step 3
correctly complied with. However, at Daimler-Chrysler’s
truck factory in Wörth am Rhein, electric tightening tools
also check whether O-rings and clips have been properly
installed in the assembly of union bolts on valve blocks. If
not, the tightening tool raises the alarm and simply will not
function.
This automatic checking of the complete assembly process is

implemented using Tensor-Electric tightening tools from the
DS range which simply acquired different software for the
control electronics (DigiTork) and special adaptors for the
offset drive known as rotation distributors with pneumatic
connections. These have a sprocket that fits the bolts from
which a peg projects. Before every tightening operation, a
pneumatic interrogation lasting only 600 milliseconds of
whether the O-ring and clips are present in the union bolt
takes place via air-passages (Ø 2 and 3 mm) in the peg. The whole secret of the Tensor
DS nutrunner, which is here
used for reaction free tighten-
The indicator is a difference between impact and flow pres- ing down of the covers of the
sures of only 0.15 bar. Accordingly the control system first canisters, is their servo-elec-
tronics known as "DigiTork". It
checks whether test air is actually getting to the rotation dis- requires no external sensors
tributor and whether the air passages are clear. The check for torque or angle but never-
whether the union bolt has all its components takes place theless controls them – using
the current, voltage, frequency
automatically every time the start button on the handle of the and temperature of the servo-
tool is operated. The control system only releases the tool motor as values.
when the required pressure differential has been measured
and the test air turned off again.

If O-Rings and clips are not

properly fitted, the tightening
tool raises the alarm and will
not start at all. To do this, the
production team simply ac-
quired the Tensor DS nutrunner
with its DigiTork Control Unit
and special adaptors for the
angle nutrunner.
If the bolt connection is OK, the whole cycle starts again

from the beginning. On the other hand, if the bolt connec-
tion is NOT OK, the tightening control system conveys this
Process
to the operator visually. The flow pressure monitoring then
quality
remains switched off. Only when the operator has disman-
Step 4
tled the parts and has confirmed this can a new cycle begin.
However, if there is no impact pressure, the O-Ring and/or

clips have not been properly installed, a visual warning sig-
nal is given and the tightening tool is not released. The op-
erator must then rectify the fault and confirm that this has
been done to the control system. Only then can tightening
continue.
6.7 The new way of tightening – beer barrels

Even though they don’t look it, tap hole bungs in beer bar-
rels are a difficult tightening job. When tightening the tap-
ping lines, locked-in stresses arise through the rubber seals
and traditional tightening tools can not carry out a proper
closure. This was what the Belgian InterBrew group and its
Stella Artois brewery found when they had to convert 1.1
million barrels to a standard closure system. Before that,
In Belgium, 1.1 million beer
barrels had to be re-equipped
each of Interbrew’s three Belgian sites had its own system.
with a standard closure sys- This entailed many empty trips between the sister breweries
tem. In order to reduce the in Brussels, Liège and Louvain since each one could only
severe locked-in stresses that
occur in the process, operator
refill its own barrels. After refitting with a standard closure,
controlled high-tech tools from which Interbrew calls “MicroMatic”, the barrels could circu-
the Tensor S class were chosen. late freely.
Although this tightening operation was not safety critical and

thus did not need to be documented, Interbrew chose the
same high-tech Tensor S Class tool as the car industry pre-
scribes for safety critical bolts on airbags, steering and safety
belt anchorages. These controlled electric tools even have an
own assigned Ethernet TCP/IP address.
Locked-in stresses from tightening are

compensated for
By doing this, hasn’t Interbrew bought itself a Lamborghini
just to do the shopping? Jan Van de Bergh refutes this firmly:
“Our self-imposed quality standards demand one hundred
percent security for every barrel”. He is the team leader from
the energy and fluids department at the Stella Artois brewery
in Louvain. “In any case, it was the only way we could get a
grip on the extreme locked-in stresses.” The MicroMatic tap-
ping lines had to be tightened into the tops of the barrels, or
“kegs”, with a torque of at least 70 but no more than 80 Nm.
Not exactly a simple tightening job, since the joint behaves
extremely soft because of the seal.
Team leader Jan Van de Bergh

with the new MicroMatic barrel
closure. To the left of the pic-
ture, the Power Focus unit. This
has its own assigned Ethernet
TCP/IP address and can be cal-
led up from anywhere, for
example to carry out a remote
analysis of the tightening job,
to make a program update or
simply to look at the tightening
results for the day.

In order to keep within tolerance at all times, the tools are

programmed for a target torque of 75 Nm. “Even with
variations of ±4 Nm, they always produce fault-free joints
that way”, Van de Bergh declares. The tapping line is first
pre-tightened, then the tool stops for 50 milliseconds at a de-
fined torque. This in order to compensate for locked-in stres-
ses in the seals. After that it switches over to the final tight-
ening of 75 Nm. If the tool control unit indicates “OK”, all
is well.
Since, in this operation, the tightening control unit also

monitors the angle of rotation, then faulty seals and damage
to threads are detected before filling. With the old barrel
closures and without the current electric tools, these faults
could only be detected in a water-bath after filling: if air
bubbles came out of the bolt connection, it had to be retight-
ened somewhat and immersed again. If the barrel still was
not gas-tight, the complete keg closure had to be exchanged.
Programming by laptop
It goes without saying that alternatives had been tried before
the investments. However, the difference in cost compared
with other compressed air or electrically based systems was
not very great.
“We were surprised how simple programming and operation

are and therefore we went for the solution with the most
functionality. From my laptop I can set the tightening para-
meters for torque and angle at any time and call up extensive
Jan Van de Bergh uploads cur- graphics and tightening statistics”, says Van de Bergh. He is
rent tightening data on his lap- aware that far from all the functionality that the system
top, with which he also pro-
grammed the tools. offers is (yet) being utilized: “We run our six Tensor S tight-
ening stations independently of one another, but we could
network them at any time, through their assigned TCP/IP-
addresses, so they can communicate with each other over an
Ethernet.”
So, at the moment, the Lamborghini is only being driven in

bottom gear, but this could change. For the future, the team
leader envisages fitting every keg with a chip that would
store all the barrel data. “When a barrel comes back to the
brewery, we could carry out a goods inwards inspection with
the tightening system. This would bring up the dates of the
last filling and the last pressure test, torques, angles of rota-
tion, etc. and direct the kegs accordingly within the brewe-
ry”, for example to the maintenance department if the regu-

lar seal change has to be carried out or damage to a thread is
detected. “Only then could we get the most out of the com-
munications capabilities of these tools”, speculates Van de
Bergh.
6.8 Ford uses the new way of tightening on its

engine lines
Process
For its 6 and 8 cylinder engines, Ford has decided on the
quality
“new way of tightening” (chapter 3). This means tightening
Step 5
systems from the new Power generation: Power Focus and
Tensor tools for hand assembly stations and PowerMACS
tightening tools on automated stations.
Johann A. Krause Maschinenfabrik in Bremen builds all of

Ford’s European lines as complete installations and acts as
Ford’s “continuous source” for planned new engine lines
around the world.
Dipl.-Ing. Detlev Kalter is in charge of tightening technolo-

gy at Krause. Because he was inclined towards Atlas Copco
and the new way of tightening he had some persuading to do
at Ford’s headquarters in Dearborn, USA. Ford had already
ruled out Japanese tightening tool suppliers, and although
Atlas Copco had better coverage of the market in Europe
and the USA, Ford were still concerned that a shift from
pneumatic to electric tools could involve risks associated
with the new technology. From his long experience with
Cylinder head assembly: Two

ten nutrunner machines are
here tightening down the
cylinder head of a V8 engine.
Tightening is controlled by
torque /angle of rotation
(40 Nm, 180°) with a precision
of ±2,5 % over 6 sigma (i.e.
Ford A10 classification).

Krause, Kalter objected to this: “Today, anyone can tighten

a bolt securely but the will and ability to provide holistic,
worldwide collaboration is something few can offer.” In
addition, he saw that the new Power generation tightening
systems fulfilled many of the wishes that Krause had been
expressing for years.
What was more, Ford required “commonality”, the stan-

dardization of all equipment for the same task and one Ford
plant in the USA had already decided on Atlas Copco’s new
generation of tightening tools for a new gearbox line. This
strengthened Kalter’s case for the new Power Focus and
PowerMACS generation of tightening tools.
“In the end, the deciding factors were the technology, the
price and the fact that Atlas Copco is present with service
and training wherever there are Ford plants”, is how Kalter
sums up the final decision to go for Power Focus and
PowerMACS tightening tools as the “global solution”.
Space saving
On a tour of the assembly stations for a V8 line, Kalter
points out the space saving wall-mounting of the
PowerMACS stand-alone modules and the fact that they
make do with only one cable to the nutrunner “whereas old
model tightening tools had three”. Since no account need be
taken of the tool equipment cabinets in the layout of the
tightening stations it was possible to make them much smal-
ler than appeared practical in the outline drawings for tender.
The control systems are accommodated where they do not
need too much additional volume. Only the network hubs,
mains safety devices and any additional mains filters re-
quired need to be housed in the equipment cabinet on the
Dipl.-Ing. Detfel Kalter, in
charge of tightening station. All this minimizes expensive container area.
technology at Krause.
Sump tightening: all twelve

M8 bolts on the sump are
tightened simultaneously to
20 Nm with torque/angle of
rotation control (60°) by the
twelve special tools. The PC to
the right of the picture was
wanted by Ford in order to be
able to interface with the
tightening system right in the
tightening station and to be
able to use the graphics capa-
bilities of the control system
simultaneously on the spot
instead of only on a PC
belonging to the line control
system.
On the assembly station for the V8 sump (see picture above)

the conversation turns to the operator interface of the new
tightening system. This has an enhanced graphics function
for programming and operator guidance via PowerMACS
HMI Annunciator (photo-quality bitmap images monitoring
in real-time tightening sequences). All stations on this line
are equipped with a control and display PC. The PC is not
necessary for the regular system, says Kalter. Programming
of the tool can be done equally well with a laptop. “But if
the customer wants, a PC can be connected to the
PowerMACS-System without problem.”
With its 6th generation of MACS automated tightening sys-

tems (PowerMACS) within ten years, Atlas Copco is now
for the first time breaking away from the convention of
always having to build the control system into the equipment
cabinet. The new systems will cope with all tightening and
monitoring and methods that have been developed over the
course of time. The software options also include fault
management (reject management), which Ford uses to re-
duce re-working off the tightening station.

Internet access is becoming important

An important thing for Ford was to be able to do program-
ming and select spindles on the station at any time, to view
the actual values and tightening curves and maintain statis-
tics. “Visualization via Windows is an important argument
for us”, stresses Kalter. The thing that is not yet provided in
the assembly line in question is the use of the TCP/IP
addresses of each tool to communicate with the systems via
the Internet. But a project for another customer is already
showing how much this type of communication, which
forms a standard part of the new way of tightening, “will
soon become the deciding factor”, as Kalter puts it.
The new lines for Ford will keep Kalter busy for a while. In
the autumn of 2001 he reckoned that the program would go
on for six or eight years.
6.9 KISSQ: Zero defect assembly in the BMW

Group
Process
quality They don’t talk about the zero-fault goal in the BMW Group
Step 5 but nonetheless, in tightening, they practice zero-fault pro-
duction according to Step 5 (chapter 1). Every work element
and equally every bolt assembly is controlled, checked and
recorded by a plant and company quality management sys-
tem. This is provided by the “Core manufacture integrated
quality management system [Kernfertigung integrierende
steurerungssystem Qualität]”, known as KISSQ. This is the
quality building block of the central KISSQ system whose
goal is to optimize working procedures and hence increase
productivity and cost effectiveness. Since 1997, KISSQ is
being introduced step-by-step at the BMW Group’s plants.
The KISSQ-System provides computer supported collection

of all quality data of every phase of production. From the
bare shell to the final assembly and driving dynamics testing
together with control of re-work. To do so it uses the auto-
motive industry package of the standard Central AQ (QSYS
standard software). This is a quality management system
developed by IBS AG. It allows collection, inspection, ana-
lysis and documentation of all production quality informa-
tion in a central database and is stored for every vehicle
assembled, including the data on the tightening process.
Here too, the data communication process between tighten-

ing tools and their control units (master and slave) and the
tool control system and the management system (in this case
KISSQ) described on page 49 applies in principle. The tight-
ening process starts only when the vehicle has been identified
and the operator is guided through visual signals for OK and
NOK, organization of the tightening sequence and settings
of parameters via socket selectors.
Everything is based on the plug-and-play IT of the new way

of tightening with Ethernet based ToolsNet server and the
Power Focus 3000 tightening systems for hand assembly
points and PowerMACS tightening tools.
The BMW Group also has two program building blocks

from Atlas Copco (based on IBS solution) that are KISSQ
specific and are known as ToolsIdent and ToolsControl.

Assembly at the BMW plant in

Dingolfing, Germany.
ToolsIdent is the interface to the automatic vehicle identifi-

cation. The barcode of the chassis number is not read in seri-
ally with a hand scanner but is transmitted wirelessly i.e.
mobile media (Station A in the diagram on page 49).
ToolsControl takes on a standardized interface function at

production management level. It is also used as an interme-
diate buffer to distribute target and actual data in communi-
cation with KISSQ. The ToolsControl software tool can
store and handle up to 400 chassis numbers and vehicle data
sets, which are communicated in advance by KISSQ via the
line management system’s SPC in accordance with the pro-
duction plan.
The tightening control system finds out what program jobs

(parameter sets and number of bolts) are to be carried out
and in what sequence via ToolsControl. ToolsControl also
handles sending back the status evaluation (OK or NOT OK)
of every tightening and job. In order to record the data, the
tightening tool control system reports directly to the
ToolsNet Server.
Defective bolts automatically appear in central re-work

management or on wireless terminals right on the assembly
line where they are immediately processed. A wireless instal-
lation is incorporated at KISSQ level for mobile re-work on
the line.
The mobile terminals are also equipped with a barcode scan-

ner for identifying the re-working operator and vehicle. The
part for re-work is indicated. The results are acknowledged
and added to the history of the work on the vehicle. Only
when all re-work demanded by the KISSQ-System has been
done, is a final OK generated and the vehicle released.
Additional Data Production

Comm System Control System
KISSQ Server TCP/IP Ethernet
Datalogic
NT Workstation
Ident M with ToolsNet/Costa
ToolsControl
& ToolsIdent PF3000 Color
Moby E
with ToolsControl
Production
as Master
Control
System
PF 3000 Compacts PF 3000 Compact as slaves

Identi-
fication
Socket Socket
Selector selector
The tightening tools communicate with the KISSQ quality management

system through an ordinary NT server.
Example in the BMW Oxford plant in England.

Atlas Copco Pocket Guides
Title Ordering No.
Air line distribution 9833 1266 01
Air motors 9833 9067 01
Drilling with hand-held machines 9833 8554 01
Error proofed production 9833 1437 01
Grinding 9833 8641 01
Percussive tools 9833 1003 01
Pulse tools 9833 1225 01
Riveting technique 9833 1124 01
Screwdriving 9833 1007 01
Statistical analysis technique 9833 8637 01
The art of ergonomics 9833 8587 01
Tightening technique 9833 8648 01
Vibrations in grinders 9833 9017 01
Recyclable paper. Jetlag 2004:1. Printed in Sweden

9833 1437 01
www.atlascopco.com

Error Proofed Production: Five Steps To Achieve Zero Fault Fastening

Uploaded by

Copyright:

Available Formats

Error Proofed Production: Five Steps To Achieve Zero Fault Fastening

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

Error Proofed Production: Five Steps To Achieve Zero Fault Fastening

Uploaded by

Copyright:

Available Formats

PocketGuideError 9833 1437 4-02-18 15.

Error proofed production

Error proofed production

The later an assembly defect is identified, the more it costs to

Atlas Copco has defined five tightening process control steps

With this five-stage classification we have tried to provide an

Error proofed production

2. Tool management and process auditing .................................................. 16

3. The new way of tightening........................................................................ 20

4. Machine capability of tightening tools:

5. The three tightening tool condition classes in

6. Error proofing cases .................................................................................. 31

ERROR PROOFED PRODUCTION 3

1000% 100% 10%

1. Five steps to zero-fault

When Henry Ford realized assembly line production, process

It is no accident that the automotive industry (see chapter 5:

Process quality also reduces costs. Generally, it is true that

4 ERROR PROOFED PRODUCTION

Step 1. To assure a correct tightening torque

Good examples of tools to be used when taking the first

At step 1, process quality is

ERROR PROOFED PRODUCTION 5

At step 2 the operator is also

Step 2. To assure that all screws are tightened

6 ERROR PROOFED PRODUCTION

in the tool at the moment of shut-off when the tool comes to

The RE-controller is also used for counting the number of

At step 3, the tightening tool is

Step 3. To assure that the joint is correct

The way to detect these types of faulty joints is to monitor

ERROR PROOFED PRODUCTION 7

8 ERROR PROOFED PRODUCTION

At step 4, all tightening data is

Step 4. To assure that safety critical joints are

This proof requires advanced controlled industrial tools such

1. Torque control using a transducer that is traceably

Fastening seatbelts is a typical

ERROR PROOFED PRODUCTION 9

If there are indications that a recall is related to the fastening

It is vital that the spindles are calibrated before delivery and

Power Focus 3000/3100 and PowerMACS controllers are

An assembly cell consists of one or several components with

At step 5, the tightening tool is

Step 5. To assure zero fault fastening

ERROR PROOFED PRODUCTION 11

Phaeton production is almost

Despite all the functionality described above, cross threads,

ERROR PROOFED PRODUCTION 13

1.1 Functionality Step 1 Step 2

Air-powered mechanical As for step 1, but with

Torque control Torque setting via Torque setting via

Step 3 Step 4 Step 4 Step 5

Tensor DS/DL electric Tensor S/ST elec- QMX spindles with

ERROR PROOFED PRODUCTION 15

2. Tool management and

How often tools need to be torque audited is depending on a

Without process auditing and tool management, let alone a

The demands can be found in quality standard ISO 9000,