2005 WJTA American Waterjet Conference

August 21-23, 2005 Houston, Texas

Paper 7B-2

STRATEGIES FOR COST- AND TIME-EFFECTIVE USE

OF ABRASIVE WATERJET CUTTING
U. Andersson, G. Holmqvist
Chalmers University of Technology
Department of Materials and Manufacturing Engineering
Gteborg, SWEDEN

ABSTRACT
Abrasive waterjet (AWJ) cutting is today an important industrial cutting technology. With an
AWJ almost any material can be cut, and the method is also considered as environmentally
friendly. However, there is no easy and practical way of optimising a cutting operation for either
maximised production rate or minimised manufacturing cost. This paper will present the cost
structure for a typical machine set up, and discuss the economical considerations that have to be
taken into account to be able to optimize the cutting operation for cost- and/or time-effective use.
An ongoing project together with the Nordic waterjet industry is focussing on reducing cost and
machining time for the project members within their operation of their abrasive waterjet cutting
machines. The project members today rely on their own experience when deciding how to use
their equipment in the best possible way. The reason for this is of course that the issue of
optimizing the process is relatively complex.
The objective for this paper is to present a strategy for how to work with the abrasive waterjet
cutting to get the most out of the process, with focus on economical considerations. The
discussions will be illustrated by economical calculations focusing on how machine settings and
cutting parameters affect the total process economy.

Organized and Sponsored by the WaterJet Technology Association

1. INTRODUCTION
Within an ongoing Nordic project called CUT- Competetive Use of waterjet Technology, the
fundamental economical issues in the matter of abrasive waterjet cutting have been studied. The
network project today has 12 members from industry, developing or using abrasive waterjet
equipment. The main issue within this network is to reduce costs for producing parts, and
through a more effective use of waterjet cutting machines improve the competitiveness of the
companies. This paper will present some important observations made during the first phase of
this project.
As a part of the project an optimization model is being developed. The objective is not just to
develop a mathematical model but most important to incorporate typical industrial setups and
issues such as limitations in pressure, flow rate and number of cutting heads. Also the project has
focussed on determining models for instance for how the pressure level influences maintenance
cost. Further the output result must be in a form so that they can be directly applied in the
practical case.
2. COST- AND TIME-OPTIMIZATION MODEL
As a part of the project, a model for determining optimized parameter settings is being
developed. In table 1 can be seen all the parameters used in the economical calculations. The
model suggests two separate optimization parameter settings for:

shortest cutting time and

lowest cost per cut meter.

The optimization for lowest cost can be of relevance in many cases where a longer cutting time
can be accepted. On the other hand, the case of shortest cutting time is of interest when a higher
cost can be accepted.
The model has three parts:
I. Input section:
Cost situation for the job shop (includes all major costs associated with the cutting
operations).
Limitations and possible set-ups (maximum pressure, maximum number of cutting heads,
utilization).
The specific cutting case (material, thickness and quality requirements).
II. Database:
Contains 500 different parameter combinations along with the costs and cutting times for
each combination.
The database is updated outgoing from the input section data.
A function for finding minimum points.

III. Output section:

Parameter settings for the two different optimization cases
Cost structure
The use of a database has the advantage of yielding discrete values, which have been chosen at
industrially relevant levels. The model is still under the course of development, but an example
of the use of the model is given in table 2 (input) and table 3 (results). The cost structures
presented later in this paper are results from the model.
3. COST ANALYSIS
Economical optimization of the abrasive waterjet process is relatively complex. If it is supposed
to be done realistically, there are a lot of parameters that have to be considered. This paper deals
with the parameters presented in table 1. To be able to perform a correct analysis of the total
costs for a specific machine set up it is essential to consider the conditions under which the
machine is to be run. Different machines of course also have their limitations in capacity
(pressure, flow rate etc.). The cost analysis made in this paper is based on the expenses for
producing a 1 meter cut. The most important economic factors are derived and later discussed
focussing on cost and/or time effective use.
3.1 Cost Split-Up
To find out the cost distribution for typical machine settings, cost split up analyses were carried
out for tree different machine settings. These settings are listed in table 4. One machine powered
with a 37 kW pump and two machines with a 73,5 kW pump. The corresponding cutting speed is
calculated basically according to Zeng and Kim (1993) but the model is adapted for higher
pressures, in this case 380 and 410 MPa.
3.1.1 Fixed costs
The cost split up makes it clear that the percentage of the fixed costs decreases with an increased
number of cutting heads as seen in figure 1, 3 and 5. This is also easy to understand if the fixed
cost is split on more than one cutting head that this part will decrease. Notice that in the
calculations no extra costs were added for extra setup time, availability and material utilization
when more than one cutting head is used. In practice this would make this difference not as
significant as calculated. If it would be possible to use the machine more hours annually, this
would also in the end generate lower fixed costs per cut meter.
To be able to produce parts with reduced fixed costs per unit the following factors has to be
considered:

Multiple nozzle machine configuration

Increased annual machine hours (increase number of shifts, workload, availability etc.)

3.1.2 Running costs

Approximately 90% of the running costs is represented by the following cost factors as can be
seen in figure 2, 4 and 6:

Abrasives (~65%)
Consumables for the cutting head (orifice, focussing tube, valve etc.) (~15%)
Pump maintenance (~10 %)

Noticeable here are the costs for the machine with four cutting heads and pump pressure 380MPa
(#3, figure 6). In this case the lower pump maintenance cost derives from the lower pressure.
To be able to produce parts with reduced running costs per unit the following factors have to be
considered:

Optimized abrasive feeding

Optimized life length versus cost of cutting head consumables
Optimized pressure

4. PARAMETER ANALYSIS
In the analysis below every parameter is discussed separately, it is important to point out that
they interact, so that for example a higher pressure changes the saturation level for the abrasive
feeding.
4.1 Abrasive Feeding
The issue to minimize and optimize the cost for abrasives has been addressed previously by for
instance Henning (2004) and Ranney (1995) and the result from the cost split up also indicates
that this is a very important area while the abrasive costs represents approximately 60-70 % of
the running costs and 22-35% of the total costs.
To be able to optimize the abrasive feeding in a model that describes the total cost for an
abrasive waterjet machine, there has to be information in the model about the abrasive feeding
saturation curve for the specific cutting head and cutting conditions. It is an extensive work to
test every cutting head for every cutting condition to produce such saturations curves. Henning et
al (2004) presents a mathematical model such as it would be possible to describe this saturation
curve with a minimum of two experiments for a specific cutting condition.
The economic model used in this paper calculates cutting speeds with a modified version of a
model presented by Zeng and Kim (1993). This model always predicts a higher cutting speed for
an increased abrasive feeding.
To be able to optimize this part of the cost it may result in a lot of tests that have to be performed
to evaluate exactly which abrasive feeding is the most cost effective for a specific machine setup.

Figure 7 and 8 shows maximum cutting depth in stainless steel for two different nozzle
configurations 0,25/0,76 and 0,35/1,1. The result from the tests shows that the maximum cutting
depth in these cases is reached for approximately 400 and 625 g/min respectively. The cutting
depths are measured for new nozzles. In practice the maximal cutting depth will decrease as the
orifice and focussing tube is worn. Both the quality of the pure waterjet when it leaves the orifice
and the focussing tube and its condition affects the jet. The policy at every company how many
hours a nozzle should run between replacement is therefore also affecting the total economy.
In practice however every company has its own policy for abrasive feeding, and even if an
optimization model generates a specific abrasive feeding, it would not be very likely that a
company will change their machine settings between almost every job.
To get a more optimized cut for every specific case a solution like a programmable abrasive
metering unit would be interesting to use, but the problem is still that information about abrasive
saturation curves for every specific parameter combination has to be generated.
4.2 Pump Pressure
The pump pressure (P) is an important factor for calculation of the cutting speed. Zeng and Kim
(1993) suggests that the cutting speed is proportional to P1,25 in the interval 138 to 276 MPa. This
is however not the same as it is economically beneficial to run at a higher pressure. To be able to
investigate which pressure to run at for a specific case the following example is computed:
Increasing pressure from 350 MPa to 410 MPa would generate an increase in cutting speed with
approximately 27% according to performed tests. If one machine is running one shift (ca 1750
hours / year) with a workload of 70%, that will result in 1225 hours of machining. Increasing the
pressure from 350 to 410 MPa will free some 260 hours (or 46 days with a workload of 70%).
One should have in mind that a higher pressure level increases the flow rate.
This would also generate more downtime and a higher maintenance cost. According to
calculations made within the project, a pressure rise from 350 MPa to 410 MPa for a pump
would almost treble the cost for spare parts (for the first 10.000 pump hours), and the pressure
rise would still be profitable. So if a company has a relatively new pump and a lot of work to do,
they ought to consider running at a high pressure. This will then also generate more downtime
and a higher maintenance cost.
In the end the question still remains for every company using waterjet equipment if it is
profitable or not to turn up the water pressure. The answer has to do with a lot of aspects such as:

The waterjet equipment itself (available pressure, flow rate, type, age, etc.)
Downtime
Maintenance cost
Workload

4.3 Number of Cutting Heads

The option to use more than one cutting head with a smaller orifice and focussing tube diameter
is widely used today. The question then arises how many cutting heads should be used, and what
savings are possible. To be able to compare the costs for different number of cutting heads two
pump sizes are compared; one 37 kW pump and one 73,5 kW pump, the machine configurations
are listed in table 5. The costs are based on realistic investment costs for such pumps. The results
are presented in table 5 and figure 9 and 10. If the costs per cut meter are compared it is
noticeable how big the difference is between one and two cutting heads.
The basis for the calculation is one 37 kW pump equipped with one 0,35/1,1 mm nozzle. The
cost per cut meter with this type of machine is set to 1,00. Four combinations are compared and
it is obvious that it is profitable to split the fixed costs on more then one cutting head. If the
number of cutting heads is increased to two 0,25/0,76 mm on a 37 kW pump the cost per cut
meter is reduced to 0,83. The difference between one 0,35 /1,1 mm nozzle on a 37 kW pump
and four 0,25 /0,76 mm nozzles on a 73,5 kW pump is a 42 % reduction in costs per cut meter.
Notice that in the calculations no extra costs were added for extra setup time, availability and
waste of material.
Generally one should always consider running with at least two cutting heads. This would in
most cases be profitable. On the other hand companies should always have in mind that a
multiple nozzle configuration implies:

Increased programming and setup time

A lower availability on the machine
Generally more scrap material
Overall a decreased flexibility

5. CONCLUSIONS

If an abrasive waterjet cutting operation is to be optimized it is essential to focus on the costs

that generates the greatest part in a cost split up.
The fixed cost is often half or up to two thirds of the total costs for producing a 1 meter cut.
Using two cutting heads instead of one gives a significant drop in the total cost for producing
a 1 meter cut.
The pump pressure is interesting to study especially if it is important to free machine time.
Notice that the flow rate can be a limitation for maximum pump pressure.
Abrasive feeding is the major part of the running costs, and is relatively complex to optimize.
It takes an extensive testing to find saturation curves for every combination of parameters
that are to be used.

6. REFERENCES
Singh, P. J., Munoz, J.: Cost Optimization of Abrasive Waterjet Cutting Systems, proc. 7th
American Water Jet Conference, Seattle, Washington, USA, 1993.
Zeng, J., Kim, T.: Parameter Prediction and Cost Analysis in Abrasive Waterjet Cutting
Operations, proc. 7th American Water Jet Conference, Seattle, Washington, USA, 1993.
Zeng, J., Munoz, J.: Optimization of Abrasive Waterjet Cutting The Abrasive Issues
Technical Paper - Society of Manufacturing Engineers, MR94-247, Dearborn, MI, USA,
1994.
Ranney W. A.: Waterjet Abrasives: Evaluating Cost Analysis Procedures, proc. 8th American
Water Jet Conference, Houston, Texas, USA, 1995.
Brandt, S., Maros, Z., Monno, M.: AWJ Parameters Selection: a Technical and Economical
Evaluation proc. 15th International Conference on Water Jetting, Ronneby, Sweden, 2000.
Henning, A., Westkmper, E., Jarchau, M.: Analysis of Cutting Performance of High Power
Abrasive Water Jets proc. 17th International Conference on Water Jetting, Mainz,
Germany, 2004.

7. TABLES
Table 1. Parameters Used in the Economical Calculations
General Costs

Investment cutting table including tubing, software etc.

Investment pump
Economic life
Interest
Rental charges
Electricity
Water

Maintenance Costs

Pump
Cutting table

Consumables

Abrasives including handling and disposal

Cutting head (orifice, focussing tube, valves etc.)

Limitations

Max pump pressure

Max pump flow rate
Working hours per year
Availability

Machine settings

Nozzle configuration
Abrasive feeding
# of cutting heads

Target material

Type
Thickness
Quality of cut

Table 2. Example of input to the optimization model

Investment

272 000

Abrasives, total cost

0,43 /kg

Operator

32,6 /h

Working hours per year

3750 h

Max pump power

73,5 kW

Max pump pressure

400 MPa

Available cutting heads

Utilization

3
70 %

Table 3. Output from optimization model, for cutting of 20 mm steel in a medium quality
Shortest
cutting time

Lowest cost

Pressure

360 MPa

400 MPa

Abrasive feed rate

500 g/min

400 g/min

Orifice diameter

0,304 mm

0,356 mm

Nozzle diameter

0,9 mm

1,0 mm

Number of cutting heads

53 mm/min

Cutting speed

68 mm/min

Table 4. Machine Settings used for Cost Split Up

#1
50 / 37
410

#2
100 / 73,5
410

#3
100 / 73,5
380

0,35 / 1,1
625

0,35 / 1,1
625

0,25 / 0,76
400

# cutting heads

Cutting speed 18 mm
stainless steel (mm/min)
Total cost /h
Total cost (/m)

88

176 (88*2)

200 (50*4)

96,1
18,1

131,2
12,4

141,4
11,7

Pump power (hp / kW)

Max pressure (MPa)
dorifice / dfocussing tube (mm)
Abrasive feeding (g/min)

Table 5. Multiple nozzle configurations

Pump power (hp / kW)

Pressure (MPa)
dorifice / dfocussing tube(mm)
Abrasive feeding (g/min)
# cutting heads
Normalized cost per meter
Normalized cost per meter

#4
50 / 37
380

#5
50 / 37
380

#6
100 / 73,5
380

#7
100 / 73,5
380

0,35 / 1,1
625

0,25 / 0,76
400

0,35 / 1,1
625

0,25 / 0,76
400

1,00
1,47

0,83
1,22

0,68
1,00

0,58
0,85

8. GRAPHICS

Running costs
36%

Fixed costs
64%

Figure 1. Cost Split Up for configuration #1, Total costs

Maintenance table
7%

Electricity
5%
Water
2%

Maintenance pump
12%

Cutting head
(Orifice, Tube, Valve
etc.)
13%

Abrasives
61%

Figure 2. Cost Split Up for configuration #1, Running costs

Running costs
47%
Fixed costs
53%

Figure 3. Cost Split Up for configuration #2, Total costs

Maintenance table
4%

Electricity
5%
Water
2%

Maintenance pump
12%

Cutting head
(Orifice, Tube, Valve
etc.)
14%

Abrasives
63%

Figure 4. Cost Split Up for configuration #2, Running costs

Fixed costs
49%

Running costs
51%

Figure 5. Cost Split Up for configuration #3, Total costs

Maintenance table
3%
Maintenance pump
7%

Electricity
4%
Water
1%

Cutting head
(Orifice, Tube, Valve
etc.)
17%

Abrasives
68%

Figure 6. Cost Split Up for configuration #3, Running costs

20

18

16

Cutting depth (mm)

14

12

10

Pressure: 350 MPa

dorifice: 0.25 mm
dfocussing tube: 0,76 mm
Cutting speed: 140 mm/min
Stand-off distance: 3 mm
Abrasive: Aluvial garnet #80

0
0

100

200

300

400

500

600

Abrasive feeding (g/min)

Figure 7. Maximum cutting depth in stainless steel for a 0,25/0,76 nozzle at 350 MPa
18

16

14

Cutting depth (mm)

12

10

Pressure: 350 MPa

dorifice: 0.35 mm
dfocussing tube: 1,1 mm
Cutting speed: 225 mm/min
Stand-off distance: 3 mm
Abrasive: Aluvial garnet #80

0
0

100

200

300

400

500

600

700

800

Abrasive feeding (g/min)

Figure 8. Maximum cutting depth in stainless steel for a 0,35/1,1 nozzle at 350 MPa

#5

#4

Figure 9. Standardized costs per cut meter for machine setup #4 and #5, compared with three
different nozzle configurations and with increasing abrasive feeding for each nozzle
configuration.

#7

#6

Figure 10. Standardized costs per cut meter for machine setup #6 and #7, compared with three
different nozzle configurations and with increasing abrasive feeding for each nozzle
configuration.