See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/225177690

Design of Experiment Based Analysis for Sheet Metal Blanking Processes

Optimisation

Article  in  The International Journal of Advanced Manufacturing Technology · April 2002

DOI: 10.1007/s001700200041

CITATIONS READS
47 4,677

1 author:

Ridha Hambli
Université d'Orléans
168 PUBLICATIONS   2,694 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

material parameters identification using new coupled methods View project

process optimization View project

All content following this page was uploaded by Ridha Hambli on 27 January 2015.

The user has requested enhancement of the downloaded file.

Int J Adv Manuf Technol (2002) 19:403–410
Ownership and Copyright
 Springer-Verlag London Ltd 2002

Design of Experiment Based Analysis for Sheet Metal Blanking

Processes Optimisation
R. Hambli
ISTIA – LASQUO, Angers, France

The blanking of metal parts for electronic components is 1. To avoid, in certain cases, additional operations such as
subjected to a variety of process parameters. In this paper, an removal of burrs to improve the geometrical quality of the
experimental investigation into the blanking process was sheared edge [5].
carried out using tools with four different wear states and four 2. To increase the fatigue life of the parts in service as shown
different clearances. The aim was to study the effects of the by Lambert et al. [7].
interaction between the clearance, the wear state of the tool 3. To increase the fatigue life of the tool [8,9].
and the sheet metal thickness on the evolution of the blanking
force and the geometry of the sheared profile. The blanking of thin metal parts is subjected to a variety
Designed experiments are an efficient and cost-effective way of process parameters. Various material coefficients and process
to model and analyse the relationships that describe process factors affect the quality of the blanked part.
variations. The main objectives of the process design in metal blanking
The results of the proposed experimental investigation show are to choose the leading process parameters in an optimal
the strong dependence between the geometrical quality of the way to ensure high-quality parts. Engineers are often required
blanked part and the magnitude of the force applied on the to examine the process parameters in order to optimise
tool as well as the variations in the process factors. production.
Inspection of the most recent studies in the field of manufac-
Keywords: Blanking; Clearance; Design of experiments; Sheet turing processes [10,11] shows that, despite the increasing
thickness; Tool wear progress on blanking analysis, there is still a lack of models
for the optimal design of sheet metal shearing processes.
Currently, correct parameter choice for a new product manufac-
tured by sheet metal blanking is determined empirically by
1. Introduction performing a large number of expensive tests.
The clearance, the tool wear state and the sheet thickness
The sheet metal cutting process for thin components using are the major factors that determine the shape and the quality
both a punch and a die (Fig. 3) is one of the most frequently of the workpiece [5,10,12]. Blanking has a large number of
used processes in industry [1]. Depending on the position of inputs. Each of these inputs has an associated variation that
the sheared surface with respect to the workpiece coordinates, leads to variations in the final part [5,13,14].
various shearing processes are used such as blanking, piercing, Owing to the high number of inputs, their associated vari-
and cutting off [1–3]. In contrast to other operations, such as ations, and the relationship between stress and strain in metal
stamping and bending where the aim is to deform the sheet forming, the blanking system behaves in a manner similar to
plastically, these operations lead to the total rupture of the a deterministic chaotic system. Understanding this system fully
metal [4,5]. As the material is processed, its behaviour is and controlling the output by adjusting the inputs is difficult
nonlinear and it is submitted to complex strain and stress states if not intractable at present.
[1,6]. Before complete rupture, the material is subjected to A more effective method of controlling the output is to
some damage and crack propagation phenomena caused by the understand the input variation more thoroughly and to account
punch penetration. for this in defining the operating point that ensures robustness
Various experimental studies carried out covering this subject with respect to this identified variation.
have shown that optimal parameter choice is crucial: An operation window in sheet metal forming is an area in
the input space that corresponds to the production of good
Correspondence and offprint requests to: Dr R. Hambli, ISTIA-LAS- parts. The size of the operating window corresponds to the
QUO, 62 Ave Notre Dame du Lac, 49000 Angers, France. E-mail: sensitivity of the part quality to variation of the input
rida.hambli얀istia.univ-angers.fr parameters. With an understanding of the process and associa-
404 R. Hambli

ted variations, the most robust operating point within this Dm − Dp

window can be identified. c = 100 (%) (1)
2t
In production, fully identifying the variation in a part pro-
duced can be very difficult to achieve and often a large number Dm, Dp, and t are, respectively, the die diameter, the punch
of experiments must be performed. This increases both the diameter, and the sheet thickness.
costs and the associated lead times. Therefore, there is a need In order to study the influence of this design parameter,
to identify this operating variation and the associated robust four tools have been designed corresponding to four different
operating point in the shortest time at a low cost. clearances, 5%, 10%, 15%, and 20%. These values correspond
Designed experiments are a systematic approach to optimis- to the most used clearances in industry.
ing process performance, they are also used for knowledge
acquisition. Traditionally, the setting of a single factor is 2.2 Influence of the Tool Wear
changed at a time until the response is improved. This one-
factor-at-a-time approach is inefficient because no measures of The design of the tool is one of the main features in the
interactions between the factors are available, and because the industrial process. Therefore it is necessary to study the effects
accuracy of the effect estimates is usually poor. of tool wear on the blanking force and the sheared profile
In this case, a multiresponse design of experiments has variations. The quality of the workpiece is governed by the
been used to identify the best compromise between conflicting state of the tool wear [5,8].
requirements. The blanking force is a cost-related response, Wear is defined as a slow degradation of the blanking tool
whereas fracture angle and fracture depth are quality-related caused by the friction involved between tool and sheet metal.
responses. Designs of experiments may also be used to improve The rate of wear is affected by parameters such as tool
the quality of a product by minimising the effect of the causes material, blanked part material, punch–die clearance, punch
of variation without eliminating the causes. The robust design velocity, lubrication, and material thickness. Generally, wear
approach is valuable to ensure that decisions that are found to takes place on the external surface of the tool. It causes the
be optimum during laboratory experiments will prove to be cutting edges to be rounded (Fig. 4). Therefore, the influence
optimum in manufacturing and in customer environments. The of the tool wear can be accounted for by changing the values
major tools used in robust design are designs of experiments of the edge radii Rwp and Rwd (Fig. 4).
in which controllable factors as well as the factors that are Experimental investigation into the blanking process was
difficult to control during production, are included. carried out using punches with different wear states. The aim
was to define the relationship between the sheared profile of
the component and the forces applied to the tool evolutions
2. Design of Experiments versus the tool wear evolution.
Four wear states of the tool were chosen corresponding to:
The accuracy of workpieces can be characterised by the follow-
ing errors: dimensional error, positional error, and form or A new die with: Rwd = 0.01 mm.
shape error (Fig. 1). A new punch with: Rwp = 0.01 mm.
The errors on blanks are influenced by material, tool shape, Three “worn out” punches with different edge radii Rwp = {0.06,
process variations and machine. 0.12, 0.2} mm.
The shape errors represented in Fig. 2 are connected to the
geometry of the sheared edge (roll-over depth, fracture depth,
2.3 Influence of the Sheet Thickness
smooth-sheared depth, burr formation and fracture angle).
Various experimental studies [3–6,8,12] have shown that for
For a given material, the energy requirement in blanking is
a given material, the blanking force and the characteristic depth
influenced by the sheet thickness. It has been observed that:
of the blanked profile (Fig. 2) are affected by the blanking
clearance, the wear state of the tool, and the thickness of 1. The blanking energy decreases with increasing clearance-
the sheet. to-sheet thickness ratio c/t and increases with increasing
sheet thickness [1,5].
2.1 Influence of Blanking Clearance 2. The proportions of the different depth characteristics of the
sheared profile are affected by the thickness [3–5].
In blanking processes, the clearance (Fig. 3) expressed as a To study the effects of the interactions between the clearance,
percentage of the sheet thickness, is defined by: the wear state of the tool, and the sheet thickness, a series of
experiments have been carried out with two thickness values
of 1.5 and 3 mm.

2.4 Design Factors

The experiment was designed to study the following variables

and the interactions between them: clearance, tool wear state,
Fig. 1. Factors affecting errors on blanked workpieces. and sheet thickness.
 Sheet Metal Blanking Processes Optimisation 405

Fig. 2. Geometry of the sheared blanked part.

Table 1. Experimental design.

Experimental Variables
point
Clearance (%) Wear radius Thickness (mm)
(mm)

1 5 0.01 3
2 5 0.01 1.5
3 5 0.06 3
4 5 0.06 1.5
5 5 0.12 3
6 5 0.12 1.5
Fig. 3. Illustration of the punch and die clearance. 7 5 0.2 3
8 5 0.2 1.5
9 10 0.01 3
10 10 0.01 1.5
11 10 0.06 3
12 10 0.06 1.5
13 10 0.12 3
14 10 0.12 1.5
15 10 0.2 3
16 10 0.2 1.5
17 15 0.01 3
18 15 0.01 1.5
19 15 0.06 3
20 15 0.06 1.5
21 15 0.12 3
Fig. 4. Wear profile of the cutting edges of the tool.
22 15 0.12 1.5
23 15 0.2 3
24 15 0.2 1.5
Thirty-two experiments were performed according to Table 1. 25 20 0.01 3
26 20 0.01 1.5
Four levels enable us to build complex cubic models. 27 20 0.06 3
28 20 0.06 1.5
29 20 0.12 3
30 20 0.12 1.5
3. Experimental Procedure 31 20 0.2 3
32 20 0.2 1.5
Experiments using devices equipped with electrical gauges and
a force transducer were performed by a 4000 kN hydraulic
press. A diagram of the experimental equipment is given 4. Results
in Fig. 5.
The problem studied here consists of an axisymetric blanking The results represented in Table 2 show the effect of each
operation of metal sheets of 1.5 and 3 mm thickness. The variable and the interaction between them on the blanking
geometrical data are shown in Fig. 6. force, the fracture angle, and the fracture zone depth.
Carbon steel (0.6% C) was blanked using a 40 mm diameter In this paper, a design of experiments is used for modelling
die and a range of punch–die clearances of 5%–20%. Punches and analysing the response of interest that is influenced by
and dies were of edge radii in the range of 0.01–0.2 mm. several variables.
406 R. Hambli

Table 2. Design points effects.

Experimental Variables
point
Clearance Wear Thickness Blanking Fracture Fracture
c (%) radius t (mm) force F angle ␪ depth
Rwp (kN) (deg.) Har (%)
(mm)

1 5 0.01 3 234 3.898 73.5

2 5 0.01 1.5 98 3.94 72.7
3 5 0.06 3 235 4.12 69.5
Fig. 5. Diagram of the testing device. 4 5 0.06 1.5 98.5 4.421 64.8
5 5 0.12 3 240 4.421 64.8
6 5 0.12 1.5 100.5 5.180 55.3
7 5 0.02 3 245 4.899 58.5
8 5 0.2 1.5 102.5 6.715 42.7
9 10 0.01 3 231 7.737 74.1
10 10 0.01 1.5 97 7.761 73.8
11 10 0.06 3 233 7.858 72.9
12 10 0.06 1.5 97.5 8.009 71.5
13 10 0.12 3 235 8.009 71.5
14 10 0.12 1.5 98.5 8.328 68.8
15 10 0.2 3 240 8.219 69.7
16 10 0.2 1.5 100.5 8.796 65.1
17 15 0.01 3 233 11.576 74.2
18 15 0.01 1.5 97.5 11.583 74.2
19 15 0.06 3 235 11.609 74
20 15 0.06 1.5 98.5 11.648 73.8
21 15 0.12 3 240 11.648 73.8
22 15 0.12 1.5 100 11.726 73.3
23 15 0.2 3 345 11.7 73.5
24 15 0.2 1.5 102.5 11.833 72.6
Fig. 6. Axisymetric model of the blanking operation.
25 20 0.01 3 235 15.416 74.3
26 20 0.01 1.5 98.5 15.405 74.4
We must address three distinct issues: 27 20 0.06 3 235 15.362 74.6
28 20 0.06 1.5 98.5 15.297 74.9
1. Optimising these three responses, or at least finding the best 29 20 0.12 3 242 15.297 74.9
30 20 0.12 1.5 101.5 15.170 75.5
compromise according to preference. 31 20 0.2 3 247 15.212 75.3
2. Finding a clearance setting that is robust with respect to 32 20 0.2 1.5 103.5 15.003 76.4
wear and to variations in sheet thickness, considering that,
during production, wear and thickness are difficult-to-control
noise factors.
sis of the clearance effect and the tool wear effect on fracture
3. Identifying a model that is valid even when the sheet
depth, fracture angle, and maximum blanking force.
thickness is modified, considering that sheet thickness is a
For each sheet metal thickness, the model ⌽ describing the
“block” factor.
variable effects x1 and x2 across their experimental ranges can
The interactions between controllable factors (clearance) and be expressed by:
noise factors (wear and thickness) are useful in reducing the
y = ⌽ (x1, x2) + ⑀ (2)
influence of the noise factors and thereby making the process
more robust against variations in tool wear and sheet thickness. Where ⑀ represents the experimental errors, ⌽(x1x2) represents
However, if the objective is to find a model that is valid even the hypothetical value of the response y that would be obtained
when the sheet thickness is changed, then it is preferable not in the absence of experimental error, and x1 and x2 are design
to have interactions between thickness and the other factors. variables that correspond to the wear radius of the cutting
It is often assumed that block factors (such as thickness) do edge of the tool and the clearance.
not interact with the other factors; otherwise, the model is The wear radius Rwp of the tool (Fig. 4) and the clearance
modified according to the block factor setting. The main effect composing the two axes of the plane will be taken as measures
of the thickness is not important in itself, what is important of the degrees of the constraint which are the fracture depth,
is to identify the optimal clearance setting and the tool wear the fracture angle, and the maximum blanking force under
states that are not too detrimental to the process, whatever the various conditions.
sheet thickness. Figures 7(a) and 7(b) are response surfaces graphs showing
Assuming that there is, most of all, a need for models that the effects of both tool wear and clearance on the blanking
are robust to differences in sheet thickness (sheet thickness is force variation for thicknesses 3 and 1.5 mm. The figure clearly
considered to be a block factor) then, based on the experimental shows that for small clearances of less than 10%, the blanking
results, three models may be elaborated allowing for the analy- force is quite high and the minimal force value corresponds
 Sheet Metal Blanking Processes Optimisation 407

Fig. 7. Blanking force evolution vs. clearance and tool wear variation. (a) Thickness = 3 mm, (b) 1.5 mm.

to a clearance of 10%. It can be shown that, starting from sensitive to variation in tool wear and clearance for a high
this optimal value, that increasing the clearance increases the sheet metal thickness.
force variation.
By comparing Figs 7(a) and 7(b), we can see the interaction
4.1 Analysis of Variance
effects of decreasing the sheet thickness by .. It is important
to note that the response surfaces have the same shape and
When mixed-level designs (in which not all factors have the
that the average force of blanking is reduced by 2.4.
same number of levels) are performed, the analysis of variance
It can be observed that the blanking force increases with
(ANOVA) method cannot be used. This is because the variance
increasing wear as a result of increased punch penetration until
of the factor effect estimates differ according to the number
the final rupture of the sheet.
of levels. The smaller the number of factor levels, the greater
The variation of the evolution of the fracture angle is
the accuracy of the estimates of the factor effect. Using the
represented in Fig. 8. It may be noted that increasing clearance
ANOVA method to discriminate between statistically significant
increases significantly the fracture angle ␤ as a consequence
effects and effects that are not significant becomes quite com-
of the increasing punch penetration corresponding to crack
plex. Since the ANOVA method cannot be used, identifying
formation into the sheet. Both figures show that the tool wear
the “real” effects from those that are due only to experimental
radius has little influence on the variation of ␤.
errors may be somewhat subjective.
To represent the relationship F(c,Rwp) between the blanking
force variation and the factors (clearance c and wear radius
Rwp) a polynomial model may be used with a Taylor series 4.2 Factor Effects on the Blanking Force
expansion [15].
The comparison between the two curves shows that there is The average blanking force when all the tests are considered
no influence of the sheet thickness on the fracture angle evol- is 168.70 kN. Table 3 indicates the differences between the
ution. general mean and the mean at each factor level. To minimise
The 3D response surface graph of Fig. 9, presents the interac- the blanking force, the clearance should be set at 10% so that
tion effects on the fracture zone depth variation. the average blanking force is decreased by 2.14 kN (see Table 3
The proportion of the fractured area increases with increasing and Fig. 7). Sheet thickness and tool wear can be considered
clearance and decreases with increasing wear. The flatter to be noise factors, since they are difficult to control during
response surface (Fig. 9(a)) means that fracture depth is less normal production. Tools wear out and different types of sheet

Fig. 8. Fracture angle evolution vs. clearance and tool wear variation. (a) Thickness = 3 mm, (b) 1.5 mm.
408 R. Hambli

Fig. 9. Fracture depth evolution vs. clearance and wear tool wear variation. (a) Thickness = 3 mm, (b) 1.5 mm.

Table 3. Main effects blanking force response.

Table 4. Clearance by wear radius interaction.
Levels Factors
Interactions
Clearance Wear radius Rwp Thickness Clearance level Water radius level Clear * Wear

1 0.484 −3.569 −69.109 1 1 0.0157

2 −2.14 −3.027 69.109 2 −0.109
3 0.234 2.133 3 0.078
4 1.42 4.325 4 0.016
2 1 0.641
2 1.016
3 −0.797
must be processed. It can be observed that the blanking force 4 −0.859
increases with increasing wear as a result of increased punch 3 1 −0.484
2 0.141
penetration until the final rupture of the sheet. That increase 3 0.078
is more noticeable for wear radius values that are between 4 0.266
0.06 mm and 0.12 mm (see Table 3). Suppose that the clearance 4 1 −0.172
is set at 10% to minimise the amount of blanking force, as 2 −1.047
the tools wears out, more force would be required. However, 3 0.641
4 0.578
we also need to consider the interactions between clearance
and tool wear, since the clearance and tool wear effects cannot
be strictly added. When the clearance is set at 10%, the tool
wear effect is slightly reduced, at lower wear radius values Table 5. Clearance by thickness and wear radius by thickness interac-
the amount of blanking force is slightly greater than when the tions.
clearance and tool wear effects are added to one another;
whereas at larger tool wear values, the amount of blanking Clearance level Thick*clear Wear radius Thick*wear
force is slightly lower than anticipated when the interaction
effects are not accounted for (see Table 4). In other words, 1 0.203 1 −1.359
2 −0.922 2 −0.984
when the clearance is set at 10%, the process is more robust 3 0.203 3 0.453
to variations in the degree of tool wear. When the thickness 4 0.516 4 1.891
is increased from 1.5 to 3 mm, a much greater blanking force
is required (+69.1 kN). Owing to the clearance caused by
thickness interactions, the effect of thickness is slightly smaller
when the clearance is set at 10% and due to tool wear caused
by thickness interactions, the influence of thickness is slightly Table 6. Main effects fracture angle response.
reduced at lower values of tool wear (see Table 5).
A clearance of 10% does not only allow for a minimisation Levels Factors
in the blanking force, it is also more robust against both the
tool wear and the sheet thickness. Clearance Wear radius Thickness

1 −5.232 −0.297 0.120

4.3 Factor Effects on the Fracture Angle 2 −1.842 −0.183 −0.120
3 1.734 0.089
The clearance effect on the fracture angle evolution is very 4 5.339 0.348
significant and almost linear (see Table 6 and Fig. 8). Increasing
 Sheet Metal Blanking Processes Optimisation 409

Table 7. Clearance by wear radius interaction. Table 9. Main effects fracture depth response.

Interactions Levels Factors

Clearance level Wear radius level Clear*Wear
Clearance Wear radius Rwp Thickness
1 1 −0.514
2 −0.288 1 −5.559 6.299 −3.578
3 0.06 2 2.641 4.879 3.578
4 0.742 3 5.453 3.176
2 1 −0.074 4 −2.534 −10.342
2 −0.016
3 0.038
4 0.052
3 1 0.181 Table 10. Clearance by wear radius interaction.
2 0.104
3 −0.020 Interactions
4 −0.265 Clearance level Water radius level Clear * Wear
4 1 0.407
2 0.120
3 −0.078 1 1 4.722
4 −0.529 2 0.672
3 −4.140
4 −1.253
2 1 −2.628
2 −2.478
punch penetration causes crack formation into the sheet. The 3 −2.241
fracture angle evolution is also increased at higher levels of 4 7.347
tool wear but that effect is barely noticeable when compared 3 1 −5.191
to the clearance effect. The fracture angle is slightly smaller 2 −3.591
when the thickness increases, again this is a very minor effect 3 −1.803
4 10.584
compared to the clearance effect. 4 1 3.097
To study the clearance by wear radius interaction, no less 2 5.397
than sixteen levels (and nine degrees of freedom) are required. 3 8.184
Note that the degree of precision on the factor effects decreases 4 −16.678
when the number of levels increases. Because the interaction
effects are so small (see Tables 7 and 8) compared to the
clearance effect and their degree of precision is not high, it Table 11. Clearance by thickness and wear radius by thickness
interactions.
was decided not to study these interactions.
Clearance level Thick*clear Wear radius Thick*wear
4.4 Factor Effects on the Fracture Depth
1 0.272 1 −3.466
The average fracture depth is 68.28% when all tests are 2 −2.453 2 −2.791
3 −3.791 3 −2.053
considered. The clearance effect on fracture depth is nonlinear 4 5.972 4 8.309
(see Table 9 and Fig. 10), the response reaches a maximum
when the clearance is set at 15% (+5.45% over the general
mean) (Tables 10 and 11). The wear radius also has a strongly
For thicker 3 mm sheets, the amount of fracture depth is
nonlinear effect, as it increases, the fracture depth decreases,
increased (+3.57%). Owing to thickness by clearance, and
this decrease is rather slow between 0.01 mm and 0.12 mm
thickness by wear radius interactions, for thicker 3 mm sheets,
wear radius values, however, for a wear radius of 0.2 mm, the
the clearance and wear radius effects are reduced, however for
response drops dramatically (−10.34% below the general mean).
thinner 1.5 mm sheets, the clearance and wear radius effects
When the clearance is set at 10% or 15% the wear radius
are reinforced.
effect is considerably reduced, whereas for a 20% clearance,
the wear radius effect is greatly reinforced.
5. Conclusion
Table 8. Clearance by thickness and wear radius by thickness
interactions. The developed experimental investigation into the sheet metal
blanking process makes it possible to study the effects of the
Clearance level Thick*clear Wear radius Thick*wear
interaction between the wear state of the tool, the punch–die
clearance value and the thickness of the sheet on the variation
1 −0.245 1 0.112
2 −0.014 2 0.067 of the blanking force and the geometry of the sheared edge.
3 0.088 3 −0.009 The design of experiments method used in this paper pro-
4 0.171 4 −0.170 vides a better understanding of the blanking manufacturing
response.
410 R. Hambli

The process signatures indicate that the maximum shearing 3. S. Kalpakjian, Manufacturing Processes for Engineering Materials,
force, the fracture angle and the fractured surface depth are 2nd edn, Addison-Wesley, 1991.
4. T. M. Chang, “Shearing of metal blank”, Journal of the Institute
influenced by the material condition as well as the geometric of Metallurgists, pp. 393–414, 1951.
characteristics of the tools and their configurations. 5. R. Hambli, “Etude expérimentale, numbérique et théorique du
The analysis of the tool wear influence allows for the découpage des tôles en vue de l’optimisation due procéde”, Thèse
monitoring of the blanking operation and so the parts quality de Doctorat, ENSAM d’Angers, 15 October 1996.
6. A. Maillard, “Etude expérimentale et théorique du découpage”,
variations during the forming process may be predicted. Thèse de Doctorat, Université Technologique de Compiègne, 1991.
This investigation shows that, in order to minimise the 7. Y. Lambert, A. Bignonnet and L. Roesch, “Prévision du comporte-
blanking force, the clearance should be set at 10%, however, ment en fatigue de structures minces en aciers à haute résistance
to minimise the fracture angle and the fracture depth, it is dual-phase et microalliés”, Mémoires et Etudes Scientifiques Revue
preferable to set the clearance at 5%. When the clearance is de Métallurgie, pp. 209–225, April 1991.
8. E. Grosset, A. Maillard and A. Turbat, “Usure en découpage”,
set at 10%, the process is slightly more robust to tool wear, CETIM, Rapport d’étude 101580, Novembre 1987.
as far as the blanking force response is concerned, and it is 9. E. Grosset, P. Peyre, P. Cherry, J. Gasnier and C. Tournier, “Les
considerably more robust (almost insensitive) to tool wear and dépôts PVD et CVD en poinçonnage et relevage de collerette”,
sheet thickness as far as the fracture depth response is con- CETIM, Informations, 111, pp. 55–61, 1989.
10. Proceeding of the 7th International Conference on Sheet Metal
cerned. Whether clearance should be set at 5% or 10% ulti- (SheMet’99), Earlengen, Germany, 25–28 September 1999.
mately depends on the priorities of the practitioners. 11. Proceeding of the 8th International Conference on Sheet Metal
As a conclusion drawn from the proposed investigation, it (SheMet’2000), Birmingham, UK, 17–18 April 2000.
is possible to optimise the sheet metal blanking process by a 12. S. Jana and N. S. Ong, “Effect of punch clearance in the high-
proper selection of the clearance. speed blanking of the thick metal using an accelerator for mechan-
ical press”, Journal of Mechanical Working Technology, 19,
pp. 55–72, 1989.
13. Y. Kasuga, S. Tsu Tsumi and T. Mori, “Investigation into shearing
References process of ductile sheet metals”, Memoir of the Faculty of Engin-
eering, Nagoya University, Japan, pp. 1–46, 1979.
1. K. Lange, Handbook of Metal Forming, McGraw–Hill, 1985. 14. T. Osaki and T. Yoshikai, “Effect of profile in basal section in
2. T. Osaki and S. S. Yamasaki “Effect of the geometry of cutting shearing process”, Memoir of the Faculty of Engineering, Kyushu
edge of tool in blanking process and punching”, Memoir of University, 38(3), pp. 249–273, 1978.
the Faculty of Engineering, Kyushu University, 38(4), pp. 371– 15. John A. Cornell, “How to apply response surface methodology”,
395, 1984. vol. 8, American Society for Quality, 1990.

View publication stats