7

1. Introduction

Micro formed parts are often components of micro system technologies (MST) or micro electro-mechanical systems (MEMS), especially read and write heads, inkjet printers, pressure sensors and micro fluidic chips. They contain leverages, connector pins, resistor caps, contact springs and chip lead frames [2]. The estimated rise in turnover from 10 to 19 billion US $ from 2004 until 2009 [1] shows a growing demand on micro formed parts, which is mainly driven by a rising trend of miniaturization, see Fig. 1. Thus, investigation and improvement of conventional micro forming processes are needed, but also the invention of new processes is desirable.

Fig. 1. Estimated market for selected MST and MEMS products with micro formed components [1]

2. Mechanical micro deep drawing

The first approach to realize micro deep drawn parts is usually suggested by downscaling the parameters of mechanical macro deep drawing processes according to the theory of similarity [3]. But this procedure can cause unexpected results, since size effects are documented for different parameters: The flow curve, the deviation of the flow curve and the friction change along with miniaturization [4]. Own investigation showed that the friction in strip drawing increases significantly, if the process is miniaturized [5].

2.1 Method In former investigations strip drawing was used to

determine size effects of friction in sheet metal forming, see Fig. 2a). Therefore a friction function f(µ) could be calculated from the process parameters and the punch force/punch travel-curve. The strip drawing was a simplification of the deep drawing process, since the tangential force Ft was excluded, see Fig. 2b).

Now, a test setup is installed for micro deep drawing including a force measurement system with accuracy of 0.01 N and a position measurement system with accuracy of 1 µm. We thus get the punch force/punch travel-curve, which can in future work be used for the calculation for friction functions for the micro deep drawing process. But size effects can already be observed

SHEET METAL MICRO FORMING

F. Vollertsen, H. Schulze Niehoff, H. Wielage, Z. Hu

Bremer Institut für angewandte Strahltechnik - BIAS GmbH

AbstractMechanical micro deep drawing becomes a more and more industrial relevant process. But due to size effects new challenges

are involved in this process compared to macro deep drawing. The size effects cause an increase of friction and thus hinder the material flow. The change of friction in mechanical micro deep drawing is subject of the presented investigations in this paper. Additionally to this, a new non-mechanical micro deep drawing process is presented, whereby a laser beam acts as a punch. This new laser deep drawing process is based on a totally different mechanism compared to thermal laser forming, e.g. forming by laser induced thermal stresses: The laser produces a pulse with an extremely high power density, which causes plasma generation at the target and thus a shock wave. The shock wave can be used as in explosive forming, but is smaller and easier to generate. Recent investigations showed that using this technology laser deep drawing is possible with a sheet metal out of Al 99.5 and a thickness of 50 µm. The deep drawing process was carried out with a die diameter of 4 mm and shows promising results.

Keywords: micro forming, laser forming, deep drawing, tribology, shock wave

SHEET METAL MICRO FORMING

F. Vollertsen, H. Schulze Niehoff, H. Wielage, Z. Hu

8

by the punch force/punch travel-curve, if the punch force is normalized by the punch diameter D, the sheet thickness s0 and the flow stress kf.

Fig. 2. a) Strip drawing and b) deep drawing in process

Storoschew [6] showed, that the maximum punch force Fmax in deep drawing can be evaluated by:

(1)

whereby

(2)

D: blank diameter, rZ: drawing radius and µ: friction coefficient.

Eq. 1 and Eq. 2 give thus

(3)

In all experiments D/d was held constant. The same applies for s0/(2rz+s0). It can thus be written that

,

(4)

whereby C1 and C2 are constants. This shows that differences of the maximum punch force are only due to friction and flow stress, if it is normalized. Within this work the value µ in equ. (4) is called effective average coefficient of friction, as it is an average of the friction

coefficient acting at different areas of the workpiece-tool interface under different contact pressures. These different contact pressures were accounted for by the strip drawing test procedure described elsewhere [5]. Further work will be done to account for that also in deep drawing.

2.2 Experimental investigations

The experimental investigations in mechanical micro deep drawing were carried out in order to enhance the investigations in strip drawing to deep drawing. The results show, that the deviation of measured parameters is comparable high in micro deep drawing. Fig. 3 shows 6 punch force/punch travel-curves for the same micro cup deep drawing process with a punch diameter of 1 mm. It can be seen that the maximum punch force ranges from 1.7 to 2.7 N, which corresponds to a deviation of 37 %. Thus, the 6 curves are averaged to one curve for further investigations.

Fig. 3. Deviation of the punch force in mechanical deep drawing with 1 mm punch diameter

In contradiction to this, the punch force/punch travel-curves of the cup deep drawing process with a punch diameter of 5 mm show a better match, see Fig. 4. The maximum of the 6 curves ranges from 50 to 52 N, which corresponds to a deviation of only 3.8 %. This shows that an increase of factor 10 in deviation by miniaturization of the deep drawing process.

SHEET METAL MICRO FORMING

F. Vollertsen, H. Schulze Niehoff, H. Wielage, Z. Hu

9

Fig. 4. Deviation of the punch force in mechanical deep drawing with 5 mm punch diameter

Fig. 5. Size effects in mechanical deep drawing

If the punch force is normalized by the diameter of the punch, the sheet thickness and the flow stress (FP/d/s0/kf) and if the punch travel is normalized by the punch diameter (sP/d), then differences in friction can be observed, see Fig. 5. The normalized punch force of the micro deep drawing process with 1 mm punch diameter is 30-50 % higher than for the process with 5 mm punch diameter, which means that the friction increases with miniaturisation. The proportion of the micro friction coefficient to the macro fiction coefficient µmicro/µmacro can be assessed by Eq. 4 to more than 2, which corresponds

to the result from strip drawing. It can also be seen that the punch force decreases, if the lubrication decreases from 8 to 4 g/m² with a punch diameter of 5 mm. This effect cannot be detected for the micro deep drawing process with a punch diameter of 1 mm, where the two curves are inconsistent and support the assumption that the amount of lubrication in micro sheet metal forming does not have the impact as in macro forming as it was observed in former investigations in strip drawing. The ratio of adhesion force to friction force within deep drawing is about 0.001.

A summary of the size dependent friction coefficients in cup deep drawing derived by Eq. 4 are given in Fig. 6. They are in average 0.129 for punch diameter of 1 mm, 0.075 for punch diameter of 5 mm and 0.081 for punch diameter of 10 mm. These friction coefficients show that if the process dimension decreases, the friction increases.

Fig. 6. Size dependent friction coefficient in mechanical deep drawing

3. Laser micro deep drawing

The process laser forming is actually known as a process, where different thermal mechanisms cause a bending of the sheet metal [7]. The most common known thermal mechanism is the temperature gradient mechanism, where inhomogeneous strains are applied within the material, which results in an incremental forming process. In contrast to this process, laser forming can also be obtained by a non-thermal mechanism, which uses the optical low-threshold surface breakdown [8, 9], which results in creation of a shock wave. Its principle

SHEET METAL MICRO FORMING

F. Vollertsen, H. Schulze Niehoff, H. Wielage, Z. Hu

10

was first shown by O’Keefe [10]. He used a Q-switch Nd:YAG-Laser with a maximum power density of 1.7 GW/cm². The experiments showed thermal damaging and mechanical deformation. First own investigations on sheet metal forming processes were presented with Al 99.5 sheets of 50 µm thickness, which were laser stretch-formed to a hemispherical dome with diameters between 1 and 12 mm. It was figured out that CO2-lasers are due to their 40 times higher wavelength more suitable for this process than excimer lasers since they do not cause ablation at the surface, but can just as well produce shock waves [11, 12]. Other investigations used Nd:glass-lasers, whose wavelength is about 5 times higher than for excimer lasers but still cause ablation [13].

3.1 Method

The shock wave is the responsible energy source for the forming process. Hence the forming velocity is mainly driven through the velocity of the shock wave. The forming behaviour can be compared to that one of a high speed forming process like electromagnetic forming or explosive forming. The process duration can not be assessed yet, but it is probably longer than the laser pulse duration of 20-80 ns since the plasma formation, shock wave propagation and forming processes occur successively. The laser induced shock wave can be used in principle for all sheet metal forming processes as long as the parts are in micro- or mesoscopic range.

Fig. 6. Schematical process of laser deep drawing

The layout of the test setup for laser deep drawing works in analogy to the laser stretch-forming process, see Fig. 6: A laser cut circular sheet metal is placed on a drawing die with a die diameter of 5 mm. The blank holder is than placed onto the blank with a blank holder force, below the clamping force. In a next step one or several short laser pulses hit the specimen with a focus located at the blank surface. The high energy density of the laser radiation initiates ionisation of the close-by atmosphere and thus plasma formation takes place. The propagation of the plasma causes a shock wave, if the energy density of the laser pulse exceeds a certain threshold.

3.2 Experimental investigations

Laser deep drawing experiments were carried out in order to identify, if the laser stretch-forming process can be transferred to a deep drawing process. The decisive question was: Does the material in the flange area flow into the die during the forming process? The first observation was, that the process is very sensitive to any non-uniform conditions: Some sections of the specimen are still in the blank holder area while other sections are completely drawn into the die, see Fig. 7. This phenomenon could be attenuated by an accurate positioning of the blank, but still not be eliminated. It is assumed that apart from the positioning of the blank anisotropy, non-uniform forces, the high forming velocity and the missing friction and the form closure at the punch are responsible for the uneven deep drawing, which has to be figured out in future work.

Fig. 7. Uneven slipped laser deep drawn cups

Fig. 8. Failure of laser deep drawn cups by a) tearing and b) wrinkling

SHEET METAL MICRO FORMING

F. Vollertsen, H. Schulze Niehoff, H. Wielage, Z. Hu

11

However, the laser deep drawing process was then carried out just before the material leaves the flange area or failure occurred. Typical failures of mechanical deep drawing processes could also be observed in laser deep drawing, such as tearing, see Fig. 8a), and wrinkling, see Fig. 8b). Especially the wrinkling proves that the process changed from stretch-forming to deep drawing, since tangential forces are needed for wrinkling, which do not occur in stretch-forming. Moreover, a reduction of the initial diameter by 14 % could be measured.

Fig. 9. Typical laser deep drawn cups within the process

Fig. 10. Process window for laser deep drawing

Sound part could be produced up to a drawing ratio of 1.5 with a reduction in diameter of 8 %, see Fig. 9a) + b). The blank in Fig. 9a) was obviously not positioned

accurately, but Fig. 9b) shows a uniform drawn part, which shows anisotropic behaviour. The variation of the blank holder force and the drawing ratio was used to set up a process diagram, see Fig. 10. But this diagram cannot be used to determine the limit drawing ratio, since the parts are not drawn in completely. It just says that the limit drawing ratio is smaller or equal to 1.5 with an initial blank holder pressure of 0.76 N/mm².

4. Conclusions

• Tribological size effects could be observed in mechanical deep drawing. The friction and the deviation of the punch force increases along with miniaturization.

• Lubrication does not have the impact on micro deep drawing as on macro deep drawing.

• Laser deep drawing was carried out successfully for the first time.

• Laser deep drawing shows the similar kind of failures as mechanical deep drawing, such as tearing and wrinkling.

• A process window was worked out for laser deep drawing, which shows that the limit drawing ratio is comparable small.

5. Acknowledgement

The investigations concerning mechanical micro deep drawing were carried out within the project VO 530/19-1 „Hochgeschwindigkeitsumformen durch laserinduzierte Schockwellen“. The investigations in laser deep drawing were part of the project VO 530/6-2 „Modellierung tribologischer Größeneffekte beim Tiefziehen“. We thank the Deutsche Forschungsgemeinschaft for there financial support.

6. References

[1] N. N.: NEXUS Market Analysis for MEMS and Microsystems III, 2005-2009, WTC Wicht Technologie Consulting (2006)

[2] M. Geiger, M. Kleiner, R. Eckstein, N. Tiesler, U. Engel: Microforming, CIRP Annals, Vol. 50/2 (2001) 445-462

[3] H. Pawelski, O. Pawelski: Technische Plastomechanik, Verlag Stahleisen, Düsseldorf (2000)

[4] F. Vollertsen: Size effects in manufacturing. In: Process Scaling, F. Vollertsen, F. Hollmann (eds.), Strahltechnik, BIAS-Verlag, Bremen, Vol. 24 (2003) 1-9

[5] F. Vollertsen, Z. Hu: Tibological Size Effects in Sheet Metal Forming Measured by a Strip Drawing Test, CIRP Annals, Vol. 55/1 (2006) 291 - 294

SHEET METAL MICRO FORMING

F. Vollertsen, H. Schulze Niehoff, H. Wielage, Z. Hu

12

[6] M.W. Storoschew, E. A. Popow: Grundlagen der Umformtechnik, VEB Verlag Technik Berlin, (1968)

[7] F. Vollertsen: Laserstrahlumformen – Lasergestützte Formgebung: Verfahren Mechanismen, Modellierung, Meisenbach Verlag, Germany (1996)

[8] A. N. Pirii, R. Schlier, D. Northam: Momentum transfer and plasma formation above a surface with a high-power CO2 laser, Applied Physics Letters, Vol. 21/3 (1972) 79-81

[9] V. Ageev, A. Barchukov, F. Bunkin, V. Konov, S. Metev, A. Silenok, N. Chapliev: Breakdown of gases near solid targets by pulsed CO2 laser radiation, Soviet Physics Journal, Vol. 11 (1977) 35-60

[10] J. O’Keefe, C. Skeen: Laser-induced deformation modes in thin metal targets. Journal of Applied Physics, Vol. 44 (1973) 4622-4626

[11] H. Schulze Niehoff, F. Vollertsen: Non-thermal Laser Stretch-Forming, Sheet Metal 2005, Advanced Materials Research, Vol. 6-8 (2005) 433-440

[12] H. Schulze Niehoff, F. Vollertsen: Laser induced shock waves in deformation processing, Keynote paper, Proc. 2nd Int. Conf. Deformation Processing and Structure of Materials (2005) 43-53

[13] J.Z. Zhou, J.C. Yang, Y.K. Zhang, M. Zhou: A study on super-speed forming of metal sheet by laser shock waves, Journal of Materials Processing Technology 129 (2002) 241-244