An Ultrasonic Knife System for MEMS Packaging by h3U571lt

VIEWS: 41 PAGES: 5

									                       An Ultrasonic Knife System for MEMS Packaging
                             Jin Sha1, a, Zhi-yuan Yao2, b and Yang Jiao3, c
    1,2,3
            State Key Laboratory of Mechanics and Control of Mechanical Structures, Nanjing University of
                              Aeronautics and Astronautics, Nanjing 210016, China
                   a
                   shajin870422@163.com, bzyyao@nuaa.edu.cn, cyang244140@hotmail.com


Keywords: ultrasonic knife cutting system; finite element analysis; cutting force; ultrasonic welding.

Abstract: This paper proposes an ultrasonic knife system for MEMS packaging. The ultrasonic knife
system is consisted of an ultrasonic transducer, a cutter and a gripper feeder. The ultrasonic transducer
engenders high frequency vibration, which lead to the resonance of the structure. Amplitude transformer can
magnify the amplitude. By the impact and collision of the cutter, the material can be cut through, and the
high temperature created by high-frequency vibration can do the welding. The structure is designed and
optimized by the finite element method, and a model machine is produced. According to the experimental
results, the ultrasonic knife system has the virtues of high cutting force and better wedding feature, which
are suitable for MEMS packaging.

Introduction
   Ultrasonic cutting and wedding, which belong to ultrasonic machining technology, rose up in the 20th.
Different from traditional machining technology, the key point of it is an ultrasonic cutter. The cutter works
as follows: firstly, the electronic ultrasonic generator sends out the ultrasonic single in 20~50 kHz range;
secondly, the ultrasound-mechanical converter would transform the single into mechanical vibration;
thirdly, Amplitude transformer can magnify the amplitude; at last, by the impact and collision of the cutter,
the material can be cut through, and the high temperature created by high-frequency vibration can do the
welding[1-6].
   In our country, the finite element simulation of the piezoelectric transducer for cutting has been done by
Y. Q. Guo[7] ; the model analysis of the ultrasonic cutting system has been done by J. Q. Liu [8]. Aboard,
novel multiple blade ultrasonic cutting device has been designed by Andrea Cardoni [9];A studied on
ultrasonic vibration cutting of low alloy steel has been done by Chandra Nath[10];an ultrasound device use
in maxillofacial surgery has been designed by Georg Eggers[11];the application of ultrasonic wedding in
MEMS and IGBT module manufacture has been done by Jongbaeg Kim[12] and Yoshitaka Nishimura[13-15].
   For there is only simulation analysis of the ultrasonic knife in our county, this paper has not only
established the finite element model and optimized the structure by the finite element method, but also made
a model machine for experiment. Furthermore, the mechanical property of ultrasonic cutting is obtained by
measuring the cutting force of different kinds of material in the designed system and preliminary wedding
experiment has been done.

1 Structural Design
   Fig.1 shows the appearance of the ultrasonic knife system designed by this paper. The shell is made of 45
steel processed by line-incise, which is very simple and compact. Fig.2 shows the inner structure of the
ultrasonic knife system, the ultrasonic cutter is fixed in the center of the inner frame by a flexure hinge, and
the vertical displacement of the inner frame is changed by a bolt above. The cutter is consisted of a cutter
head, 4 pieces of PZT8, 2 electrode plates and tail. As it runs, the voltage at ultrasonic frequency is applied
on 2 electrode plates, which would cause longitudinal resonance of the cutter. And the cutter head is designed
just as an amplitude transformer to magnify the vibration, so the system would be more efficient.
 Fig.1 The appearance of            Fig.2 The inner structure of the         Fig.3 Finite element model of the
the ultrasonic knife system             ultrasonic knife system                             cutter

2 Dynamic Analysis
   The core part of the ultrasonic knife system is the ultrasonic cutter. This paper designed it by finite
element software ANSYS. Firstly, the model is established with parametric method, the element type of the
PZT8 is solid5; other (45steel) is solid45. Fig.3 shows the finite element model of the cutter. Secondly, the
grid is meshed for the model analysis and harmonic response analysis. The position of the hinge is
determined by the node of the longitudinal vibration, which is the intersection of the red line and green line in
Fig.4. At last, sensitivity analysis is performed, by which we find that the length of amplitude transformer is
the major influence on the objective function, but the width and thickness can be ignored; we changed the
length of amplitude transformer to achieve the optimal design. Finally results of simulation are showed in
Fig.5. We can see that the frequent of the first longitudinal resonance mode are 42387Hz and the amplitude
of the cutter head, whose maximum value has reached 1.7μm, has been effectively magnified.




Fig.4 Vibration mode shape of the       Fig.5 Colored atlas of the         Fig.6 The result of modal
              cutter                      vertical displacement                   experiment

3 Experimental Study
   3.1 Modal Experiment
   Base on the dynamic analysis above, we made a model machine. Then we did the modal experiment of the
cutter by the PSV vibrometer: Set the sweep frequent range from 0 to 50 kHz and the voltage be 100V. The
experiment results are showed in Fig.6, we can see that the frequent of the first longitudinal resonance mode
are 41880Hz which is much closer to the simulation result and the peaks of interfering frequency are so low
that can be ignored, so the optimal design is on target.
   3.2 Studies on the Cutting Force
   3.2.1 Introduction to the Experiment System
   Base on the dynamic analysis above, we made a model machine. Then we did the modal experiment of the
cutter by the PSV vibrometer: Set the sweep frequent range from 0 to 50 kHz and the voltage be 100V. The
experiment results are showed in Fig.6, we can see that the frequent of the first longitudinal resonance mode
are 41880Hz which is much closer to the simulation result and the peaks of interfering frequency are so low
that can be ignored, so the optimal design is on target. Base on the dynamic analysis above, we made a model
machine. Then we did the modal experiment of the cutter by the PSV vibrometer: Set the sweep frequent
range from 0 to 50 kHz and the voltage be 100V. The experiment results are showed in Fig.6, we can see that
the frequent of the first longitudinal resonance mode are 41880Hz which is much closer to the simulation
result and the peaks of interfering frequency are so low that can be ignored, so the optimal design is on target.
                                                                   LabVIEW interface


                               Force sensor
                                                                    Data acquisition               Ultrasonic cutter
                  Clamp A                                                 card
                                                                                                   Cutting
                                                                                                    force
                  Ultrasonic                                                 D/A       Digital
                    cutter                                                             signal
                                            Clamp B                     LC1015                       Force sensor




 Fig.7 Measurement of cutting          Fig.8 Functional block diagram of        Fig.9 LabVIEW interface
               force                             the test system
   3.2.2 Test Results and Discussion
   Firstly, the paper has measured the cutting force of 65Mn steel and 45 steel cutter heads at the voltage of
200V separately, as the preloading change. As showed in Fig.10, the cutting force of 65Mn steel is 50N
higher than the 45 steel on average. In consideration of the stiffness of 65Mn steel is higher than 45 steel, we
infer that the cutting force is proportional to the stiffness of material.
                                                      65Mn steel                                                                                    Force
                                                       45 steel                                                                                             Force
                                                                      F(N)
  Cutting force




                                                                                                                       Cutting force




                                                                                                                                       Preloading
                               Preloading                                                  T(μs)

  Fig.10 Cutting force of 65Mn and          Fig.11 The characteristic of      Fig.12 The relationship between
      45 steel cutter head(200V)                   cutting force                cutting force and preloading
   Secondly, we consider of the cutting force of 65Mn steel cutter heads with preloading 16N, the voltage at
200V and the sampling frequency at 100 kHz. The experimental results and fitting results are showed in
Fig.11. It can be found in Fig.11 that the cutting force vibrate at small amplitude in period, just as the sine
wave, which fits with the characteristic of longitudinal vibration very well.
   Thirdly, we consider on the relationship of cutting force and preloading. Keep the voltage at 200V then
change the preloading, the measure result is as Fig.12 shows. It can be found that the cutting force increased
linearly with the preloading, a dramatic rise of about 60N can be achieved. The cutting force of ultrasonic
cutter is greater than the common cutting whose cutting force is equal to preloading only. This would
improve the cutting efficiency.
   3.3 Wedding Experiment
   Before the experiment, we replace the sharp cutter head with the welding head with flat face, so the
ultrasonic cutter can do the wedding. When it is working, the flat face of welding head press down on the
weldment and solder, as it is showed in Fig.13.
   Firstly, we weld the end of a copper wire, whose diameter is 70μm, on an aluminium plate. Heat sealing
and ultrasonic wedding are used respectively, with tin to be the solder. After the wedding, the high-power
microscope is used to observe the welding spot feature. The result is showed in Fig. 13, we can find that the
top of solder is167μm higher than copper wire in in Fig. 13(a) and 97μm higher in Fig. 13(b). Moreover the
welding spot of the ultrasonic wedding is much smoother than the heat sealing. So the ultrasonic wedding can
control the size of the welding spot well. Therefore we can control the volume of solder precisely.
                                                         Top of solder
                                                                                                         Top of solder
                                                                              Copper wire



                                                      Copper wire




                                        (a) Welding spot feature of          (b) Welding spot feature of ultrasonic
                                                                                           wedding
                                                   heat sealing
Fig.13 The ultrasonic                              Fig.13 Welding spot feature comparison between
 cutter for wedding                                       heat sealing and ultrasonic wedding

                   Surface relief




                                                                                                    e
                                                                                                   zon
                                    e




               Copper sheet A
                                 zon




                                                                           Copper sheet A




                                                                                              ing
                             ing




                                    Copper sheet



                                                                                              dd
                                        B
                             d




                                                                                            We
                          ed




         Overflowing                                                                               Copper sheet B
                         W




           Solder



      (a) Side boundary feature of heat sealing              (b) Side boundary feature of ultrasonic wedding
            Fig.13 Side boundary feature comparison between heat sealing and ultrasonic wedding
    Secondly, in order to study the difference in the side boundary of heat sealing and ultrasonic wedding, two
pieces of copper sheet are chosen for the wedding experiment, with tin to be the solder. Side boundary feature
of two kinds of wedding under the high-power microscope is showed in Fig. 14 and the side length of the
grid is 5μm. It can be found in Fig. 14(a) that the side boundary feature of heat sealing is irregular and there
is overflowing solder and bump shape which would lead to bad wedding quality; but the boundary feature of
ultrasonic wedding is much smoother and with less overflowing in Fig. 14(b), so the wedding would be of
high quality. Moreover, the working temperature of ultrasonic wedding is low and local, so it is much more
suitable for MEMS packaging.

4 Conclusion
      This paper has designed a kind of ultrasonic knife system for MEMS packaging. The finite element
model has been established and the structure has been optimized by the finite element method, and a model
machine has been made for the experiment. We have studied on the cutting force and wedding feature by a
designed experiment. To be more specifically, it can be summarized as follows:
      1) At the same voltage, the cutting force of 65Mn steel is higher than the 45 steel on average; we infer
that the cutting force is proportional to the stiffness of material.
     2) Keep constant the preloading, the cutting force vibrates at small amplitude in period, just as the sine
wave, which fits with the characteristic of longitudinal vibration very well.
    3) The cutting force increased linearly with the preloading, a dramatic rise of about 60N can be
achieved. It is greater than the common cutting whose cutting force is equal to preloading only. This would
improve the cutting efficiency.
    4) The ultrasonic wedding can control the size of the welding spot well. Therefore we can control the
volume of solder precisely. And the boundary feature of ultrasonic wedding is much smoother and with less
overflowing than the heat sealing. Moreover, the working temperature of ultrasonic wedding is low and
local.
     This ultrasonic knife system is smart, lower energy consumption but with high cutting force and low
working temperature. It can be integrated in precise working platform for MEMS packaging.

Acknowledgements
    This work is supported by the National Basic Research Program (973 Program) (2011CB707602) , the
Key Program of National Natural Science Foundation of China and the program of National Natual Science
Foundation of China (50975136).

References
[1] Susann Zahn, Yvonne Schneider and Harald Rohm: Innovative Food Science and Emerging
Technologies Vol. 7 (2006), p. 288–293
[2] G. Sinn, B. Zettl, H. Mayer, S. Stanzl-Tschegg: Journal of Materials Processing Technology Vol. 170
(2005), p. 42–49
[3] J. Rimkevičienė, V. Ostaševičius, V. Jūrėnas, R.Gaidys: MECHANIKA Vol. 1(2009), 1(75).
[4] Reimund Neugebauer, Andrea Stoll: Journal of Materials Processing Technology Vol. 149 (2004), p.
633–639
[5] G. Arnold, L. Leiteritz, S. Zahn, H. Rohm: International Dairy Journal Vol.19 (2009), p. 314–320
[6] M. Robiony, F. Polini1, F. Costa1, N. Zerman and M. Politi: Int. J. Oral Maxillofac. Surg. Vol. 36 (2007),
p. 267–269
[7] Y. Q. Guo and L. Y. Zhang: Modern Machinery Vol.6 (2008), p. 44-46
[8] J. Q. Liu, J. C. Yan and S. Q. Yang: Journal of Harbin Institute of Technology Vol.33 (2001), p.
435-438
[9] Andrea Cardoni, Margaret Lucas , Matthew Cartmell and Fannon Lim: Ultrasonics Vol. 42 (2004), p.
69–74
[10] Chandra Nath, M. Rahman, S.S.K. Andrew: Journal of Materials Processing Technology Vol. 192–193
(2007), p. 159–165
[11] Georg Eggers, Johannes Klein, Julia Blank, Stefan Hassfeld: British Journal of Oral and Maxillofacial
Surgery Vol. 42 (2004), p. 451—453
[12] Jongbaeg Kim, Bongwon Jeong, Mu Chiao, and Liwei Lin: IEEE TRANSACTIONS ON ADVANCED
PACKAGING Vol.32 (2009), p. 461-467
[13] S.-F. Ling,X. Li,Z. Sun: ASME International Mechanical Engineering Congress and Exposition,
(2007)
[14] Misugi Hongoh, Masafumi Yoshikuni, Hiroyuki Miura, Tetsugi Ueoka, Jiromaru Tsujino: 2004 IEEE
International Ultrasonics, Ferroelectrics and Frequency Control Joint 50th Anniversary Conference p.
2326-2329
[15] Yoshitaka Nishimura, Kazumasa Kido, Fumihiko Momose and Tomoaki Goto: Proceedings of The
22nd International Symposium on Power Semiconductor Devices & ICs, Hiroshima p. 293-296

								
To top