田口方法用於探討錫膏印刷 ... - engh.kuas.edu.tw

1 120~133 E-mail (Surface Mount Technology SMT) (Surface Mount Devices SMD) (yield) (reliability) SMT (Taguchi Methods) / (Signal-to-Noise Ratio S/N) (process capability index Cpk) SMT 1. (Print Circuit Board PCB) PCB (Surface Mount Technology SMT) (Surface Mount Devices SMD) PCB (Fine pitch) (Ultra fine pitch) (yield) (reliability) SMT PCB PCB (bridge)

2 SMT 60% [1] PCB [2-3] 2011 National Kaohsiung University of Applied Sciences, ISSN 1813-3851 121 (SMT) 2. SMT SMT ( 1) [4] SMT ((process capability index Cpk) (orthogonal array) / (S/N) Cpk (flux) Life time (flux) Life time 1.)

3 3. SMT (stencil) (squeeze) (solder paste) PCB (pad) , 122 pad [5] ( 2) SMT 50%-60%[6] 2. ( [6]) 4 4 3 (level) 4 3 L9(34) 9 1 9 PCB(Print Current Board) 9 (Mean) (Standard Deviation) Cpk 2 123 1. L9(34) - (Nt) (mm/sec) (mm/sec) (mm)MeanStd Deviation115 15 30 65 15 30 65 15 30 65 2. - MeanStd DeviationRangeSpec width 4 Mean (1) (Standard Deviation ) (2) Cpk [7-8] = = niixn1221 (1) Cpk = () 6*1TK (2) n = xi = ( ) = K = ca Ca = MS (3) 124T = (USL-LSL) MS = =()LSLUSL+21 USL = LSL = SMT cpk 9 9 cpk 9 MIntab 3 20Nt 15mm/sec Mean of Standard EKRA Main Effects Plot for Standard DeviationsMean of Standard EKRA Main Effects Plot for Standard Deviations 3.

4 4 S/N 20Nt 15mm/sec 3 2 5 10 10 20Nt 15mm/sec 125 Mean of SN ratios321242322321321242322321 EKRA Main Effects Plot for SN ratiosSignal-to-noise: Nominal is best (10*Log(Ybar**2/s**2))Mean of SN ratios321242322321321242322321 EKRA Main Effects Plot for SN ratiosSignal-to-noise: Nominal is best (10*Log(Ybar**2/s**2)) 4. S/N 3. - 2 Mean Std Deviation Range Spec width Cpk 10 5. - 10 SMT (Electro polishing) 6 [9-11] 126 6.

5 ( [12]) (1) (2) (3) (4) (5) (6) (7) (8) (life time) 8 (1) , 10~15kg/cm2 (2) pulse 800Hz~1200Hz (3) 3 DCV DCV (4) laser , mm/sec 6~12mm/sec (5) 80~100A 127(6) 120sec 150sec 180sec (7) 40 60 (8) (level) (1) 2 level 3 level L18(21*37) 4 18 18 SMT 18 SMT 20 mm 18 5 4.

6 (Kg) (Hz) (DCV) (mm/sec) (A) (sec) ( C) ( )1 15 800 6 80 120 40 20002 15 800 3 9 90 150 50 25003 15 800 12 100 180 60 30004 15 900 6 90 150 60 30005 15 900 3 9 100 180 40 20006 15 900 12 80 120 50 25007 15 1000 9 80 180 50 30008 15 1000 3 12 90 120 60 20009 15 1000 6 100 150 40 250010 12 800 12 100 150 50 200011 12 800 3 6 80 180 60 250012 12 800 9 90 120 40 300013 12 900 9 100 120 60 250014 12 900 3 12 80 150 40 300015 12 900 6 90 180 50 200016 12 1000 12 90 180 40 250017 12 1000 3 6 100 120 50 300018 12 1000 9 80 150 60 2000 5. 128 S/N

7 4 (4) (Sm) (5) (Ve) S/N( ) (6) 6 7 ( 7) [8] ySm = = n*2 (4) yVe = = 11 n ( 2- S) (5) =201iiymS/N( )= 10log(n1emVS-n1) (6) n = ( 20) 129 6. SmVe 7. A. (Kg)B. (Hz)C. (DCV)D. (mm/sec)E. (A)F. (sec)G. ( C) H. ( )Level 6531542 1302021222324252627282930A1A2B1B2B3C1C2C 3D1D2D3E1E2E3F1F2F3G1G2G3H1H2H3dB 7. E3D1H3G2F2C3B3A1 (Max-Min) Rank 5 (E3H3D1G2F2) S/N = +( E3- )+( H3- )+( D1- )+( G2- )+( F2- ) =181( + + +..+ + ) = : E2H1D1G3F2C1B3A1 = + + + + E3D1H3G2F2C3B3A1 = + + + + = S/N ( ) ( ) S=10log(n1(S - V)) 18 (S) ( 8 ) me 8.

8 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 131 ( 9) ( 8) 9. A. (Kg)B. (Hz)C. (DCV)D. (mm/sec)E. (A)F. (sec)G. ( C) H. ( )Level (dB) 8. 8 S( ) D2C3G1H1F1B1E2A1 S/N S/N E3D1H3G2F2C3B3A1 (S) D2C3G1H1 F1B1E2A1 S/N S/N E3H3D1G2F2 (S) D2G1 S Sout = SE3+SH3+SD1+SG2+SF2-4S = +(4* )= m= STA=10log(m)2 = = S=STA- Sout= ( )= S B S/N E3H3D1G2F2C3B1A1 out S/N 132 ou t = + + + + = = E3H3D1G2F2C3B1A1 10 SMT 11 9 Cpk Cpk 10.

9 (Kg) (Hz) (DCV) (mm/sec) (A) (sec) ( C) ( ) 11. Mean Std DeviationRange Spec widthCpk 9. - 4. SMT (cpk ) 133(1) 15Kg 800Hz 6mm/sec 100A 150sec 50 C 3000 (2) SMT 20Nt 15mm/sec (1) S/N Cpk (2) SMT [1] 3-10 2005 [2] Richard, S.

10 C., The complete solder paste printing processes, Surface Mount Technology, Vol. 13, No. 1, , 1999. [3] Steve Hall, Screen printer intelligence requirements for low defect process capability, Surface Mount International Conference & Exposition - Proceedings of the Technical Program, , 1993. [4] - 2010 [5] - 5-7 2005 [6] 15-16 2003 [7] 2001 [8] 2000 [9] 10-23 2004 [10] Kozak, J., Domanowski, P., Inverse problem of shaping by electrochemical generating machining. J. Mater. Process. Technol. 109, pp. 347 353, 2001. [11] An, Q. Shong, Research on the Combined Electro Chemical Mechanical Polishing, Electro Machining, No. 2, 2002. [12] 9-11 2009