FI RST MGPA V2 TEST RESULTS wit h I PNL SETUP And MGPA V2 TEST - PowerPoint PPT Presentation

FI RST MGPA V2 TEST RESULTS wit h I PNL SETUP And MGPA V2 TEST PRODUCTI ON R. Della-Negra, M. Dupanloup, J . Fay, S. Gascon, H. El Mamouni, B. I lle, H. Mat hez 1 CMS FRANCE MAI 2004 h. mathez

MGPA V2 TEST SETUP (0) Dif f erent ial t o single ended 50 Ω out put 50 Ω input MGPA V2 I 2C Scope or CSA803 test board 3 Gains C inj = 1. 0 pF Calibrat ed Bessel Filt er 30 ns Trigger PC PicoPulse Gen Labview ± 40 V in 400 ps Data 2 CMS FRANCE MAI 2004 h. mathez

MGPA V2 TEST SETUP (1) ARI ES SOCKET 3 CMS FRANCE MAI 2004 h. mathez

FEW PLOTS (0) Q=40pC, Gain X1 Rs1 = 3.3 Ohms Rs2 = 4.7 Ohms X1 Rs1 = 10 Ohms Rs2 = 10 Ohms 4 CMS FRANCE MAI 2004 h. mathez

FEW PLOTS (1) X1 Qin = 4.5pC X6 Rs1,2 = 10 Ohms Af t er Reset , I 2C = 64 Pkt X1 ≈ 50ns X12 Pkt X6 ≈ 50ns Pkt X12 ≈ 50ns Pkt PA ≈ 25ns PA 5 CMS FRANCE MAI 2004 h. mathez

FEW PLOTS (2) X12 m PA X12 p Qin = 4.5pC Rs1,2 = 10 Ohms Af t er Reset I 2C = 64 X1 plus and minus out put Base line are : • 799.2 mV = 0.8 V (ADC input ) • 1488.8 mV < 1.7 V (ADC input ) 6 CMS FRANCE MAI 2004 h. mathez

FEW PLOTS (3) X1 m PA X1 p Qin = 40pC Rs1,2 = 10 Ohms Af t er Reset I 2C = 64 X1 plus and minus out put Base line are : • 670 mV < 0.8 V (ADC input ) • 1527 mV < 1.7 V (ADC input ) 7 CMS FRANCE MAI 2004 h. mathez

FEW PLOTS (4) Peaking Time Gain X1 and X6 8 CMS FRANCE MAI 2004 h. mathez

FEW PLOTS (5) Power Supplies 573 mW 9 CMS FRANCE MAI 2004 h. mathez

PRELI MI NARY YI ELD Chips t est ed :133 Bad chips : 7 YI ELD ~ 95% 10 CMS FRANCE MAI 2004 h. mathez

MECHANI CAL PROBLEMS (0) Socket : wit h microst rip, same t echnology as previous (FPPA) NO problems package EDQUAD 52 pins ASAT New problems with Aries socket and ASAT TQFP100 package o The socket was t est ed wit h At lant ic package at ARI ES house : Some devices present an excess of plast ic in t he corners � Some small chips goes under t he microst rip : no elect rical cont act , t he MGPA does not work ! � Microst rips could be t wist ed due t o t his small chips, t he MGPA does not work ! Solutions � Very caref ully clean up t he PCB and t he socket 11 CMS FRANCE MAI 2004 h. mathez

MECHANI CAL PROBLEMS (1) Microst rip 12 CMS FRANCE MAI 2004 h. mathez

MGPA V2 TEST PRODUCTI ON 13 CMS FRANCE MAI 2004 h. mathez

MGPA V2 TESTI NG PRODUCTI ON REQUI REMENTS PARAMETER MEASUREMENTS TO BE DONE DC power supply Current I2C register Internal Default Value (64) after reset Scan Chain Functionality Put some logic input to several values and clock them Offset Current Generation I2C register Base line on 3 differentials gains amplifiers 3 I2C values must be tested Calibration Pulse Generator 3 I2C values which represents 3 different Output amplitude on 3 differentials gains amplifiers charge injection (with lookup table values) Qin = 4.5 pC Rise and Fall Time and peak amplitude on G12 differential gain amplifier Qin = 8 pC Rise and Fall Time and peak amplitude on G6 differential gain amplifier Qin = 50 pC Rise and Fall Time and peak amplitude on G1 differential gain amplifier Noise RMS output amplitude on G12 differential gain amplifier St art t est ing product ion : middle of Sept ember (if product ion st art s on middle of j une) 14 CMS FRANCE MAI 2004 h. mathez

TESTI NG HOUSES SUM UP (0) MICROTEC ASAT EDGETEK TEST PROGRAM DEVELOPMENT € 15 500,00 ? € 12 000,00 HARDWARE DEVELOPMENT € 6 900,00 ? PRICE PER COMPONENT € 0,27 ? € 0,28 TIME PER COMPONENT [s] 10,00 ? TOTAL TIME [h] 361,11 ? TOTAL COST € 57 500,00 ? € 48 400,00 FORECAST TIME SCHEDULE 5/6 weeks >= 8 weeks after hardware dev Chip to be tested 130 000 130 000 15 CMS FRANCE MAI 2004 h. mathez

TESTI NG HOUSES SUM UP (1) •ASAT Europe business cont act is current ly overloaded •Microt ec and Edget ek appear t o be very good candidat es Weakness St rengt h • Experienced •Flexibilit y ? Microt ec • Already used inside CERN • I N2P 3 part ner f or small quant it ies • Background f or high volume Edget ek • French company product ion • Flexibilit y f or cust om services • Price We would suggest t o give t o Microt ec t he MGPA product ion t est s 16 CMS FRANCE MAI 2004 h. mathez

TESTI NG CONCLUSI ON (MEETI NG 10 MAY CERN) Buf f er Test ing House : EDGETEK Forecast Schedule : � 100 buf f ers f or J une 10 t h � 5000 buf f ers f or J uly 15 t h MGPA Microt ec or Edget ek but we are wait ing t he Asat ’s answer Edget ek seems t o be t he best way t o t est t he MGPA Forecast Schedule : � 1700 MGPA f or j uly 15 t h � St art t est ing product ion : middle of Sept ember 17 CMS FRANCE MAI 2004 h. mathez