Flex review. Outline. Preproduction: layout and status QA during the production Electrical tests done/foreseen Details on R signals Details on R LV

Transcription

1 Flex review C.Gemme, G.Gariano, E.Ruscino INFN Genova Rui de Oliveira, Alexandra Grindis Cern Outline Preproduction: layout and status QA during the production Electrical tests done/foreseen Details on R signals Details on R LV Next steps: Usage of batch3 Moving towards the production 1

2 Preproduction 6 sheets with 3 flex on each. 3 sheets A side, 3 for C side 9 sets On each sheet, one flex has the 5 um LVDS_Ref plane uniform, two have it hatched. Possible contribution to material reduction, to be measured the implication on impedence difficult to simulate in 2D software. Production launched beginning of November. Delivery of Batch 1 and 2 (12 flexed, i.e. 6 sets) mid-march. (~3 m.u.) Batch3 (3 sets) left on hold. 08/05/2012 Flex Review 2

3 Produced Flexes ID Batch Delivered LVDS_Ref Hatched Main Production comments Connectors Loaded (who/when) Current Location 223A1 10-Mar NO Wings Fixed CernSMD /19-March CERN - FX 223A2 10-Mar Yes CernSMD /8-March SLAC 223A3 10-Mar Yes Phoenix/22-March Genova Extra wing thickness 1 615C1 10-Mar NO (no modules) Wings Fixed CERNSMD/27-March CERN - FX 615C2 10-Mar Yes CernSMD /8-March Yannick 615C3 10-Mar Yes Phoenix/22-Marh Genova 161A1 19-Mar NO CernSMD /20-March Genova 161A2 19-Mar Yes CernSMD /20-March CERN - FX 161A3 21-Mar Yes CernSMD/23-March Yannick Wing Thickness 2 ~200um 444C1 19-Mar NO CernSMD /20-March Genova 444C2 19-Mar Yes CernSMD /20-March Genova 444C3 21-Mar Yes CernSMD/22-March CERN - FX 3

4 Produced Flexes: Naming ID Batch Delivered LVDS_Ref Hatched Main Production comments Connectors Loaded (who/when) Current Location 223A1 10-Mar NO Wings Fixed CernSMD /19-March CERN - FX 223A2 10-Mar Yes CernSMD /8-March SLAC 223A3 10-Mar Yes Phoenix/22-March Genova Extra wing thickness 1 615C1 10-Mar NO (no modules) Wings Fixed CERNSMD/27-March CERN - FX 615C2 10-Mar Yes CernSMD /8-March Yannick 615C3 10-Mar Yes Phoenix/22-Marh Genova 161A1 Foil TOP/LVDS2 19-Mar Foil GND1/LVDS1 NO Number stack Cu Number Stack Al Number Final stack CernSMD /20-March 1A 6A 16A 1A 161A Genova 161A2 19-Mar Yes CernSMD /20-March CERN - FX 2A 2A 22A 3A 223A 161A3 21-Mar Yes CernSMD/23-March Yannick Wing Thickness 2 3A 3A ~200um 33A 6A 336A 444C1 19-Mar NO CernSMD /20-March Genova 4C 4C 44C 4C 444C 444C2 19-Mar Yes CernSMD /20-March Genova 5C 5C 55C 2C 552C 444C3 21-Mar Yes CernSMD/22-March CERN - FX 6C 1C 61C 5C 615C 4

5 Stack as-built Layer Name Type Material Thickness (um) Dielectric 12.5 GLUE 10 TOP (HV) Conductor Copper 15 Dielectric Pyralux 25 µm 25 LVDS2 Conductor Copper 15 GLUE 12.5 Dielectric Kapton 25 µm 25 GLUE 12.5 GND1 Conductor Copper 5 Dielectric Pyralux 75 µm 75 LVDS1 Conductor Copper 15 GLUE 10 Dielectric Kapton 50 µm 50 GLUE 10 Aluminium Plane Aluminium 50 GLUE 10 Dielectric Kapton 12.5 µm 12.5 GLUE 10 Aluminium Plane Aluminium 50 GLUE 10 Dielectric Kapton 12.5 µm ?? 2 Al Layer Name Type Material Thickness (um) Coverlay Dielectric KAPTON 12.5 GLUE 12.5 TOP (HV) Conductor Copper 19 Dielectric Pyralux 25 µm 25 LVDS2 Conductor Copper 18 GLUE Dielectric GLUE 25 1 GLUE Dielectric GLUE 25 GND1 Conductor Copper 5 Dielectric Pyralux 75 µm 75 LVDS1 Conductor Copper 19 GLUE Dielectric GLUE 25 Dielectric Kapton 25 µm 25 2 GLUE Dielectric GLUE 25 Aluminium Plane Aluminium 50 GLUE Dielectric Kapton 12.5 µm from 40 to 60 GLUE Aluminium Plane Aluminium 50 Coverlay Dielectric GLUE 12.5 KAPTON 12.5 Cu thickness um due to plating step. Thickness checked on any foil during production. Per Rui: why the dieletric between the first and second Cu layer has been changed? 496 5

6 Thickness checks Per Rui: I do not understand the thicknesses Number Should be 99?? GND1/LVDS1 Layer Without glue Should be 64?? Top/LVDS2 Layer Without glue Copper stack before copper drill step Final copper stack After copper drill and plating Aluminium stack Gnd/Vcc stack After Cu plating (top&bot ) coverlay top & bot 223A C C A A C Layer Name Type Material Thickness (um) Coverlay Dielectric KAPTON 12.5 GLUE 12.5 TOP (HV) Conductor Copper 19 Dielectric Pyralux 25 µm 25 LVDS2 Conductor Copper 18 GLUE Dielectric GLUE 25 1 GLUE Dielectric GLUE 25 GND1 Conductor Copper 5 Dielectric Pyralux 75 µm 75 LVDS1 Conductor Copper 19 GLUE Dielectric GLUE 25 Dielectric Kapton 25 µm 25 2 GLUE Dielectric GLUE 25 Aluminium Plane Aluminium 50 GLUE Dielectric Kapton 12.5 µm from 40 to 60 GLUE Final Aluminium Plane Aluminium 50 Aluminium Final stack GLUE 12.5 Coverlay Without Dielectric KAPTON /05/2012 Flex Review 6 1?? 2 Al

7 Thickness checks N Thicknes s at connecto rs area (µm) Thickness of the stack under wing From number 1 to 16 (µm) Δ 161A A A C C C A A A C C /05/2012 Flex Review 7

8 Produced Flexes: wing thickness On Batch1 extra thickness under the wings: This prevents us to use them for module loading. Two flex ~ successfully reworked (1 signal failure). 8

9 LVDS2 Lines properties Measured track width and clearance during production assembly for each layer. As an example LVDS2: Number Track 76 µm Clearance 76 µm Vias 0.6 mm Vias 0.9 mm Vias 1.27 mm 161A A A C C C Top width: width um width um width um LVDS2 width um spacing um LVDS1 width um spacing um 9

10 Lines properties Line width and spacing Cross-checked during the qualification. 444C1 LVDS _Ref uniform visible 08/05/2012 Flex Review 10

11 Lines properties Line width and spacing Cross-checked during the qualification. 444C2 LVDS _Ref Hatched visible 08/05/2012 Flex Review 11

12 Effect on the impedence Nominal: 78 Ohm Width Spacing Z0 Calculate impedence with width1=width2=67um (2.63 mil) and spacing 87um (3.42 mil) width1=width2=72 um (2.83 mil) and spacing 82um (3.22 mil) Difference of few % on the impedence. 08/05/2012 Flex Review 12

13 161A1 wing 12 TOP and LVDS2 layers Shift VIA SHIFTING (1)

14 VIA SHIFTING (2) 444C1 First wing Last wing LVDS2 layer LVDS2 layer

15 Electrical Tests List of test done in Genova Search for shorts on the connectors and on the wings Resistance of signals connector-wings Resistance of LV connector-wings To be fully implemented Impedence (on samples) Cross talk (on samples) HV BER in SLAC (on samples) 15

16 Test results: Overview ID R lines R LV RVcc RGnd wing bend Continuity/Isolat ion before Comments Continuity/Isol ation after 223A1 y y y FAIL 1 high R on VCC PASS 223A2 y n FAIL 1 FE VCC open 223A3 y y PASS 615C1 y y y FAIL 615C2 y y y FAIL 615C3 y y PASS 161A1 y y PASS 1 open on signals due to reworking; 3 high R on VCC 1 FE GND open (known and repairable); 1 short on signals PASS PASS 161A2 y y y FAIL 1 high R on VCC PASS 161A3 y y FAIL 3 high R on VCC 444C1 y y FAIL 3 shorts on signals 444C2 y y FAIL 1 shorts on signals 444C3 y y y PASS PASS 16

17 Signals lines Failures 1 signal open (615C1, due to wing repair) 5 signal shorts (1 on 615C2, 1 on 444C2, 3 on 444C1) 615C2 Short DO4n/p: short close to connector: 0.67 Ohm 444C2 SHORT (37-38) P/N DI9-DI10 the only one visible short close to connector: conn. 0.1 ohm; w9 12,4 ohm; lw10 11,5 Ohm 444C1 SHORT (37-38) P/N DI9-DI10 short close to connector: conn. 0.4 ohm; w ohm; lw10 10 ohm 444C1 SHORT(34-35) P/N CK11-CK12 short Close to wing 11: conn. 9.4 ohm; w ohm; w12 0,9 ohm 444C1 SHORT(29-30) P/N CK13-CK14 short close to wing 14: conn ohm; w13 3.1; w ohm 17

18 Short vias on 444C2 (on pins 37 and 38 next to connector) See next slide

19 Short vias on 444C2 (on pins 37 and 38 next to connector)

20 Signal lines: Resistivity Line resistivity is ~ 20% higher than nominal (to be checked with measured width and thickness). Not Understood systematic difference between differential lines (Ckn/p; Din/p; Don/p) C3 - Ckn 444C3 - Ckp 615C3 - Ckn 615C3 - Ckp 444C1 - Ckn 444C1 - Ckp /05/2012 Flex Review 20

21 Signal lines: Resistivity Line resistivity is ~ 20% higher than nominal (to be checked with measured width and thickness). Not Understood systematic difference between differential lines (Ckn/p; Din/p; Don/p): ~ +0.2 Ohm C3 Ckp-Ckn 615C3 Dip-DIn 444C2 Ckp-Ckn 444C2 Dip-DIn /05/2012 Flex Review 21

22 LV lines Resistivity Tested during production: On the Al stack ~60-80 mohm (expected ~ 60 mohm) After the wings on the final stack ~ mohm Resistivity tested in Genova not very uniform in Genova due to the contact? Some extremely large values One flex tested in real conditions with loads. The other flex are measured with ohmmeter (4 points) 22

23 Power supply connector Load each wing with 3.3 Ohm

24 FlexBus 615C2 measured shape: W4 W3 W2 W1 Connector L2 L1 Measurements with a load on each wing of 3.3 Ohm and applying Vpower supply = 0.5 V: I tot = A; DV_VCC (from connector to wing1) = 0.052V DV_GND (from connector to wing1)=0.042v DV round trip nominal = V/0.456 A* 2 A = V 300 mv budget Now calculating the contribution to the Resistance of L1 and via/wings

25 FlexBus 615C2 measured shape: L2 Wing+via L1 DV_VCC = R_VCC L1 * I tot + R_VCC wing +via * I wing DV_GND = R_GND L1 * I tot + R_GND wing +via * I wing Assuming the resistance of vias and wing from Alexandra/Rui calculation (wing) or direct measurements (Via) R_VCC wing +via = =0.03 Ohm R_GND wing +via = =0.03 Ohm Measured the voltage drops on the other wings and reasonable to assume I wing = I tot /4 R_VCC L1 = (DV_VCC - R_VCC wing +via * I tot /4)/ I tot R_GND L1 = (DV_GND - R_GND wing +via * I tot /4)/ I tot R_VCC L1 = ( (0.03*0.456/4))/0.456) = Ohm R_GND L1 = ( (0.03*0.456/4))/0.456) = Ohm Rather consistent with Alexandra measurements.

26 R VCC and GND shape Calculated L1 L3 L2 RL1 = RL3 = 0,014 Ohm RL2 =0,0067 Ohm Rtot Wing1-Wing2 = 0,014+0,014+0,0067 =0,0347 Ohm

27 LV resisitivity In general resistivity larger than expected (could be due to contact or oxide). Not understood why RGnd can on average be larger than RVcc on two flex. Few (8/(16*5) values very large on wings (>0.2 instead of 0.1 Ohm) Not observed slightly increasing R from first to last wing in the same group. RVcc (Ohm) RGnd (Ohm) /05/2012 Flex Review 27

28 LV inside a group A1 223A3 161A1 161A2 161A A1 223A3 161A1 161A2 161A VCC VCC A1 223A3 161A1 161A2 161A3 VCC /05/2012 Flex Review 28

29 First test: impedence 19/10/2011 C. Gemme, INFN Genova, Module Flex 29

30 Production steps Not used Qualified Steps Task Place Schedule Responsible 1 2 Flex manufacturing (Full flex equipped with flat wings+qa) Flex connectors mounting Manual process followed by special flux cleaning 3 1 st electrical test Cable saver is loaded Test tracks integrity + HV qualification + High current test Rui s workshop, at CERN Phoenix (I) Cern N. Wauquier Lab Genova DONE DONE DONE Rui/Claudia Claudia Gemme Claudia Gemme Wings bending Fast polymerisation method 2 nd Electrical Test Test tracks integrity CERN Reception test Cable saver has to be loaded Genova DONE beginning of April Claudia Gemme Genova DONE 20/04/2012 Claudia Gemme CERN Between 23 rd and 24 th of April Daniel Dobos 7 Cable saver PP0 CERN (PH/DT) 24 th of April Francois-Xavier Nuiry 8 Flex cleaning with flange tool (Remove all grease) Drying &protection for shipment to SR1 CERN Nicole Wauquier Lab 24 th of April Francois-Xavier Nuiry 9 Cable saver PP0 CERN (PH/DT) Not necessary Francois-Xavier Nuiry 10 Glue flex on stave Calibrated thickness of glue, flex position is accurate CERN Between 10 th and 11 th of May Francois-Xavier Nuiry CERN leave test (stave on the handling 08/05/ Flex CERN Review 14 th of May Daniel Dobos 30 frame)

31 Feedback from Francois Xavier Some more feedback is needed from thermal stave (Flex batch1 + steva +heaters) and stave-0 (flex Batch2). My main comment is the design modification for Batch 3, I would need to test: -Addition of flat ears on the opposite side of the wings. This ears will be removed after gluing. -Removal of the microns coverlay on the to, but addition of a 25+25microns coverlay instead. This 2 modifs could be cancelled for the production, and in any case it does not change anything in the production time. Removal tool to unplug cable savers? 31

32 Feedback to layout From Susanne ok Better to change the naming convection on A side to have consistency both on A and C side? A side / Wing 4 C side / Wing 1 VCC/GND4, NTC1, HV1 VCC/GND4, NTC4, HV4 08/05/2012 Flex Review 32

33 Sapres 08/05/2012 Flex Review 33

34 161A2 Signal Resistance values (from connector to wings) (1) w16 w15 w14 w13 w12 w11 w10 w9 w8 w7 w6 w5 w4 w3 w2 w1 ckn 2,98 2,61 4,45 4,14 4,31 4,01 6,13 5,79 5,86 5,51 8,01 7,65 7,05 6,68 9,6 9,2 ckp 2,92 2,53 4,5 4,08 4,21 3,81 6,15 5,76 5,81 5,39 8,03 7,64 6,84 6,47 9,37 8,96 ck delta n wing 0,57 0,56 0,58 0,55 0,58 0,59 0,55 0,57 ck delta p wing 0,59 0,58 0,6 0,6 0,6 0,61 0,58 0,58 din 2,97 2,65 4,56 4,18 4,3 3,87 6,31 5,91 5,85 5,58 8,23 7,81 7,05 6,63 9,66 9,28 dip 3 2,54 4,54 4,13 4,19 3,77 6,25 5,76 5,71 5,39 7,99 7,61 6,86 6,49 9,4 8,97 di delta n wing 0,57 0,56 0,58 0,58 0,59 0,63 0,57 0,67 di delta p wing 0,58 0,57 0,6 0,58 0,6 0,63 0,59 0,66 dop 1,94 2,26 3,28 3,61 3,11 3,48 4,9 5,28 4,71 5,06 6,96 7,3 6,06 6,42 8,54 8,92 don 2,02 2,25 3,18 3,54 2,99 3,32 4,77 5,23 4,58 4,92 6,79 7,12 5,86 6,22 8,32 8,66 ntc 1,69 2,84 4,04 5,3 ntc ret 1,76 2,89 4,08 5,4 hv 0,62 0,76 0,92 1,05 1,15 1,29 1,46 1,6 1,81 1,97 2,17 2,36 2,68 2,87 3,08 3,29 A side Calcolato w16 w15 w14 w13 w12 w11 w10 w9 w8 w7 w6 w5 w4 w3 w2 w1 ckn 2,43 2,13 4,78 4,47 7,35 7,033 ckp 2,43 2,12 4,77 4,47 7,34 7,029 ck delta n wing 0,47 0,47 0,48 ck delta p wing 0,47 0,47 0,48 din 2,44 2,13 4,77 4,47 7,36 7,02 dip 2,44 2,13 4,77 4,46 7,35 7,02 di delta n wing 0,48 0,48 0,5 di delta p wing 0,47 0,48 0,5 dop 1,6 1,87 3,86 4,12 6,43 6,73 don 1,61 1,86 3,86 4,13 6,42 6,73 ntc 1,39 3,54 ntc ret 1,42 3,56 hv 0,86 1 2,02 2,16 3,01

35 w1 w2 w3 w4 w5 w6 w7 w8 w9 w10 w11 w12 w13 w14 w15 w16 Signal Resistance values (from connector to wings) (2) ckn 3,43 3,07 4,17 3,86 4,99 4,64 5,71 5,4 6,23 5,93 6,97 6,63 7,75 7,43 8,31 7,99 ckp 3,55 3,19 4,36 4,08 5,16 4,85 5,91 5,61 6,44 6,27 7,24 6,99 8,04 7,65 8,7 8,36 ck delta n wing 0,59 0,57 0,59 0,57 0,56 0,56 0,6 0,62 ck delta p wing 0,61 0,62 0,61 0,59 0,58 0,59 0,64 0,61 din 3,5 3,16 4,28 3,95 5,04 4,69 5,77 5,43 6,29 5,99 7,04 6,69 7,79 7,42 8,47 8,07 dip 3,62 3,29 4,31 4,06 5,23 4,86 5,86 5,55 6,52 6,17 7,25 6,87 8,17 7,74 8,75 8,41 di delta n wing 0,59 0,61 0,65 0,6 0,6 0,62 0,63 0,61 di delta p wing 0,6 0,62 0,65 0,61 0,61 0,63 0,63 0,62 dop 1,89 2,26 2,54 2,87 3,15 3,53 3,7 3,95 4,06 4,38 4,79 5,07 5,34 5,5 5,82 6,14 don 1,94 2,32 2,61 2,95 3,25 3,54 3,8 4,06 4,17 4,49 4,87 5,08 5,41 5,64 5,95 6,31 hv 0,77 0,92 1,64 1,78 2,8 2,94 ntc 1,44 2,52 3,6 4,78 ntc ret 1,44 2,56 3,6 4,79 hv 1,01 1,19 1,36 1,52 1,73 1,94 2,09 2,28 2,45 2,64 2,85 3,01 3,29 3,49 3,68 3,88 C side Calcolato w1 w2 w3 w4 w5 w6 w7 w8 w9 w10 w11 w12 w13 w14 w15 w16 ckn 2,71 3,03 5,01 4,71 7,26 6,95 ckp 2,72 3,03 5,02 4,72 7,27 6,96 ck delta n wing 0,48 0,47 0,5 ck delta p wing 0,48 0,47 0,5 din 2,76 3,08 5,03 4,72 7,33 7 dip 2,77 3,09 5,04 4,73 7,34 7,01 di delta n wing 0,49 0,48 0,51 di delta p wing 0,49 0,48 0,51 dop 1,88 2,19 3,53 3,77 5,44 5,7 don 1,89 2,2 3,53 3,77 5,44 5,69 ntc 1,25 ntc ret 1,22

36 Comparing measured R_VCC L1 with theoretical expectation geometrical dimension of L1 Shape : L=77mm Width =1.625 mm L=118.4 mm Width=2.08 mm L =14.38mm Width =10.9 mm resistivity Al (10E-5 Ohm mm) 2,82 resistivity Cu (10E-5 Ohm mm) 1,7 Max current (A) 2 Flex width d (mm) Thickness GND COPPER (mm) 0,005 Thickness GND ALUMINIUM (mm) 0,05 Thickness BOTTOM - Low Voltage (mm) 0,05 ALLUMINIO R(Ohm) Width(mm) length(mm) VCC1=GND ,9 14, ,08 118, , R_VCC L1 = Ohm while expected value was Ohm

37 R VCC and GND shape Calculated Between connector and via of first of 4 group of vias Wing 1: 0.06 Ohm Rtot = 0.12 Ohm Vdrop = 0.12 * 2 Amp. =0.24 V Wing 5: Ohm Rtot = Ohm Vdrop = V Wing 9: Ohm Rtot = Ohm Vdrop = V Wing 13 : Ohm Rtot = Ohm Vdrop =0.264 V In the calcolous it is not included the via resistance.