1 / 8

Experience with C4 flip-chip

Experience with C4 flip-chip. …..from the point of view of chip design & testing and o n behalf of the GBT project Motivation Implications for design & testing Controlled Collapse Chip Connection. Motivation: GBT-SERDES (prototype GBTX). dOut [29:0]. DES. Digital Core. Parallel

lanai
Download Presentation

Experience with C4 flip-chip

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Experience with C4 flip-chip • …..from the point of view of chip design & testing and on behalf of the GBT project • Motivation • Implications for design & testing • Controlled Collapse Chip Connection ACES, March 10th 2011

  2. Motivation: GBT-SERDES (prototype GBTX) dOut [29:0] DES Digital Core Parallel Out/ BERT rxDataValid 120 120 Serialinput rxClock40 rxClock160 4.8 Gbit/s serial data More details from Paulo Moreira today Phase Shifter ClkOut3 60-bit I/O bus at 160 MHz …….. 169 pads (but many more for GBTX) ClkOut2 ClkOut1 ClkOut0 Clock Generator Clockreference rxRdy Control Logic txRdy I2C JTAG AUX[n:0] RST SER dIn [29:0] Parallel In/ PRBS 120 txDataValid Serial out 120 txClock40 txClock160 PROMPT Power On RESET reset ACES, March 10th 2011

  3. Motivation: GBT-SERDES (prototype GBTX) • High speed lines: minimise inductance • Direct powering to blocks • Direct cooling on back-side (although coverplate is material eg Cu) • Die size is pad-limited; extra silicon (400 I/Os in GBTX) • Mount on BGA package ACES, March 10th 2011

  4. Technical Information IBM 130nm library - C4 pad only in LM metal stack - pad & drivers separate blocks IBM deposit bumps on wafers; - more masks => extra cost - MPWs: split between C4 & wirebond Size/pitch = “5 on 10” IBM standard (recommended for yield) = 125/250mm Package vendor recommended High-Temperature Solder (97Pb3Sn) for reliability 146mm ACES, March 10th 2011

  5. Design Implications Floorplan at start of design Communicate early with package/hybrid designers -> we iterated a few times -> placement of signal/power pads Routing by hand is difficult -> we used the CERN-VCAD ‘digital design flow’ (analog blocks incorporated as black-box abstracts) Solve ALL DRCs (even if not physical!) ACES, March 10th 2011

  6. Layout & Bumped Die ACES, March 10th 2011

  7. Testing Results Package designed by Endicott Interconnect Assembly by Endicott No cover-plate mounted No C4 problems found on chips tested so far Chip testing issues: Need complex package/hybrid immediately Package design was significant part of project cost Chip orientation is bad for radiation testing attenuation of 10keV X-rays in CERN facility back-grinding needed for heavy ion tests ACES, March 10th 2011

  8. Summary C4 has many performance advantages Standard industrial process Requires careful planning: floorplan, package/hybrid Implies extra financial cost upfront Extra cost may be offset by savings elsewhere ACES, March 10th 2011

More Related