A modules misgraded as B

[drberry85]

Hey Greg and John,

It looks like LessWeb is grading grade A modules as grade B because it's looking at the total number of bad bump bonds instead of the bad bumps per ROC. Examples are M-B-4-44 and M-G-2-48. These are grade A modules, but it looks like LessWeb is saying they are grade B because they have 44 bad bumps and 83 bad bumps, respectively. While this does place a single ROC pass the 0.1% threshold, these bumps are distributed across multiple ROCs. You can check in MoReWeb and see that M-B-4-44 and M-G-2-48 should be grade a modules.

Doug

[jstupak]

These grades look correct to me. ROC 13 on M-B-4-44 has 48 (44+4) defects, which is over the 1% threshold, so B is correct grade.

[lantone]

The DB/MoReWeb discrepancy comes from the bump bonding analysis done in MoReWeb [1]. In the latest commits from @gfunk723 this issue has been addressed. Once we do an update on Purdue's side (in the coming days) everything should be consistent.

Hey, I just had a thought - MoReWeb is like the cross-check analysis that never gets officially used but helps the mainline analysis validation its analysis code!

[1] http://www.physics.purdue.edu/cmsfpix/MoReWeb/Results/REV001/R001/M-B-4-44_FPIXTest-m20C-FNAL-160125-1241_2016-01-25_12h41m_1453747305/QualificationGroup/ModuleFPIXTest_m20_1/Chips/Chip13/TestResult.html#CMSPixel_QualificationGroup_Fulltest_Chips_Chip_BumpBonding_TestResult_BumpBonding

[gneeser]

With the new update of MoReWeb a few days back, has this discrepancy been resolved?

[lantone]

yes indeed! 44 = 44:

MoReWeb: http://www.physics.purdue.edu/cmsfpix/MoReWeb/Results/REV001/R001/M-B-4-44_FPIXTest-m20C-FNAL-160125-1241_2016-01-25_12h41m_1453747305/QualificationGroup/ModuleFPIXTest_m20_1/Chips/Chip13/TestResult.html#d

LessWeb: http://www.physics.purdue.edu/cmsfpix/Submission_p/summary/summaryFull.php?name=M-B-4-44

[gneeser]

Excellent!