i.mx6ULL : BGA ball crack failures observed in Dye and Pry Test

[JayPatel_16792]

Hi NXP support team,

We have blocked in Mass production of our Gateways due to a very weird issue with i.MX6ULL part in the field. We feel that it is currently linked with either product handling or PCB assembly process.

We wanted some expert views on this to expedite the issue RCA, & need to communicate with some FAE or expert who can guide us with necessary steps to discover the Root-Cause & derive a concrete solution to avoid it in future production batches.

Problem Statement:

Few of our Production Test PASS boards are not working (booting) in the Field. Looking at initial debug logs, Uboot don’t getting PMIC I2C address & got stuck.
As a part of Physical analysis of Processor BGA solderability, we capture X-ray & conducted Dye & Pry test.
X-Ray has no clues relavent to this.
Attached is the Snap of Dye & Pry test report of i.mx6ULL processor BGA chip (Mfr PN # MCIMX6Y2CVM05AA) analysis. Test observations describing about “extremely thin crack failures on solder pads below the BGA balls”, and could be the probable cause for infant mortality of Failed Gateways.

Let us know if you need more information & even we can arrange a meeting to dip-dive the Dye & Pry Test report.
Hope to crack RCA with amicable solution ASAP!

[vicentegomez]

Please accept my ap[ologies for the late response

I will contact you by email, in order to request more information about the problem

Have a good day

Vicente