SDRAM module "zero window" breakage

sinara-hw / Kasli

Kasli is a powerful FPGA carrier, capable of controlling 12 Eurocard extension modules.

Other

16 stars 1 forks source link

SDRAM module "zero window" breakage #27

Closed sbourdeauducq closed 5 years ago

sbourdeauducq commented 5 years ago

This happened after shipping on a Kasli (mounted in a subrack) that was working before, without touching the firmware:

Initializing SDRAM...
Read leveling scan:
Module 1:
00000000000000000000000000000000
Module 0:
00000000000111111111111000000000
Read leveling: Zero window: 1: 0-0 (0)
SDRAM initialization failed

Has anyone seen this issue before?

marmeladapk commented 5 years ago

We may have seen something similar when the FPGA wasn't soldered correctly.

sbourdeauducq commented 5 years ago

Why did that affect the SDRAM in particular? Which solders were broken?

marmeladapk commented 5 years ago

Part of the FPGA was lifted and some FPGA pins (particularly lower left quadrant) didn't make proper contact. It was easy to see visually, FPGA wasn't aligned horizontally with the board. We didn't check pin by pin as it was obvious that the FPGA had to be resoldered.

sbourdeauducq commented 5 years ago

That failing Kasli also developed breakage on EEMs 9 and 10.

jordens commented 5 years ago

I have one where SFP0 is not receiving and where this SDRAM zero window happens sporadically. It's also a speed grade 3.

hartytp commented 5 years ago

@gkasprow what do we do about this?

gkasprow commented 5 years ago

This happened on a PCB series where FPGA was soldered later on manually. There is nothing to do with PCB design. The tests were probably not sufficient or were not conducted at all.