820-2849 No POST

[ayilm1]

Full disclosure, this thread is also on advanced reworks but it seems there's not much activity there these days. I have an 820-2849 here that won't POST. Full symptoms:

• No chime on case speaker.
• Fans spin up and stay on. Doesn't power cycle on its own.
• Fans run high if trackpad is disconnected validating SMC function.
• No absent RAM beeps when RAM is removed.
• Mighty mouse lights up when plugged into USB port.
• PM_PCH_PWRGD high. Probing CPUIMVP_PGOOD with anything, even a disconnected multimeter lead however causes the board to turn off (fans stop). I thought R2850 might be floating, but value is good and verified at 10k. PP3V3_S0 source for pull-up is also good.
• PCH_CLK25M only runs briefly at startup before stopping.
• PPVCORE_S0_CPU holds at 1V (0.9968V) and doesn't transition, which is why I believe it's failing POST.
• CPUIMVP_VID[6..0] is 0101000 which is consistent with the IMVP6.5 spec for a 1.0V VRM output.

The only things connected in this setup are the case speaker and magsafe board (and trackpad if I don't want the fans to spin high). I have not tried swapping U6100 yet, mostly because I don't have a part that can just drop in. Closest sibling boards I have are 2850s which are unlikely to be compatible and swapping blindly may do more harm than good. I guess I wanted to rule out everything else first but I've kind of hit a brick wall. Anyone else seen similar behaviour? TIA

[piernov]

Re: 820-2849 No POST

Well I guess everyone moved to rossmanngroup forums . :°

What's the history of the board? Liquid damage, physical damage…?

I guess your USB mouse always lights up even when USB controller is not initialized, as it shouldn't light up without RAM.
It can be EFI corruption so you can always try the 820-2850 SPI ROM on this board, they are very similar so it might work.

[ayilm1]

Re: 820-2849 No POST

What's the history of the board? Liquid damage, physical damage…?

Thanks for the reply. Unfortunately this has been sitting and gathering dust for a good while now so history on the board is a mystery. There are no visible signs of ingress though and the board is actually in very good condition. No contamination aside from flux residue in the areas I've already checked/reworked.

I guess your USB mouse always lights up even when USB controller is not initialized, as it shouldn't light up without RAM.
It can be EFI corruption so you can always try the 820-2850 SPI ROM on this board, they are very similar so it might work.

This is very valuable information actually. I didn't know removing RAM should prevent the host controller from enabling. I'm fairly certain I checked the USB function with RAM installed so I'll check again with it removed.

I'll also give swapping the SPI flash a quick test too. Thinking about it, there shouldn't be anything that can get damaged. The VRM is configured using resistive straps on the CPU, and I'm pretty sure all other low level config is handled by the SMC firmware, which isn't stored on the SPI flash. At the very least, I should see some change in behaviour which will hopefully tell me if I'm on the right track. I'll test this out as well and post back. Thanks for the input!

[MobileXT]

Re: 820-2849 No POST

It could be a Bios Issue

[ayilm1]

Re: 820-2849 No POST

Thanks for the insight guys. It looks like the mouse still turns on without RAM. I also tried installing U6100 from an 820-2850. Symptoms are unchanged. I'm leaning toward the CPU being dead on this board.