BIOS flash gone wrong - CPU or component damage possible?

[voodoodrul]

A BIOS flash gone horribly wrong due to flaky RAM left me with a dead motherboard. The machine hard locked during writing of the main block. The subsequent boot attempts failed, obviously.

Anyone with intimate knowledge of VRM circuits chime in on what follows.

To everyone else I ask this: "Should I be overly paranoid about possible component damage due to a BIOS flash gone wrong?"

My question from here is more of a technical one. I am approaching this from an engineering standpoint rather than one of previous experience.

During the POST attempt, voltages are set according to the relevant BIOS settings. Without a functioning BIOS, or one in a partially corrupt state, is it possible for the VRM to overvolt the processor or other components on the motherboard?

I consider that VIDs stored in the BIOS and directed by the BIOS are references to the voltage table entries and that the VRM itself takes the VID code and applies that voltage. Lets assume that the BIOS never even got that far and no VID code was issued. The VRM probably would never power up the CPU to begin with.

If however the VRM was issued a corrupt VID, it wouldn't reference anything in the table and the VRM would likely also stop short of powering up the CPU.

If the VRM was issued an undesired VID, then we might have a problem if that VID was in the VRM's table and resulted in a high overvoltage scenario.

 

[myocardia]

While anything's possible, it's not very likely that you've done any damage. Just call EVGA, and have them send you a new CMOS chip.

 

[LittleNemoNES]

I freaked when I had a bad flash with my old p5wDH
I bought a new chip from bios man and my machine is up and running (sold!)

Ask evga for a new chip or buy one from bios man

 

[myocardia]

Originally posted by: gersson
Ask evga for a new chip or buy one from bios man

Oh yeah, I had forgotten about BIOS man. He'll even put whatever BIOS on it that you want. Here's the link.

 

[Modelworks]

Originally posted by: voodoodrul
A BIOS flash gone horribly wrong due to flaky RAM left me with a dead motherboard. The machine hard locked during writing of the main block. The subsequent boot attempts failed, obviously.

Anyone with intimate knowledge of VRM circuits chime in on what follows.

To everyone else I ask this: "Should I be overly paranoid about possible component damage due to a BIOS flash gone wrong?"

No.
The reason is that when you first power on the pc the actual voltage supplied to the processor is not read from the bios. It is determined by the motherboards design and is not editable in the bios.
Once the cpu has started and is ready then it looks to load the bios and change the default voltage.

If it was not done this way then using wrong settings when trying to overclock would kill the board, requiring a new bios every time you used a bad setting.

 

[bryanW1995]

this is good info. I knew that I HADN'T destroyed any mobos over the years, now I know WHY! 🙂

 

[voodoodrul]

Originally posted by: Modelworks

No.
The reason is that when you first power on the pc the actual voltage supplied to the processor is not read from the bios. It is determined by the motherboards design and is not editable in the bios.
Once the cpu has started and is ready then it looks to load the bios and change the default voltage.

If it was not done this way then using wrong settings when trying to overclock would kill the board, requiring a new bios every time you used a bad setting.

Sounds like some solid reasoning. It would make sense that the VRM circuit would have to decide initial conditions and once the CPU is initialized, it could ask for overriding BIOS settings. Thanks Modelworks!