Jump to content

PCIE devices not showing up in BIOS after "hot" swap - did I break two 3080s?

coolzach

I have a dual boot system with a phanteks revolt 1000w power supply, and did not power off the power supply at the wall when changing taking my graphics cards in and out. The system itself was powered off (clicked shutdown in Windows) but my Unraid dual boot was still powered on, so the power supply stayed on. I took the graphics cards out of my Windows machine to reapply the thermal pads and thermal paste, then replaced them. The system rebooted fine and worked as normal, but I put my graphics cards in in the wrong order and couldn't close the side panel, so I had to take them out again and rearrange them. The system booted fine after changing the thermal pads, so I don't think I broke them doing that.

 

This is when they stopped showing up in BIOS, and my NVME drive is also not showing up. I have tried a different motherboard and a different power supply, cleared CMOS, tried each device individually and tried each PCIE slot individually, all to no avail. I have a fresh Windows install on the system (on a SATA SSD) and none of the devices show up in device manager, but I am prepared to troubleshoot if there's anything I can do in Windows. Graphics cards both still turn LEDs on and fans spin, that is the only sign of life.

 

It really seems like all my devices are borked, but I want to reach out to the internet for help before I start funeral arrangements.

 

Specs:

Phanteks Revolt 1000W

Windows 64 bit -

i9 9900K

32 GB Corsair 3600 MHz ram

Gigabyte Z390 Designare

EVGA RTX 3080 FTW3

ZOTAC RTX 3080 Trinity

Unraid -

i9 9900KF

32 GB Corsair 3600 MHz ram

ASRock H310 cm

Also tested on Gigabyte Aorus Pro Z390 and Corsair CX750M

 

Newest BIOS versions on both mobos

Link to comment
Share on other sites

Link to post
Share on other sites

Never... Ever... EVER hotswap PCI, PCIx or NVME (just pcie by annother name)

There are certain storage controllers designed for that, but thats not your standard PCIe.

That said... You can only cold boot and wipe the bios to find out.

1 - pull the cards.
2 - Wipe the bios
3 - turn it on (does it stay on? ok keep going)
4 - Does the internal GPU post? (well if it does...)
5 - turn it off, and put in ONE card (does it post?)
6 - THE OTHER CARD then (does it post?)
7 - If the IGPU gets you to Windows, but it cant see the PCIe GPUs Then there are 2 possibilities.
A.) Either the GPUs are for whatever reason: No longer working.
B.) The PCI ports are no longer working.
(pending the BIOS was wiped, then the BIOS is not the problem.)

You need a second board to make sure of the cards status. (broken, or not)
Also to check the PCIe port status. (broken or not, by deductive reasoning)

My bet is actually less catastrophic.
I think you MAY have damaged your board, before id think youve damaged the cards.

Link to comment
Share on other sites

Link to post
Share on other sites

# Extra Note: Thermal module failure would result in a [short-post and sudden power-off], but not in a [no-post] nor a [non-detection.] in windows.

AND If you shorted a component on the local gpu bus, then you can trace it with a multi meter.

Link to comment
Share on other sites

Link to post
Share on other sites

Yes, my laziness caused the problem, I will be a lot more careful with my PC in the future. I am currently in Windows (on a new install on a SATA SSD) and I have tried one GPU at a time in each individual slot on two boards, all to no avail. The NVME has been tried in 3 different slots on two different boards. I have cleared the CMOS on both boards and they are on the newest BIOS updates, should I try a roll back?

 

The system was at least powered off, just with the power supply still on.

Link to comment
Share on other sites

Link to post
Share on other sites

29 minutes ago, coolzach said:

The system was at least powered off, just with the power supply still on.

This still can cause some issues, its always better to have the PSU turned off fully ("better safe than sorry), as some power is still going to the motherboard and other devices, just not as much as when the system is completely on. 

Chicago Bears fan, Bear Down

 

Link to comment
Share on other sites

Link to post
Share on other sites

46 minutes ago, coolzach said:

Yes, my laziness caused the problem, I will be a lot more careful with my PC in the future. I am currently in Windows (on a new install on a SATA SSD) and I have tried one GPU at a time in each individual slot on two boards, all to no avail. The NVME has been tried in 3 different slots on two different boards. I have cleared the CMOS on both boards and they are on the newest BIOS updates, should I try a roll back?

 

The system was at least powered off, just with the power supply still on.

If the nvme doesn't work anymore in all those systems it's simply dead. Try your gpu in other working systems too (one at a time make sure the psu is sufficient) and HOPE you didn't kill them too.

 

Pcie is hotswap CAPABLE but it has to be specifically SPECIFIED that the board is hotswap COMPATIBLE. That and the devices are ok with hotswapping.

Link to comment
Share on other sites

Link to post
Share on other sites

TBH it's never a good idea to start swapping stuff with the PSU still on, if it is that means it's still supplying power to the system and whatever else is connected to it.

 

Always make sure it's off before doing any of that kind of work.

"If you ever need anything please don't hesitate to ask someone else first"..... Nirvana
"Whadda ya mean I ain't kind? Just not your kind"..... Megadeth
Speaking of things being "All Inclusive", Hell itself is too.

 

Link to comment
Share on other sites

Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×