PC just caught fire

[Quartz]

Wonder if an unnoticed watercooling failure could cause one or more of those things and the case to start sparking.

@Viberious, were you watercooling? Or using an AIO / CLC?

there’s burns around the pci-e socket and obvious damage to the connector.

That would tally with @EsaT's thoughts upthread. I think there are other GPUs (RX 480 comes to mind) that have been caught pulling too much power from the PCIe slot. You were running the two PCIE power cables on different cables from the PSU, weren't you?

 

[EsaT]

That would tally with @EsaT's thoughts upthread. I think there are other GPUs (RX 480 comes to mind) that have been caught pulling too much power from the PCIe slot. You were running the two PCIE power cables on different cables from the PSU, weren't you?

Slot is used as current/power source by different VRM phases than PCIe connectors.
So properly working card should never overload it even without PCIe power cables.

 

[Quartz]

Slot is used as current/power source by different VRM phases than PCIe connectors.
So properly working card should never overload it even without PCIe power cables.

True, but if Viberious had just been using one cable for all 300W then an overload there might have caused problems on the slot.

 

[Viberious]

It was powered by 3 separate cable runs only thing I can think of is a water leak or something shorted. But it’s dead so I can’t change that. Just need to sort out a rebuild over the next week or three with the new parts.

 

[varkanoid]

Pictures please

 

[Purgatory]

@Viberious Sorry to hear that (be happy you were there to catch it and no one got hurt), did the fire start at the PCIE slot on the motherboard ? If so maybe the card was not pushed all the way in and started to short sadly. Also did you have seperate power cables to each PCIE power connector on the gpu coming from the PSU? Not the second cable with connector that comes off the main cable on each PCIE power cable.

Sadly a pc fire can easily happen if something is a miss and the PSU doesn't detect a direct short or an OCP situation. This is why I also warn people with PSU power supplies to make sure to get a good one and one that is right for their power use of their components. Your power supply was perfectly fine for the system you have power wise but sadly it didn't detect an issue and kept pushing power to the components that were not working right and started a fire.


@xPETEZx @Insolent mint See why I make sure people are careful now with PSUs and using right ones from day one and not risking using an underpowered psu maybe for the job, because fires can happen and the risk of electric shock even with a good brand PSU sadly and one with more than enough power for the system. But as we have seen the protection on PSUs can't be 100% trusted as in this case.

 

[Purgatory]

Well what an evening my gaming rig just caught fire, destroyed my MSI 3080 and Z590-E motherboard. Looks like i wont be gaming for a while.

Least the SSD / Memory / CPU are okay.

Better start looking for a new Z590 motherboard.

What brand and model was the motherboard mate and full model of the MSI 3080 card please? Just good information in case it happens to anyone else.

 

[BallistixOnZ490]

Hmmm from 100c+ vram to this.

Could have just been a single faulty card or motherboard though.

 

[BallistixOnZ490]

I just realized there could be another issue here.

While some high end PSUs have plenty of individual PCI-E outputs, they are often paired into rails.

My EVGA has 3 rows of 2 PCI-E outputs, it could be that even if using two cables, they need to go onto different rows.

Looking at a pic of the outputs on OP's PSU, it just has 6 PCI-E and CPU outputs all in a block, it could be that the outputs are shared in the same manner that a daisy chain splitter would be?

 

[wookiee87]

when you take it apart look closly at the motherboard and especially where the screw holes are, it sounds like the board earthered its self via the case which caused a catastropic short and took it and the gpu out, i'd be suprised if the other components are all well and good but testing will confirm this.

where did the sparks roghtly come from, start in that gerneral area and look for signs of damage to the board or the gpu, check your aio/clc for leaks as above too if you run that trype of cooling

 

[Mr Mcgoo]

i had a very similar issue with a gtx 1070 custom loop. randomly the system caught fire from the pci-e slot and system shutdown. after a couple days as i went to send off ek gpu block back to ekwb for refund, i noticed the screw holding the ports in place had rusted and it was then i found my issue. the plexi had cracked and thats where the leak came from. ekwb confirmed this and while it took 6 months they finally agreed to refund me for the gpu and block. thankfully asrock replaced mobo free of charge as a good will gesture.

 

[Quartz]

when you take it apart look closly at the motherboard and especially where the screw holes are, it sounds like the board earthered its self via the case which caused a catastropic short and took it and the gpu out, i'd be suprised if the other components are all well and good but testing will confirm this.

This can be caused by an insect getting between the board and the case.

 

[Alan McFall]

Holy crap, gutting. Stories like this reinforce my behavior of paying house insurance that I never claim on.. Hope you manage to get a replacement fast.

 

[Danny75]

i had a very similar issue with a gtx 1070 custom loop. randomly the system caught fire from the pci-e slot and system shutdown. after a couple days as i went to send off ek gpu block back to ekwb for refund, i noticed the screw holding the ports in place had rusted and it was then i found my issue. the plexi had cracked and thats where the leak came from. ekwb confirmed this and while it took 6 months they finally agreed to refund me for the gpu and block. thankfully asrock replaced mobo free of charge as a good will gesture.

That's the first thing I would check tbh.

 

[4k4uNuNu]

Holy crap, gutting. Stories like this reinforce my behavior of paying house insurance that I never claim on.. Hope you manage to get a replacement fast.

Also never leave electricals on when you aren't around.

Maybe miners have found a new way to heat their room. And the sparks can save on the lighting.

 

[xPETEZx]

@Viberious Sorry to hear that (be happy you were there to catch it and no one got hurt), did the fire start at the PCIE slot on the motherboard ? If so maybe the card was not pushed all the way in and started to short sadly. Also did you have seperate power cables to each PCIE power connector on the gpu coming from the PSU? Not the second cable with connector that comes off the main cable on each PCIE power cable.

Sadly a pc fire can easily happen if something is a miss and the PSU doesn't detect a direct short or an OCP situation. This is why I also warn people with PSU power supplies to make sure to get a good one and one that is right for their power use of their components. Your power supply was perfectly fine for the system you have power wise but sadly it didn't detect an issue and kept pushing power to the components that were not working right and started a fire.


@xPETEZx @Insolent mint See why I make sure people are careful now with PSUs and using right ones from day one and not risking using an underpowered psu maybe for the job, because fires can happen and the risk of electric shock even with a good brand PSU sadly and one with more than enough power for the system. But as we have seen the protection on PSUs can't be 100% trusted as in this case.

Not sure why I didn't get alerted to your @ mention. So just seeing this now by chance...
Also not sure how your point stands here? The OP had a RM1000 PSU, more than enough even by your standards to power the system he had. A quality brand unit too.
Schim reading the thread, sounds like the issue occurred at the PCIe slot, either due to overload or possibly liquid ingress there. So again, not sure how the PSU would be at fault here?
If this was a fault of the PSU... it kinda shows the opposite of what you have been saying. IE that even buying a quality PSU that is more than enough for the system you have can still lead to faults and failures.
Quiet how this shows the risk of electric shock I dont know either.

OP, you seem resigned to just having to pay to replace these parts? I would have thought you can either claim on warranty of the part that caused the fault, or at the very least claim on your home insurance?
I wish you luck getting the system back online. Perhaps share some pictures as it might help determine what caused the issue.

 

[Purgatory]

Not sure why I didn't get alerted to your @ mention. So just seeing this now by chance...
Also not sure how your point stands here? The OP had a RM1000 PSU, more than enough even by your standards to power the system he had. A quality brand unit too.
Schim reading the thread, sounds like the issue occurred at the PCIe slot, either due to overload or possibly liquid ingress there. So again, not sure how the PSU would be at fault here?
If this was a fault of the PSU... it kinda shows the opposite of what you have been saying. IE that even buying a quality PSU that is more than enough for the system you have can still lead to faults and failures.
Quiet how this shows the risk of electric shock I dont know either.

OP, you seem resigned to just having to pay to replace these parts? I would have thought you can either claim on warranty of the part that caused the fault, or at the very least claim on your home insurance?
I wish you luck getting the system back online. Perhaps share some pictures as it might help determine what caused the issue.

The reason I mentioned you was because you stated you never heard of a PSU causing a fire or not just shutting down if there was a problem. It was regarding the thread we spoke on before, showing that OCP and OPP doesn't always work on PSUS and the safety features can be hit and miss or set to high to the point they can cause a fire.

 

[xPETEZx]

The reason I mentioned you was because you stated you never heard of a PSU causing a fire or not just shutting down if there was a problem. It was regarding the thread we spoke on before, showing that OCP and OPP doesn't always work on PSUS and the safety features can be hit and miss or set to high to the point they can cause a fire.

Ok, but this isn't an example of that?
The fault occurred on the motherboard/PCIe slot.

I never said I don't believe a PSU can cause a fire. That for sure can and has happened. While again, its most likely that the fire is caused further down-stream from the PSU. Like using cheap aftermarket wires, PCIe risers or something else. Anything but the cheapest of PSUs should trip OCP before its own wires cause a fire. (Again... outside of an engineered situation... like maybe connecting a bunch of SATA-PCIe adapter to a single cable from the PSU... )
I was sceptical of how anybody would shock themselves or even kill themselves.
Getting a shock maybe... but youd need to be fiddling around with the PC while its ON. And the most basic of pre-caution which anybody should follow if they suspect an issue is to turn the thing off before fiddling round inside.
While I am sure it is technically possible, I just don't think its a scenario likely to happen. Sure you CAN be struck by lightning, but its not something anybody should be worrying about daily.

 

[Deleted member 77746]

Pictures or it didn’t happen.

 

[Jay85]

Wonder if an unnoticed watercooling failure could cause one or more of those things and the case to start sparking. Bit of liquid in the PCIe slot that melted for example.

This is what happened to me once, the flow meter shorted out which caused the wiring to heat up and then melt which inevitably melted other things around it which then caught fire but luckily the psu tripped and it never destroyed anything else other than a fan cable.