Topic: Computer Caught Fire! (Read 2118 times)
Friends don't let friends use Gigashyte.
Who said btc isn't hot?
God, be safe, you should move your mining rig to underground.
Hope that is not normal. I don't want to see in the news "bitcoiners are burning all the cities" 
WE ARE GOING TO BURN ALL THE CITYS DOwn !!!!
Anarchy !@!!!!
Well as much as a group of nerds with video cards will achieve ..lol
LOL we had a gigabyte motherboard at work have that happen, except it burnt directly through the PCB. It was still ammusing.
It was replaced with another gigabyte motherboard (RMA - but upgraded from a 1157 to a 1155 motherboard and a new chip, so it was a nice upgrade for a 2 year old MB).
No troubles with any other gigabyte stuff so far.
It was replaced with another gigabyte motherboard (RMA - but upgraded from a 1157 to a 1155 motherboard and a new chip, so it was a nice upgrade for a 2 year old MB).
No troubles with any other gigabyte stuff so far.
Unfortunately, this does happen once in a while. I had the exact same thing happen to my First AsRock P55 Extreme board but the replacement has been running fine for years now and I've abused it pretty bad lol. I'd RMA it and grab a multimeter to test the PSU. Hopefully it didn't fry your other components when it went.
I have a PSU tester but I've got a 750 watt power supply test good but nothing I plug into it will work. I'm a little skeptical of my PSU tester. How do I test w/a multimeter?
You just did test it, If nothing it plugs into works, It does not work.
Unfortunately, this does happen once in a while. I had the exact same thing happen to my First AsRock P55 Extreme board but the replacement has been running fine for years now and I've abused it pretty bad lol. I'd RMA it and grab a multimeter to test the PSU. Hopefully it didn't fry your other components when it went.
I have a PSU tester but I've got a 750 watt power supply test good but nothing I plug into it will work. I'm a little skeptical of my PSU tester. How do I test w/a multimeter?
Comp deaths like this are not uncommon. I've had many a brand spanking new motherboard die during a 24 hour burn in. Motherboards dying are pretty tame, bit of a smell.
But compares to a fully loaded KW power supply exploding while your head is in the case looking for fan vibration -_-
But compares to a fully loaded KW power supply exploding while your head is in the case looking for fan vibration -_-
Hope that is not normal. I don't want to see in the news "bitcoiners are burning all the cities"
Unfortunately, this does happen once in a while. I had the exact same thing happen to my First AsRock P55 Extreme board but the replacement has been running fine for years now and I've abused it pretty bad lol. I'd RMA it and grab a multimeter to test the PSU. Hopefully it didn't fry your other components when it went.
Do the AMD APUs have separate vcore for the iGPU? Usually the whole CPU only has a single voltage but either way it looks like the stress caused some of your vcore switching MOSFETs to die catastrophically.
Based on the fact that the vcore regulators don't have any kind of heatsinking I imagine it was probably a cheaper board. Usually cheaper boards only use the minimum for the CPU regulator (usually only 2-3 regulator phases) but enthusiast grade boards will have 6-9 phases or more.
Edit - just looked up the board, looks like it has a pretty decent cpu regulation layout so it was likely just an unlucky failure
Based on the fact that the vcore regulators don't have any kind of heatsinking I imagine it was probably a cheaper board. Usually cheaper boards only use the minimum for the CPU regulator (usually only 2-3 regulator phases) but enthusiast grade boards will have 6-9 phases or more.
Edit - just looked up the board, looks like it has a pretty decent cpu regulation layout so it was likely just an unlucky failure
well that component is part of the VRM for the CPU, which part of the CPU can be the tough part to figure out. If its a phase for the cores, RAM or integrated graphics.
Was the CPU / RAM or integrated GPU oc'd? the phases could have been pumping more current than designed if so.
Was the CPU / RAM or integrated GPU oc'd? the phases could have been pumping more current than designed if so.
Nothing oc'd (although the board supposedly supports easy oc with 3 oc levels).
yeah, Im gonna go out on a limb and say the mfr skimped somewhere on the RAM / integrated GPU VRM current handling capabilities(if scrypt mining both of those would be heavily under load, if just sha256 then the integrated GPU would be)
Or, alternatively, one of those mosfets was just defective to begin with.
I did have Litecoin running. I was actually catching up the Litecoin blockchain when it caught fire. I'm wondering if the integrated mining was set to start on startup; could have done it. I'm going to stop mining using the GPU; 80 MH/s isn't worth this kind of headache.
Hope you solve ASAP. I had lot of Gigabyte mobo at home and sold many, but no problem.
Maybe a bad stock of some boards.
Maybe a bad stock of some boards.
well that component is part of the VRM for the CPU, which part of the CPU can be the tough part to figure out. If its a phase for the cores, RAM or integrated graphics.
Was the CPU / RAM or integrated GPU oc'd? the phases could have been pumping more current than designed if so.
Was the CPU / RAM or integrated GPU oc'd? the phases could have been pumping more current than designed if so.
Nothing oc'd (although the board supposedly supports easy oc with 3 oc levels).
yeah, Im gonna go out on a limb and say the mfr skimped somewhere on the RAM / integrated GPU VRM current handling capabilities(if scrypt mining both of those would be heavily under load, if just sha256 then the integrated GPU would be)
Or, alternatively, one of those mosfets was just defective to begin with.
well that component is part of the VRM for the CPU, which part of the CPU can be the tough part to figure out. If its a phase for the cores, RAM or integrated graphics.
Was the CPU / RAM or integrated GPU oc'd? the phases could have been pumping more current than designed if so.
Was the CPU / RAM or integrated GPU oc'd? the phases could have been pumping more current than designed if so.
Nothing oc'd (although the board supposedly supports easy oc with 3 oc levels).
well that component is part of the VRM for the CPU, which part of the CPU can be the tough part to figure out. If its a phase for the cores, RAM or integrated graphics.
Was the CPU / RAM or integrated GPU oc'd? the phases could have been pumping more current than designed if so.
Was the CPU / RAM or integrated GPU oc'd? the phases could have been pumping more current than designed if so.
wow! hope you and your property are safe... looks like some defective components.
Yep, I caught it as it was happening, no big issue (except for about 5 seconds of abject terror). I'm just trying to figure out which components were faulty.
wow! hope you and your property are safe... looks like some defective components.
Here's my build:
Power: CORSAIR Enthusiast Series TX650 V2 650W ATX12V v2.31/ EPS12V v2.92 80 PLUS BRONZE Certified Active PFC High Performance Power Supply
Motherboard: GIGABYTE GA-F2A55M-HD2 FM2 AMD A55
APU: AMD A10-5800K Trinity 3.8GHz FM2 100W Quad-Core Desktop APU (CPU + GPU) with DirectX 11 Graphic AMD Radeon HD 7660D
Video Card: SAPPHIRE 100314-5L Radeon HD 6870 1GB GDDR5 PCI Express Video Card (One of the rebranded 5870s)
I was mining on the 6870 @ 950 MHz with around 400 MH/s using GUIMiner
I was mining on the APU with around 80 MH/s
Saturday evening, this happened to my motherboard:
These are two chips labeled DEQ1 and DEQ2 just above the CPU socket.
I've RMA'd the board, but I was only using 1 video card (two if you count the onboard one). One other thing, my power supply would run even if the power switch was switched to off, so maybe my power supply is doing something funky. I'm wondering wtf went wrong here, this build is only 4 months old and has been running just about nonstop since then with no issues.
Power: CORSAIR Enthusiast Series TX650 V2 650W ATX12V v2.31/ EPS12V v2.92 80 PLUS BRONZE Certified Active PFC High Performance Power Supply
Motherboard: GIGABYTE GA-F2A55M-HD2 FM2 AMD A55
APU: AMD A10-5800K Trinity 3.8GHz FM2 100W Quad-Core Desktop APU (CPU + GPU) with DirectX 11 Graphic AMD Radeon HD 7660D
Video Card: SAPPHIRE 100314-5L Radeon HD 6870 1GB GDDR5 PCI Express Video Card (One of the rebranded 5870s)
I was mining on the 6870 @ 950 MHz with around 400 MH/s using GUIMiner
I was mining on the APU with around 80 MH/s
Saturday evening, this happened to my motherboard:
These are two chips labeled DEQ1 and DEQ2 just above the CPU socket.
I've RMA'd the board, but I was only using 1 video card (two if you count the onboard one). One other thing, my power supply would run even if the power switch was switched to off, so maybe my power supply is doing something funky. I'm wondering wtf went wrong here, this build is only 4 months old and has been running just about nonstop since then with no issues.
Jump to: