Computer crashes -- psu issue?

[79wjd]

My desktop will randomly crash while I'm playing a game and then won't turn on for a couple of minutes. Sometimes it happens after I've been playing a game (dota 2) for a couple hours and other times I can go significantly more time without it happening. So far it's happened three times. The PSU feels warm to the touch -- not hot but pretty warm I'd say. If I press on the top of the PSU then I hear a kind of static-y noise which I only hear the first couple of times and then it goes away. 

 

I'm assuming it's some PSU related issue, as it doesn't appear to be a cpu/gpu stability issue, not to mention any cpu/gpu/memory related issue wouldn't likely cause a failure to turn back on for a few minutes.

 

Specs: 

• 4670k @ 4.3ghz 
• MSI Z87m Gaming
• GTX 780
• RM650

@STRMfrmXMN @Energycore @Comic_Sans_MS

 

[STRMfrmXMN]

12 minutes ago, djdwosk97 said:

My desktop will randomly crash while I'm playing a game and then won't turn on for a couple of minutes. Sometimes it happens after I've been playing a game (dota 2) for a couple hours and other times I can go significantly more time without it happening. So far it's happened three times. The PSU feels warm to the touch -- not hot but pretty warm I'd say. If I press on the top of the PSU then I hear a kind of static-y noise which I only hear the first couple of times and then it goes away. 

 

I'm assuming it's some PSU related issue, as it doesn't appear to be a cpu/gpu stability issue, not to mention any cpu/gpu/memory related issue wouldn't likely cause a failure to turn back on for a few minutes.

 

Specs: 

• 4670k @ 4.3ghz 
• MSI Z87m Gaming
• GTX 780
• RM650

@STRMfrmXMN @Energycore @Comic_Sans_MS

 

Does the issue go away if you take the GPU out and run on the iGPU? If so then I'd assume a PSU issue and you can just RMA your PSU with Corsair.

[Guest]

Check event viewer?

[79wjd]

8 minutes ago, STRMfrmXMN said:

Does the issue go away if you take the GPU out and run on the iGPU? If so then I'd assume a PSU issue and you can just RMA your PSU with Corsair.

I've yet to figure out what exactly causes it to happen -- sometimes it shuts down after two hours of gaming and other times it needs more like six hours. So it's not really predictable enough to do any kind of testing like that. 

6 minutes ago, deXxterlab97 said:

Check event viewer?

There is a critical event -- kernal-power, event id 41 -- that occured right about when my computer crashed. Although I don't see any other critical events from earlier today (when it also shut down with the same set of conditions).

 

I think all crash events also started occuring after Windows downloaded the Creators update.

[Guest]

Just now, djdwosk97 said:

I've yet to figure out what exactly causes it to happen -- sometimes it shuts down after two hours of gaming and other times it needs more like six hours. So it's not really predictable enough to do any kind of testing like that. 

There is a critical event -- kernal-power, event id 41 -- that occured right about when my computer crashed. Although I don't see any other critical events from earlier today (when it also shut down with the same set of conditions).

yeah kernal power are usually psu related

[STRMfrmXMN]

4 minutes ago, djdwosk97 said:

I've yet to figure out what exactly causes it to happen -- sometimes it shuts down after two hours of gaming and other times it needs more like six hours. So it's not really predictable enough to do any kind of testing like that. 

There is a critical event -- kernal-power, event id 41 -- that occured right about when my computer crashed. Although I don't see any other critical events from earlier today (when it also shut down with the same set of conditions).

 

I think all crash events also started occuring after Windows downloaded the Creators update.

Have you checked to see if your motherboard VRMs are overheating? Perhaps reapply thermal paste on your 780? It's getting a bit old now and might need it.

[79wjd]

13 minutes ago, STRMfrmXMN said:

Have you checked to see if your motherboard VRMs are overheating? Perhaps reapply thermal paste on your 780? It's getting a bit old now and might need it.

I ran Heaven for about ten minutes and gpu temps slowly climbed up to 80ªc, which is perfectly normal. Then I ran Prime95 and again, nothing is overheating. 

[STRMfrmXMN]

Just now, djdwosk97 said:

I ran Heaven for about ten minutes and gpu temps slowly climbed up to 80ªc, which is perfectly normal. Then I ran Prime95 and again, nothing is overheating. 

80C for a 780? I've built I think one PC with a 780 and it was rare it breached 75C with the fan curve set to go to about 50% at 70C. Perhaps replace the paste for the heck of it? Of course I'd take the GPU out of the system and see if it crashes before going to the trouble of doing the paste swap.

[79wjd]

1 minute ago, STRMfrmXMN said:

80C for a 780? I've built I think one PC with a 780 and it was rare it breached 75C with the fan curve set to go to about 50% at 70C. Perhaps replace the paste for the heck of it? Of course I'd take the GPU out of the system and see if it crashes before going to the trouble of doing the paste swap.

80ºc is pretty normal for my 780 and my fan curve (that I've been using for years) -- 25% until 40ºc, 40% until 60ºc, 50% until 75ºc, and then 60% at 85ºc.

[STRMfrmXMN]

Just now, djdwosk97 said:

80ºc is pretty normal for my 780 and my fan curve (that I've been using for years) -- 25% until 40ºc, 40% until 60ºc, 50% until 75ºc, and then 60% at 85ºc.

Hmm, OK.

 

Well I guess just take the GPU out and see what happens. The RM650 is a pretty decent PSU so I'd be surprised if it started degrading that quickly unless it were exposed to a ton of heat without enough airflow for the fan to intake.

[79wjd]

12 minutes ago, STRMfrmXMN said:

Hmm, OK.

 

Well I guess just take the GPU out and see what happens. The RM650 is a pretty decent PSU so I'd be surprised if it started degrading that quickly unless it were exposed to a ton of heat without enough airflow for the fan to intake.

I'll try taking out the GPU eventually, but since I can't consistently reproduce the problem -- short of gaming for a few hours (at least as of now) -- removing the GPU won't tell me much. Tomorrow I'll try running Heaven and P95 to try and create a reproducible set of circumstances. 

 

But airflow shouldn't be an issue. I'm also wondering if the Creators Update has something to do with it -- there have been three crashes so far and all of them have been in the last few days. Two of them have definitely been after the CU, but the first one may have been before or right after it -- I'm not exactly sure when the first crash happened. So, I may try doing a W7 install and seeing if the issue occurs there -- but again, I need to find some way of reproducing the issue consistently. 

[79wjd]

@STRMfrmXMN @deXxterlab97

 

It looks like it's the the PSU, for all of the reasons mentioned before, but I also noticed that the power button on the motherboard (that's lit up as long as the PSU is providing power -- even when the system is shut down) is NOT lit up after the crash occurs (and thus the psu is not providing any power, so the system won't turn on). Previously, it would light up again if I waited like ten minutes (I assume for the PSU to cool off) and the system would turn on. So I think the PSU was overheating. Although, now I've waited like a half hour and the light on the motherboard still hasn't turned back on (and I still can't power up the system). So I think the PSU might be completely dead now.

 

I grabbed a very old PSU to see if the motherboard power button would light up, and it does (although I can't really test much else with that PSU because it's a low wattage and very old unit (430w w/ 230w over the 12v rail and not even 80+) plus the EPS power connector is blocked by my AIO and I really don't feel like taking that out (especially since i'm pretty confident my RM650 is the cause of the problem).

 

I emailed Corsair support on Monday, but I haven't heard back from them. 

 

I also forgot that my motherboard only has HDMI/DP outputs (and I'm using an old monitor), so I can't really test the iGPU only, and I don't really have 

[Guest]

32 minutes ago, djdwosk97 said:

@STRMfrmXMN @deXxterlab97

 

It looks like it's the the PSU, for all of the reasons mentioned before, but I also noticed that the power button on the motherboard (that's lit up as long as the PSU is providing power -- even when the system is shut down) is NOT lit up after the crash occurs (and thus the psu is not providing any power, so the system won't turn on). Previously, it would light up again if I waited like ten minutes (I assume for the PSU to cool off) and the system would turn on. So I think the PSU was overheating. Although, now I've waited like a half hour and the light on the motherboard still hasn't turned back on (and I still can't power up the system). So I think the PSU might be completely dead now.

 

I grabbed a very old PSU to see if the motherboard power button would light up, and it does (although I can't really test much else with that PSU because it's a low wattage and very old unit (430w w/ 230w over the 12v rail and not even 80+) plus the EPS power connector is blocked by my AIO and I really don't feel like taking that out (especially since i'm pretty confident my RM650 is the cause of the problem).

 

I emailed Corsair support on Monday, but I haven't heard back from them. 

 

I also forgot that my motherboard only has HDMI/DP outputs (and I'm using an old monitor), so I can't really test the iGPU only, and I don't really have 

I got notification just now but the post says 30 minutes ago. Is it my end or you really just posted it 30 mins ago. 

I haven't gone reading through your post properly yet.

[79wjd]

Just now, deXxterlab97 said:

I got notification just now but the post says 30 minutes ago. Is it my end or you really just posted it 30 mins ago. 

I haven't gone reading through your post properly yet.

I posted it 30 minutes ago, but I corrected something and mentioned both you and STRM to let you know how things turned out (not that I'm done since Corsair still hasn't answered).

[Guest]

1 minute ago, djdwosk97 said:

I posted it 30 minutes ago, but I corrected something and mentioned both you and STRM to let you know how things turned out (not that I'm done since Corsair still hasn't answered).

Ah that's why i only get notification now. Anyway Corsair RMA would be your best choice that's free of charge (or low cost)

Else me and Strmfrmxmn will be here helping you with suggestions for a new PSU.

[79wjd]

7 minutes ago, deXxterlab97 said:

Ah that's why i only get notification now. Anyway Corsair RMA would be your best choice that's free of charge (or low cost)

If only they would answer. 

7 minutes ago, deXxterlab97 said:

Else me and Strmfrmxmn will be here helping you with suggestions for a new PSU.

I've got that part handled myself Not that I have any intention of buying a new PSU seeing as I still have two years left on this one. 

[Guest]

Just now, djdwosk97 said:

If only they would answer. 

I've got that part handled myself Not that I have any intention of buying a new PSU seeing as I still have two years left on this one. 

Call them directly would be a good choice for an impatient person like me. 

[79wjd]

1 minute ago, deXxterlab97 said:

Call them directly would be a good choice for an impatient person like me. 

I hate calling support, it's always such a pain. Also, I noticed that the light on the motherboard came back on, so I guess the PSU isn't completely dead, but it took 30~ minutes for it to supply power again.

[Guest]

Just now, djdwosk97 said:

Also, I noticed that the light on the motherboard came back on, so I guess the PSU isn't completely dead, but it took 30~ minutes for it to supply power again.

Better safe than sorry however

Just now, djdwosk97 said:

I hate calling support, it's always such a pain. 

All support I called was nice however.

[STRMfrmXMN]

1 hour ago, djdwosk97 said:

I hate calling support, it's always such a pain. Also, I noticed that the light on the motherboard came back on, so I guess the PSU isn't completely dead, but it took 30~ minutes for it to supply power again.

Corsair support I do believe is in America so if you call them it won't be the biggest headache in the world.

@Corsair Blake @Corsair George @Corsair James @Corsair Johan @Corsair Joseph

Probably more but I tagged a few

[79wjd]

1 hour ago, STRMfrmXMN said:

Corsair support I do believe is in America so if you call them it won't be the biggest headache in the world.

@Corsair Blake @Corsair George @Corsair James @Corsair Johan @Corsair Joseph

Probably more but I tagged a few

I still hate calling support, it's always annoying -- even when they speak english. Corsair has a Live chat though, and they're going to prioritize my ticket (apparently). So, I'll give it another day and then call. 

 

Also, I grabbed my heat gun, and the fan of my RM650 isn't spinning after 45 minutes of being on (and playing Dota) and the temps measured from the exhaust side of the PSU range from 100-125ºF (temps measured from the intake side are about 90-105ºF)

[79wjd]

@STRMfrmXMN Okay, another update. I haven't even touched my computer in the last hour or so (but it was still on, sitting on the desktop with only steam open -- just the display fell asleep). I moved the mouse to wake up the display (again, the computer was running just the display was asleep), and the second I moved the mouse I heard a static shock type noise and the computer shut down, and the motherboard light went out. 

 

So, it isn't an overheating problem, just some other problem with the PSU.

[STRMfrmXMN]

6 hours ago, djdwosk97 said:

@STRMfrmXMN Okay, another update. I haven't even touched my computer in the last hour or so (but it was still on, sitting on the desktop with only steam open -- just the display fell asleep). I moved the mouse to wake up the display (again, the computer was running just the display was asleep), and the second I moved the mouse I heard a static shock type noise and the computer shut down, and the motherboard light went out. 

 

So, it isn't an overheating problem, just some other problem with the PSU.

That sounds bad, as in that could have killed your CPU/motherboard bad. You posted this from your troublesome desktop, I hope?

[79wjd]

9 minutes ago, STRMfrmXMN said:

That sounds bad, as in that could have killed your CPU/motherboard bad. You posted this from your troublesome desktop, I hope?

I posted it from my laptop. I won't know until Corsair replaces my PSU (once they respond to my RMA request) -- I could technically remove my AIO and test my system with my ancient PSU but that's just a lot of work to find out the same thing I'll know once Corsair replaces my current PSU (and everything except the PSU is out of warranty anyway, so time isn't a critical factor). 

[79wjd]

I decided to stop being lazy and just call. They're sending me an RMx. So free upgrade, yay.