• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

PSU causing lockups?

E

Ebernanut

[H]ard|Gawd
Joined
Dec 15, 2010
Messages
1,926
I've been experiencing some random lockups(screen loses signal, sound usually goes on for 5-10 seconds and then starts looping) with the only way to recover being a hard reset, I never get a BSOD and event viewer doesn't show anything except the PC shutting down unexpectedly. I'm hoping someone can either confirm my thoughts or tell me I'm crazy because I've been pulling my hair out trying to figure this out.

The problem has been going on for about a year but at first it would go weeks or months between incidents so it's been hard to diagnose, since about Jan it's been getting progressively worse and I've been able to do a bit more but it will still go days to weeks without an issue at times. Everything has been at stock since last fall and the problems still occur with the BIOS set to fail safe defaults.

It happens somewhat randomly but almost always when I'm actually doing something though it's often something minor like clicking over to another tab in my browser, when it's acting up games have a good chance of locking it up almost immediately or at least more frequently. It's also more likely within the first half hour of turning the PC on if ambient temps are below 70f or after several hours if ambient temps are above 80f, component temps are all within range(cpu, gpu, gpu vrm, chipset). A couple times after lockups I had to remove my gpu drivers in safe mode or the monitor would lose signal right before windows got to the desktop but I'm guessing the lockups just corrupted the drivers somehow.

I don't think it's a software issue because I've tried multiple drivers for every component(several for my GPU w/driver cleaner) with no change and was even able to replicate it with the win 7 install I still had on my old spinner drive. I've tested the ram with memtest and by removing one stick at a time and it seems fine, unfortunately I don't have a spare PSU or GPU I trust to swap out. I've also reseated everything and visually checked for bad caps.

Now, on to why I think it might be the PSU and why I'm not sure. Most of the time when I've checked voltages through software or a multimeter they've semed fine(3.28, 4.8-4.9, 12.25-12.35) but the last week or so it's been acting up more and I've seen varied readings on the 12v rail using my multimeter. When it's been acting up shortly after boot I've seen readings at 12.00-12.08(within atx specs but different than when it's stable) and when it's been acting up after it's been on a while I've seen readings of 2.45-12.54(outside of specs by a hair). Earlier today it was actually showing the highest seemingly reliable reading I've seen at just over 13v but then 5 minutes later it was at 12.25.

The multimeter is a cheap one and I've had a couple odd things occur that have me questioning it(I did replace the battery a couple months ago). One is that I've seen some really high readings that then go out of the 20v range, the first time it happened I checked the 12v rail and it read ~17v and quickly rose to out of range and the second time I wanted to monitor it while I played a game and it started out a normal reading and over 30-45 minutes it slowly rose and got out of range. The weird part is that when I removed the probes and put them back it returned to a normal reading, it then started to raise slowly again but it hasn't happened since then. The other is that when I turn my pc off the 12v rail doesn't completely power down, it drops to about 0.8v and then over a minute or two drops to 0.4v(5v on molex shows 0.0v); the fact that the 5v rail doesn't do that makes me think it's the psu but with the other weirdness I'm not 100% trusting my multimeter.

I'm thinking it's the power supply but that it could be the GPU or MB and I'd rather not start throwing money at the problem until I can be fairly sure of what the problem actually is, especially since I halfway plan on retiring it once Haswell-E comes out. NE has the corsair HX750 for a good price right now, any reasons to go with something different if I decide to replace the PSU?

I apologize for the long winded post, I tried to make it concise but I didn't want to miss too many details. Thank in advance for any input.

Specs:

i7 920@stock
AMD 6950 2gb@stock with stock BIOS(was shader unlocked)
Gigabyte ex58-ud4p(f13 BIOS)
G Skill Trident 3x2gb 2000mhz@1066
Corsair TX850(about 5 years old)
Crucial M4 256gb
WD Black 1tb
WD Red 3tb
Hauppauge Colossus(capture card)
 
I guess I should have mentioned that it's the v1 TX 850, I forgot there was a v2.
 
I was able to determine that that the odd high readings are an issue with my multimeter but I was also able to determine that outside of that intermittent issue the readings have been correct.

There are a couple of things I forgot to mention. The first is that there has been a handful of times the last month that the pc has restarted instead of locking up(always within the first hour of booting) which I know is a more classic sign of a bad psu. The other is that whenever the pc locks up the video card makes a whirring sound for a second that sounds like when the fan rpm changes, this card has always made the same sound during post. I can also confirm that the pc is completely locking up, it's not just the display.

The pc has been stable the last couple of days and the 12v rail has been at 11.98v-12.1v, I haven't seen any of the 12.5ish readings(or the single 13v reading) like I saw when it was acting up a few days ago. I also tested it this morning after it had been off all night and the 12v rail was still showing 0.4v with the pc off, this seems really odd unless I'm mistaken about the only standby power coming from the purple +5v wire on the 24 pin connection.

I think I'm going to wait until the pc starts acting up again and see if the voltages are higher again as well. Once again I apologize for the rambling OP, I've been dealing with this for so long that my thoughts on the subject are a bit scattered.
 
It's the PSU... I can remember having almost similar problems like sound looping and system locks up on a previous PC I used, then eventually the PSU just died and wouldn't start. Memory can give you that same problem, but seems you tested it out. My advice, save up now for a PSU, this one is slowly dying out.
 
I don't mean to keep bumping this thread but I thought I should update it in case someone with a similar problem finds it.

My hunch and ravensteel's comment appear to be correct. I finally managed to monitor the 12v rail right before the PC locked up a couple times and the voltage was fluctuating quite a bit and spiking well outside ATX standards for a few seconds and then the PC would lock up. In addition to that the PC had been stable for a couple of weeks and that entire time the 12v rail was 11.99v-12.01v whenever I checked it.

I'm pretty sure that the PSU I ordered(EVGA Supernova 750 G2) will fix the issue but if it doesn't I'll update the thread again.

@ravensteel: Thanks again for the input.
 
You must log in or register to reply here.
Back
Top