Can't figure out which part is causing my problems, help!

Toggle sidebar Toggle sidebar
S

SildricStormwind

Honorable
Jun 26, 2013
10
0
10,510
Hello all! I just put together my new PC and was playing some games on Steam that it should definitely be able to handle and the PC just reset, it's happened twice now and I'm guessing that it's the power supply, but I wanted some other opinions because I feel like this power supply should be able to handle what I have. It only happens while I'm playing games, so I'm pretty sure it's not the motherboard. Also, right after the PC did it the second time it did it I booted into the BIOS and the CPU was running at 44C, so I'm pretty sure it's not overheating. Anyways, here are my specs, let me know what you think!


GIGABYTE GA-990FXA-UD3 AM3+ AMD 990FX SATA 6Gb/s USB 3.0 ATX AMD Motherboard
ASUS GTX660 TI-DC2O-2GD5 GeForce GTX 660 Ti
FSP Group AURUM GOLD 700W (AU-700) ATX12V/EPS 12V
AMD FX-8350 Vishera 4.0GHz (4.2GHz Turbo) Socket AM3+ 125W
SAMSUNG 840 Series MZ-7TD120BW 2.5" 120GB SATA III Internal Solid State Drive
G.SKILL Ripjaws Series 8GB (2 x 4GB) 240-Pin DDR3 SDRAM DDR3 1600 (PC3 12800)
TOSHIBA PH3300U-1I72 3TB 7200 RPM 64MB Cache SATA 6.0Gb/s 3.5" Internal Hard Drive


Thank you for your time!

Alright, as an update it just did it while installing a game, any help would be greatly appreciated.

Another update, just ran a torture test on it and the CPU was temping at 67C after 30 minutes and holding stable, I think that I've ruled out the CPU and GPU, I also highly doubt it's a RAM issue, but my original RAM was DOA and I'm sending that back anyways. That only leaves the SSD, Motherboard and PSU. I really doubt that it's the motherboard, I feel like during the torture test that would have pooped out were it the problem. This is my first time with an SSD, but it doesn't look like a normal HDD problem, so unless someone tells me otherwise I'm going to rule that out too. It must be the PSU, does anyone else have any other ideas? Thanks ahead of time.

Link to torture test stats: http://postimg.org/image/ebge0dv3t/

Link to GPU test stats: http://postimg.org/image/8p67pinp9/
 
Sort by date Sort by votes
L

liquidsnaske

Honorable
Jun 26, 2013
55
0
10,660
Hi Mr Stormwind,

My experiences of intermittent restarts have been either something metallic shorting out on the motherboard, overheating or a faulty disc drive. My suggestion is that you make sure that there is nothing shorting out on the motherboard and try ti with the disc drive disconnected. Also checking the GPU temperature could provide a clue to.

The last time I saw this it turned out that the BIOS battery had worked it's way loose and was rattling around inside the PC! Thankfully no lasting harm was done from it and it also answered why the bios was randomly resetting. Let me know how you get on.

Hope that helps
 
Upvote 0 Downvote
S

SildricStormwind

Honorable
Jun 26, 2013
10
0
10,510

Alright, thank you for the advice. I have checked the motherboard acouple of times, so I doubt that that is it. I've run Power Supply and GPU tests using OCCT and didn't see anything strange, so I'm not quite sure what is wrong. I'll try unplugging the disc drive, but I don't think that that would be the problem. I'll keep you up to date, thank you.
 
Upvote 0 Downvote
L

liquidsnaske

Honorable
Jun 26, 2013
55
0
10,660
Ok, a couple of other things spring to mind then. I agree it doesn't sound like a storage problem but some ssd controllers can be finicky with motherboard storage controllers so I suggest updating your SSD firmware and bios. You can also run a stress test on the ssd using something like Crystal disk (http://crystalmark.info/software/CrystalDiskMark/index-e.html) if that doesn't solve or diagnose the fault then perhaps a memory test might reveal something.

Finally can you check Event Viewer in windows for any critical / Errors reported and see if they have any clues in?
 
Upvote 0 Downvote
S

SildricStormwind

Honorable
Jun 26, 2013
10
0
10,510

I can definitely try testing the SSD. The event viewer has a kernel-power critical event. O.O

Here is a link: http://postimg.org/image/4iuiq46nr/

Thank you for all of your help so far! It looks like we're making progress!
 
Upvote 0 Downvote
S

SildricStormwind

Honorable
Jun 26, 2013
10
0
10,510


I just tried running the OCCT power supply test and it did it again, I'm going to attempt it again, although I'm still not 100% convinced that it's the Power supply.

Also here is the SSD test: http://postimg.org/image/qk209q04j/
 
Upvote 0 Downvote
L

liquidsnaske

Honorable
Jun 26, 2013
55
0
10,660
The kernel power error happens whenever the system unexpectedly loses power, so it is to be expected and shows us only that it isn't locking up just before it shuts down so it doesn't give many hints. The SSD test just confirms that they are being written to and read from with no errors as well at max speed.

I think the best way to proceed is to follow the advice given here: http://support.microsoft.com/kb/2028504#method3 particularly disabling overclocking and running memtest http://www.memtest86.com/ .

Edit: Just thought I'd add that we can probably rule out graphics and storage problems here as they would have prompted an error log to be written specifically for them.
 
Upvote 0 Downvote
S

SildricStormwind

Honorable
Jun 26, 2013
10
0
10,510


Okay, I ran the memtest and it seems as though there was 1 error during test 5 and 2 errors during test 6. I had a friend come over who is good with this kind of stuff and he tweak some of the bios voltage settings and we re-ran the tests and there were no failures. We assumed everything was good, but as soon as he left I loaded up a game and it almost instantly went down... I'm not sure what the problem is here now that it passes the memtests. :/
 
Upvote 0 Downvote
L

liquidsnaske

Honorable
Jun 26, 2013
55
0
10,660


Ok I'm running short on diagnostics now, but if you're tweaking the voltages then I wonder about other changes you've made to the system. Can you remove all overclocking (easiest way is to reset the bios to defaults, make a note of the voltages your friend has used) and test your game again? Also can you confirm that there is no blue screen of death before it restarts please.
 
Upvote 0 Downvote
S

SildricStormwind

Honorable
Jun 26, 2013
10
0
10,510
Okay, we have only changed the North Bridge Voltage by a bit and it fluctuates between freezing mid-game and rebooting. There is no blue screen and we have not overclocked anything (we reset the bios right before changing the NB voltage). Currently the voltage is at 1.16 on the NB and the game lasted about 30 more seconds than before. Thanks for all of your help so far, you're my hero today.
 
Upvote 0 Downvote
L

liquidsnaske

Honorable
Jun 26, 2013
55
0
10,660


Ok as I see it there are 3 options left which will reveal the culprit one way or the other. The 1st thing to try I think is to start testing the PC with only 1 stick of RAM at a time (preferably with default voltage on NB, or try both) to see if there is a faulty stick there. If that fails then do a fresh install of windows and re-test, this really doesn't sound like a software problem but we haven't ruled it out completely. The final thing to do if those fail is find a similar PC / parts and begin swapping components until it works which will definitely tell you which part is faulty.

Anyway I'm going to catch my 40 winks now, so good luck and I'll check back in the morning to see how it goes.


 
Upvote 0 Downvote
S

SildricStormwind

Honorable
Jun 26, 2013
10
0
10,510


Okay, thank you for all of your help and I'll keep you posted.
 
Upvote 0 Downvote
nostall

nostall

Splendid
Nov 28, 2012
3,244
12
23,865
This is beginning to sound like a power supply issue. The 12v rails on that PSU can only supply 18amps per rail and your video card uses only one 6pin power connector. The minimum recommendation for your video card is 38amps. My bet is that as soon as the video card starts to work (gaming) it draws more power then your PSU can deliver through the 1x6pin power connector.
If your friend has one, try a power supply with a single 12V rail rated in excess of 38 amps. If you find you need to send your Aurum back, here're a few recommendations:
http://www.newegg.com/Product/Product.aspx?Item=N82E16817139012
http://www.newegg.com/Product/Product.aspx?Item=N82E16817182068
http://www.newegg.com/Product/Product.aspx?Item=N82E16817151096
http://www.newegg.com/Product/Product.aspx?Item=N82E16817207013
http://www.newegg.com/Product/Product.aspx?Item=N82E16817182071
http://www.newegg.com/Product/Product.aspx?Item=N82E16817207014
Any of these should have enough juice to power your rig.
6/28; 8:54PDT My mistake, one 660Ti needs 24amps, TWO (SLI) need 38A.
If you are monitoring this thread, Please post when you have solved your problem or need more help; and I hope my suggestions have helped you.
 
Upvote 0 Downvote
S

SildricStormwind

Honorable
Jun 26, 2013
10
0
10,510


Hello, sorry for the delay, but I was waiting on a new heatsink and have been very busy with my summer classes. As an update the problem is a lot better, but not fixed. Following your advice I went and tested each individual stick of RAM and everything ran fine with each stick. My friend told me that he thought that the memory controller in the CPU may be overheating (at the time we were running 4 2-GB sticks out of his machine) Each stick worked better on their own than the 4 sticks combined. With the 4 sticks in playing games would cause the machine to crash in > 4 minutes, but with 2 sticks in it often took 1-3 hours before it went down. So I ordered a new heat sink. This new one is a beast and I just installed it today, cleaning all of the old thermal paste off the old CPU and applying new thermal paste in addition to the heatsink.

Since doing this I have been playing my games and yet again, about three hours into it the computer reset. While playing the game the CPU stayed in the 30's (Celsius), but occasionally would spike to the upper 40's. I decided to run the torture test and it is stabilizing at 44C right now. I'm not sure what the problem may still be. Thank you guys for all the help and I'm sorry I kept you all in the dark so long.

Here is a link to the new heatsink on Newegg: http://www.newegg.com/Product/Product.aspx?Item=N82E16835103099
 
Upvote 0 Downvote
nostall

nostall

Splendid
Nov 28, 2012
3,244
12
23,865


HAVE you tried a different PSU yet. You need to try a 'known' good PSU; a brand new out of the box unit might still be bad. Borrow one from a friends computer or one that you are sure is working. It should be one with a single +12V rail rated for 24 amps or more. If your problem is PSU related, then everything else you have tried or will try will be simply 'chasing ghosts'.
 
Upvote 0 Downvote
You must log in or register to reply here.

TRENDING THREADS

Latest posts

Moderators online

Share this page

COMPANY
RESOURCES
FOLLOW
Top Bottom