Sign in with
Sign up | Sign in
Your question

New build is random shutting down!

Last response: in Systems
Share
January 31, 2010 3:40:00 PM

I'd like to get some opinions on my problem I'm having. Maybe someone can even tell me for sure. First, let me tell you what's in my machine:
RAIDMAX SMILODON ATX Case
MSI NF750-G55 Motherboard
AMD Phenom II X4 955 BE
G.SKILL 4GB (2 x 2GB)DDR3 1600
Rosewill RBR1000-M 1000W Power Supply*
EVGA 896-P3-1257-AR GeForce GTX 260*
ARCTIC COOLING Freezer 7 Pro
2 IDE DVD burners and 2 SATA HDDs (1x320g, 1x1tb)

*=Likely culprits in my problem

Here's the deal. I throw my computer on after installing the PS and it runs w/o problem for a good day and a half on general computer use. I go to play Star Trek Online (STO) and it runs fine for about 20 minutes or so. Then my whole system shuts down in the middle of the game. No "Windows is shutting down" screen, no error messages, not even a word in the event log when I reboot. It shuts down like someone is just pulling the plug out of the wall. After I fire it back up it runs for about 10 or 15 minutes and does the same thing, again while playing STO. I reboot, try playing Borderlands, shuts down after about 20 minutes.
The highest temp on the GPU that CPU-z has clocked is 78c and I've seen online my card should tolerate up to 85c. But I did notice that it shut down twice around 78c. I throw my older PS in the system just to create a variable (it's a 550w) and I can play STO ALL NIGHT! Which is double weird because the Watt Calculator I used said I'd need 685w for my system minimum, but I won't get in to that. So I think it's likely the PS, but I can't run it hard enough to get the GPU temps up to 78c with the new, apparently cooler running, PS. I even tried my roommate's PS (600w or 650w) and same story. Played all night, can't get temps that high. So I KNOW it's the PS, but is it because it's generating too much heat and causing the video card to have problems or is mine defective? I've had 3 people try to help me IRL and can't figure it out. I've already RMA'd one PS (bad cables when I pulled it out of the box) so I know the shipping on these things is $15 and I'd rather not ship it back when it's not the problem.

Can anyone see what I can't see? I appreciate any help I can get.

More about : build random shutting

January 31, 2010 5:43:39 PM

I wouldn't really vouch for Rose will's power supplies, but that's all good. It does seem to be a problem with the PSU, but don't take my word on that...

Your graphics card gets hotter when you have the Rosewill power supply in there? That's strange... Might wanna look into that. Also, I've had an older graphics card run hotter than 78 degrees, and no shutdown, so I don't think that's the problem.
m
0
l
February 1, 2010 9:19:35 AM

I know Rosewill isn't king of computer parts, but the last one of theirs that I had worked without issue (in fact, that's the 550w that's powering my machine now) but I will bookmark that OCZ if I can't get any further with this one.

I assumed the video card being hotter with the 1000w in there was do to it being a more powerful PS and generating more heat. Not sure how true that is, but it made sense at the time.

I'll try contacting Rosewill today and see if I can get some answers from them and maybe not have to pay another $15 for Newegg RMA.

Also, thanks for responding. I have 30+ views on this thread at the time I'm writing this and you're the only response. Either I've stumped a lot of people or no one felt like trying to help. Thanks for at least throwing in your opinion! :na: 
m
0
l
Related resources
February 1, 2010 7:23:26 PM

Haha, no problem. And scratch that power supply I linked you to, another more experienced member of this forum said he wouldn't recommend it.

I keep reading over your original post and I dunno. I'm not as knowledgeable as some of the guys here. Sorry I can't help.

What were your cpu temps and mobo temps?

Try underclocking your graphics card, generate less heat, maybe take less voltage, idk. For testing's sake?
m
0
l
a b B Homebuilt system
February 1, 2010 7:39:18 PM

There's a reason nobody on these forums recommends Rosewill PSU's. The specs might look good on paper, but they are not quality units. You're best off with a quality unit from a known good manufacturer like Corsair, Antec, PC Power & Cooling, Seasonic, etc... OCZ also makes some decent PSU's that are priced very low. Are you planning on adding another GPU in SLI in the future? Here's a few PSU suggestions.

CORSAIR CMPSU-650TX 650W $89.99 - $20 MIR

OCZ ModXStream Pro OCZ700MXSP 700W ATX12V V2.2 / EPS12V SLI Certified CrossFire Ready 80 PLUS Certified Modular Active PFC Power Supply $89.99 - $30 MIR + an extra $10 off w/promo code EMCYZNW76, ends 2/3

SeaSonic S12D 850 Silver 850W ATX12V V2.3/EPS 12V V2.91 80 PLUS SILVER Certified Active PFC Power Supply - Retail $109.99

The PSU above would be plenty even if you add another GPU in SLI.
m
0
l
February 1, 2010 8:03:58 PM

"Planning" sounds a little optimistic, but I'd like to have the option to SLI.

So we're sure it's the PSU that's the issue and not the GPU overheating? That's the conclusion I came to, but I don't want to go barking up Rosewill's tree and find out it's something else.
m
0
l
a b B Homebuilt system
February 1, 2010 8:12:22 PM

Yes, it certainly sounds like the PSU is faulty. That's nearly always the case when the system simply shuts off. The fact that the GPU runs hotter with that PSU is not a good thing at all. It probably means that POS Rosewill PSU isn't running the 12v rail within specs and is sending too much voltage to components.

The SeaSonic PSU linked above is a high quality unit that will easily power the system with any upgrades you could ever do. I wouldn't use another Rosewill PSU if you value your components.
m
0
l
a b B Homebuilt system
February 1, 2010 8:16:18 PM

I wonder if the 1000W PSU is having problems blowing out enough hot air, or otherwise dissipating heat? The symptoms honestly sound exactly like an overheating problem.



m
0
l
February 1, 2010 8:21:41 PM

Download speedfan and check voltages. Or check them in your bios, which gives a more reliable reading.
m
0
l
February 2, 2010 3:57:12 AM

Great stuff. A flood of responses. I'll check my voltages. As far as the heat goes I definitely see what you mean about possibly sending too much voltage through although I had the same opinion that it was probably just the PSU sending it's heat straight down, but the voltage thing makes sense too.

IDK if I can flat out return it to Newegg still and get my money back to buy a new one, but I can try. Might see what Rosewill can offer as far as a return goes. Like I said, my old 550w was awesome so I never thought twice about it. After this experience and the overall poor things said about them here I don't think I want to make a third attempt on this PSU.

I'll keep an eye on that SeaSonic while I'm settling this with Rosewill/Newegg, but while I'm doing that is there a similar model that is SLI ready? I didn't see that in the spec and I was hoping to keep that option open.
m
0
l
February 6, 2010 1:21:50 PM

Looks like I've found the culprit thanks to the help from a Rosewill technician.

My video card requires 36 amps on the 12v rail and the power supply I got puts out a max of 30. Newegg was wonderfull and is going to return the item for me so I'll definitely be looking to buy one of the above items. Thanks for the recommendations everyone and now I need to go find out what the hell a rail is lol
m
0
l
!