• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

Symptoms of a weak power supply?

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.

dim1014

Member
Joined
Feb 27, 2008
Hey guys: I've been having some serious issues with the build in my signature and am wondering if my power supply is just not strong enough to support my recent hard drive upgrade (I had 2 before, added 2 more to create RAID 10)

The problem is: Recently (within the past month or so) while gaming(and only while gaming) occasionally, usually after at least an hour or two and not every time, my computer will just shut off as if I pulled the plug. Upon restarting, the HD array verifies/rebuilds because of the unexpected shut off (this happens if it freezes or gets shut down unexpectedly regardless of the cause).

Looking in the windows system logs, the sole error is that the system experienced an unexpected shut down, literally as if I pulled the plug on the machine. I always monitor my GPU temps in game, and have tried monitoring my CPU temps using rivatuner as well, and they seem to be within normal range.

Things I have tried:

  • taking some of the components out and reseating them, including cables
  • monitor internal temperatures/fans
  • cleaning out any dust inside the case
  • upgrading video drivers/bios/chipset drivers, etc
  • just bought a multimeter and tried testing the voltages off of both the 24pin motherboard connector and a loose molex connector during bootup and under load (Furmark + Prime95). They all seemed within reasonable ranges
  • visually checking capacitors on the motherboard for leaks

I suspect the problem pretty much has to be either the power supply or the motherboard because of the error I get. Typically Memory errors give some sort of blue screen and video errors do the same or at least freeze the computer. This is a complete low-level cut in power.

I thought the power supply was the most likely culprit but it seems to have checked out with the multimeter (if there is more I should try here, I am open to suggestion). I can't even RMA it because BFG went out of business, so I would have to buy a new one to try it out. Other than the possibility of it being bad, I'm pretty sure it provides enough power to run my components, especially considering the shutoffs didn't start till a few months after I put in the new HDs.

However, I did just used newegg's power supply calculator (how accurate is this thing? http://educations.newegg.com/tool/psucalc/index.html) and it looks like before adding the 2 extra hard drives I was just under 800 watts, and after adding I am in the high 800's (see here ). My power supply is only 800w so if this is accurate then I am definitely over.. Would shutting off under load periodically be a symptom of this?

Sorry for the long post, and I appreciate you guys reading, and I am open to any suggestions, thanks!
 
Possibly a bad memory stick too. Might want to run a Memtest and Prime/Orthos again. They don't always leave a message on a BSOD and assuming you checked with your multimeter while system is under a load. Many times idle and load readings can be quite different.

That power supply isn't even close to its capacity for what you have and the additional drives are insignificant for additional power needs.
 
I did check the system under load as I listed (using Prime95 and Furmark simultaneously). I will run memtest86 when I get home, leave it running throughout the night and report back.

In the meantime, any other suggestions/ideas are welcome as well!
 
no worries, I appreciate the help! Assuming the memory tests pass, any ideas on what else it could be?

Another note: When I added the two extra hard drives, I was short a sata power connector and not able to hook them all up. I bought a splitter on newegg and used that to connect two of them. Could splitting the power to 2 hard drives possibly cause this issue?
 
Last edited:
How did you plug a molex into those drives? Surely that power supply has 6 sata power connectors
 
I used one of these:
http://www.newegg.com/Product/Product.aspx?Item=N82E16812123119

Looking at the power supply specs and you're right (I also meant sata in the previous post, correcting it). Looking back I think what actually happened was that the sata power cables couldn't read all the components so that's why I got the adapter (for extra reach). When I get home I will look at the cable layout and see what I actually did because it was a number of months ago.
 
thanks. So after reading around, it is seeming more and more like the symptoms I am experiencing (power cut under heavy load after a while with no temperature spikes) is typically due to an underpowered power supply. I'll keep looking around though and run memtest tonight
 
If your power supply is faulty yes but that PS running normally is not in any way underpowered.
 
Possible causes:

1: The processor reached a critical temp. (most likely) (Your cooling sucks)

2: Bad caps.

3. Weak battery. (laptop)

4. Bad power connections.
 
Possible causes:

1: The processor reached a critical temp. (most likely) (Your cooling sucks)

2: Bad caps.

3. Weak battery. (laptop)

4. Bad power connections.

1. been monitoring all temperatures under heavy loads and they are all within low ranges (GPU gets in the low 90s sometimes but only during graphically intensive games, and it is just a hot card).

2. I checked the caps visually, at least on the motherboard, they look fine, otherwise I am not sure how else to check them, any suggestion here would be welcome

3. not a laptop

4. i considered this and will make some more adjustments and see, but I have reconnected everything securely and it does not seem to make a difference.
 
Update: ran memtest for about 8 hours with no errors. I actually did have to run it twice because it froze after 4-5 hours one time (I googled that usb legacy settings in BIOS and cause that and sure enough the problem was fixed when I turned them off).

Also rechecked all the voltages for power cables connected to each HD and they are all normal. I haven't had a crash in a few days but I haven't been playing any games so that's probably why. I'll try to get some gaming in tonight and see if it happens.

***
I did do a bit more research and found some more info though. Looking through the windows system event logs for more details on events occurring before the crash(since the actual error is not informative), I found a few things.

The actual crash event I get is
Critical 3/15/2011 9:25:51 PM Kernel-Power 41 (63)

The system has rebooted without cleanly shutting down first. This error could be caused if the system stopped responding, crashed, or lost power unexpectedly.

Right before it, every time usually less than a minute before, the following event occurs:
Information 3/15/2011 9:25:46 PM FilterManager 6 None

File System Filter 'FileInfo' (6.1, ‎2009‎-‎07‎-‎13T19:34:25.000000000Z) has successfully loaded and registered with Filter Manager.

Sometimes, this one also occurs:
Information 3/15/2011 9:25:51 PM FilterManager 6 None

File System Filter 'MpFilter' (6.1, ‎2010‎-‎09‎-‎14T20:19:28.000000000Z) has successfully loaded and registered with Filter Manager.

Upon googling around for these keywords, I found that tons of people with various configurations/hardware have the exact same symptoms as me. The problem is almost no 2 solutions were identical. In all of the threads/posts I found people seemed to fix it with various combinations of the following:
  • replacing memory
  • replacing motherboard
  • replacing power supply
  • resetting CMOS(tried this)
  • replacing graphics card
  • reinstalling various drivers
  • replacing heatsinks/fans (this one is definitely not my problem though..)

Basically, it appears that this error/shutdown problem can occur for just about any reason, even including driver conflicts/issues (even though I thought it can't possibly be a software issue).

The next thing I will probably try is a complete Nvidia driver wipe and reinstall using driversweeper.

I'm weary of replacing random hardware components blindly so I am interested in narrowing this issue down first. As usual, all suggestions and ideas are welcome. Thanks for reading!
 
Dim,

Then if it only happens during games, I would concentrate there (drivers) first. Memory seems ok as does your power supply and temps. A bad power supply won't show in event viewer.
 
Thanks. Next thing I will try is reinstalling my drivers one by one. I'll report back!
 
The usb thingy i also had issues with but it was on the ud3r it took me some playing around to get even get the pc to post. Turning them off helped out.
 
the legacy usb setting seemed to only give me trouble running memtest. I actually toggled it earlier inadvertently by resetting my cmos at some point and I had crashes both ways so it doesn't seem to make a difference otherwise. Glad you solved your issues though!
 
update:
What I have done since my last post:
1-unplugged and replugged all power cables
2-did a complete wipe of video/physx drivers useing driver sweeper

since doing these 2 things I have had no crashes, but it still may be too soon to tell. If I don't respond back here again for anyone else that encounters this issue, then my problem has been fixed.
 
almost forgot, one other thing I did that may be of significance: I swapped out the power connectors on my video card for the other spare set coming out of my power supply. If this was the fix then I'm thinking either the first set of connectors is faulty, or it has something to do with overloading a rail (although all voltages seemed to come back normal).
 
almost forgot, one other thing I did that may be of significance: I swapped out the power connectors on my video card for the other spare set coming out of my power supply. If this was the fix then I'm thinking either the first set of connectors is faulty, or it has something to do with overloading a rail (although all voltages seemed to come back normal).

Dim,

If your power supply isn't faulty, and it appears to be ok, any excess power needs on a rail will be passed of to another rail. That shouldn't be a problem whatsoever. Connectors can be faulty but it would be unusual.

FYI, here is a comprehensive "Sticky" in this section on single and multiple rail power supplys if you need a great explanation:

http://www.overclockers.com/forums/showthread.php?t=548862
 
Back