Hey guys, I've been suffering some odd random shutdowns of my computer, in that the it powers off completely, like someone just pulled the plug and I have no idea why its doing it. I have noticed however that my 12v Rail does seem to get a little low, sometimes drops to 11.2v Screenie This was running the Aida stability test, you can see the 12v reading. Wondering if this has something to do with the random shutoffs, and also what would cause the 12v Rail to be so low? One thought I had was that I'm running One Molex cable with 5 fans connected to a Lampton fan controller, another 3 fans running off the NZXT controller on the Swith 810, a SSD, and finally my XSPC Pump. All this is connected to a single molex cable. Now I know that the HX1050 is a big Single 12v Rail, but I'm wondering what the actual molex cable is capable of? I don't know anything about this stuff, is it worth putting a second molex cable in to share some of the load?
Yep that 12v is dropping waaaaaaaay below spec, looks like your PSU is to blame for the random shutdowns, and no - extra molex cables etc won't make a difference. Time to RMA... Btw - is your case really hot inside or does the PSU exhaust unusually hot air?
Cheers for the response dude, was hoping not to RMA. I've not had this PSU long, think its about 10months. Surprised its having issues especially as its a Corsair. Don't think my case is hot inside, its a Switch 810, I've got two watercooling rads in it. My CPU and both GPUs are under water, my NB/QPI heatsink does get very warm to touch, did in my Antec1200 also. But the PSU's fan is pointing to the floor so it gets air from outside the case anyway.
Yeah I've been told that before, I have no idea how to use one or what one I should even get. Any chance you could recommend one?
any cheap DMM model will do often find them in B&Q for about 7 quid this would do http://www.amazon.co.uk/dp/B000LFXG...de=asn&creative=22218&creativeASIN=B000LFXGFK easy to use just set the dial on the meter to 20 DVC which means 20 volts then stick the red probe + on meter connector to the molex connector on your psu with the yellow wire which is the 12v line then the black probe - or com port connector to any ground source meaning any black wire on the molex or even the case can act as ground connector then watch the meter readings should say on display then stress an idle your system to see how much it alters etc under load conditions
It's not hard to put 2+2 together and see that he has a low 12vrail, besides I don't think the reading will that far off even if not measured with a multimeter...
I've personally seen software say my voltage was 11.1v for 12v rail then cracked out the DMM measured it, an it was in fact 12.2v software only as reliable as the voltage sensor readings on the board, actually software might not even be reading the right sensor at all. Hence why software readings when comes to PSU's should be taken with large pinch of salt.
Right, so you're saying the fact his PC shuts down randomly for no reason is in no way related to his 11.2v 12v rail? Which is below ATX spec of 11.4v. Oh please....
could be motherboard just saying I wouldn't take that software reading as gospel seem's its idling at 11.3v if you trust that software so should be dying like all the time ^
Ok, cheers matey, I'll probably order that one. I'll prob get more in the future, one day hoping gather enough cash for a benching rig and sub zero cooling, so I'll need a few I guess. Heres a screen shot of HWMon and Aida whilst at idle, IE was running and usual crap in the background, but nothing stessful like a game or stability test They both seem to report differing results by around 0.15v So I guess it won't hurt to spend £5 on a DMM just to be sure, I have no other PSU here and can't afford to buy another in the meantime so RMA is my last resort I guess. Its been running fine today, but then it ran fine for a few days last time and then last night just boom, went down, no reason whatsoever. I have flashed my BIOS to the lastest this morn, so it'll be good to leave it for a few days to see how it gets on, by then that DMM may have arrived and I'll take it from there. Cheers for your help though fellas, I'll report back anything I do or experience in the meantime.
well DMM is good diagnostic tool an will aid you in RMA return simply if you say you done the readings using a DMM they will take that over any software reading ^ however from your other SS does appear your psu might not be functioning properly Notice though how two different software's are giving different readings for 12v rail
Depends, 11.3 may not be enough to shut it down - it could be due to fluctuation causing the 12v to drop even lower....*shrugs*. Anyway I guess there's no harm in checking with a MM to make sure... Btw OP I find HWMonitor's 12v readings differ from most other software for some reason...just saying...in other wors I don't really trust it. Try HWiNFO64
Well if you want to see if it's psu try this http://www.ocbase.com/ install an select occt PSU test if you have flaky psu chances are it will shutdown during this test
aida has always been spot on for me. think you need an rma, but if you have a mm handy try that at load. that psu working right should power your system no problemo. you might to want to try the psu tester in occt, but do that at your own risk, think it would shut down your psu fairly quickly
Usually a diferrence in voltage bigger than 5% can cause shutdown. Especially if it´s under the reference. I always consider a variation of 3% to be safe. 11.3v is way too low.
Hey Matt It was only recently you purchased that psu so it would be surprising if it is to blame, I would also recommend a DMM as software can be off and it's quite cheap. It may be helpful in the future too. Another thing to look into could be GPU temps, as I have, on occasion, had a strange issue since adding the water block. What happens is that software like afterburner/gpu-z/ccc all suddenly start to see GPU temps as either very high (1000+ C) or very low (-1000c), this seems to be caused by some software conflict and this effects only the 4 GPU temp readings not the VRM temp. On occasion when the temps are read as too high the driver attempts to put the fan to 100% and after a while will suddenly shut the pc down without warning. This is what the GPU temperature sensors typically look like when it happens, here it's a negative reading on other occasions when it shuts down it's a positive reading of a similarly unbelievable number.
Test with a multimeter. Chances are it is dropping outside the ATX spec (+/-5%, 11.4V to 12.6V). If you confirm it is outside spec with a meter, it should be returned.