computer shut off during a benchmark

[toyota]

I was running the Hard Reset game benchmark and my pc just completely shut off. I have the alarm muted on my UPS but the lcd on it was active which only happens if manually click on or there is a power problem. there was no issue in the house itself as for a power goes so maybe the power supply had an issue and caused the pc to turn off? or is possible the UPS caused the problem? anyway now I am paranoid as hell.


PSU is in sig and here is the UPS I have http://www.newegg.com/Product/Product.aspx?Item=N82E16842102132

 

[BFG10K]

Try the same thing without the UPS?

It could be the PSU’s protections shutting it down, or it could be other things like the UPS or an unstable overclock.

 

[toyota]

I have tried to make it do it again by doing loops of the benchmark and nothing happens. nothing is overclocked and I show only about 325 watts being used at the wall in that benchmark so the system was not overly loaded.

yeah it had to be UPS or power supply related because a regular crash would not have triggered the UPS alarm.

 

[Rifter]

Almost all issues like this i have seen over the years led to a faulty PSU.

 

[toyota]

this is really going to bother me because I know something is likely wrong with either the psu or ups.

 

[BoomerD]

No chance the benchmark overheated your CPU? I don't see your CPU cooler listed in your system specs...but the stock cooler should be good enough to keep it cool...IF the cooler is seated properly and the TIM is applied correctly.

 

[toyota]

No chance the benchmark overheated your CPU? I don't see your CPU cooler listed in your system specs...but the stock cooler should be good enough to keep it cool...IF the cooler is seated properly and the TIM is applied correctly.

zero chance of that as the cpu never gets out of the 50s or 60s in games. and the benchmark had just started too.

again a regular crash would not have triggered the UPS alarm. it had to be power related in some way.

 

[BFG10K]

and I show only about 325 watts being used at the wall in that benchmark so the system was not overly loaded.

OCP doesn't need PSU overloading to trip, just an overloaded rail. A momentary short from a flaky connector or wire could cause it. If OCP tripped there’s nothing wrong with the PSU or the UPS.

I wouldn’t worry about it unless it happens again in the near future. It could just be a freak occurrence.

 

[Stuka87]

This was happening to me whenever I clocked my CPU up high and had some heavy GPU load. Turned out to be the PSU I had used three 12V rails, so the rail that the CPU and GPU were on was having too much current drawn, so it switched it self off.. When it did happen, I had to flip the switch on the PSU off and then back on in order to turn the system back on. With stock CPU clocks it was fine.

Replaced it with one that has a single 12V rail, and no longer have any issues. And I am running an even higher CPU clock than I did then.

EDIT: Oh, I now have the same PSU as you. I highly doubt the 12V rail is getting over loaded. with your setup.

 

[Rvenger]

Check your event viewer. It might say a thermal shutdown has occurred. This happened to me when I was running a benchmark and the computer just shut off. I went to the event viewer and found that it performed a thermal shut down. I think one of the sensors glitched while it was benchmarking and hit the ceiling on my shutdown temp in the bios. Check for that.

 

[toyota]

all it says is

Source: EventLog
Date: 10/19/2011 1:11:57 PM
Event ID : 6008
Task Category: None
Level: Error
Keywords: Classic
User: N/A
Computer: ROB-PC
Description:
The previous system shutdown at 1:10:46 PM on ‎10/‎19/‎2011 was unexpected.
Event Xml:
<Event xmlns="http://schemas.microsoft.com/win/2004/08/events/event">
<System>
<Provider Name="EventLog" />
<EventID Qualifiers="32768">6008</EventID>
<Level>2</Level>
<Task>0</Task>
<Keywords>0x80000000000000</Keywords>
<TimeCreated SystemTime="2011-10-19T18:11:57.000000000Z" />
<EventRecordID>28013</EventRecordID>
<Channel>System</Channel>
<Computer>ROB-PC</Computer>
<Security />
</System>
<EventData>
<Data>1:10:46 PM</Data>
<Data>&#8206;10/&#8206;19/&#8206;2011</Data>
<Data>
</Data>
<Data>
</Data>
<Data>160938</Data>
<Data>
</Data>
<Data>
</Data>
<Binary>DB070A00030013000D000A002E00BA02DB070A000300130012000A002E00BA023C0000003C000000000000000000000000000000000000000100000020610100</Binary>
</EventData>
</Event>

 

[Rvenger]

I would try to replicate the issue before you blame the power supply. I would watch for voltage dips on particularly the 12v rail while benchmarking. Have a program run in the background to capture such.


EDIT: Come to think of it, you may be cutting it close with your Neo Eco 520watt. I think the Nvidias are a little more power hungry. My Antec BP550+ would dip down to 11.68v under load with a Radeon 5850.

 

[toyota]

I would try to replicate the issue before you blame the power supply. I would watch for voltage dips on particularly the 12v rail while benchmarking. Have a program run in the background to capture such.


EDIT: Come to think of it, you may be cutting it close with your Neo Eco 520watt. I think the Nvidias are a little more power hungry. My Antec BP550+ would dip down to 11.68v under load with a Radeon 5850.

I have tried running the bench over and over and even with overclocking, it will not shut off again . and no I am not even close to pushing the psu too hard. I already said power usage hit a peak of about 325 watts at the wall during that benchmark and that was including my monitor. so in actual watts, I am not even pushing the 12v rail much more than 50&#37;.

 

[Rvenger]

Do you have a shut down temp set in your Bios? I think there could still be a chance that Windows may not of noticed a thermal shutdown but your Bios did. It may not be overheating but a sensor could of thrown a misread.

 

[toyota]

Do you have a shut down temp set in your Bios? I think there could still be a chance that Windows may not of noticed a thermal shutdown but your Bios did. It may not be overheating but a sensor could of thrown a misread.

I will have to check later but if that was the cause I still do not see why the UPS would have be throwing off its alarm. it has to be power related in some way or the UPS would have done nothing.

 

[sm625]

There is seriously a game called "Hard Reset"? Maybe you found an easter egg? Every time it runs there could be a random low probability chance that the game will call rundll32 and force shut down your pc.

 

[Rvenger]

I will have to check later but if that was the cause I still do not see why the UPS would have be throwing off its alarm. it has to be power related in some was or the UPS would have down nothing.

Aw crap, this is what happens when I scan posts. Sorry about that Toyota. I would run without your UPS until you can replicate the problem. If the issue doesn't occur again, then it is your UPS. We had an issue at work where a bunch of 10 year old APCs were cutting out with the alarm beeps a couple years ago. Something is probably faulty on your UPS.

 

[toyota]

well if I could make it do it again at least once then I would try running without the UPS for a couple days. seeing as it wont do it again though that will not rule it out. plus our neighborhood has shitty unreliable power so I really want to keep using the UPS just in case we have an outage or brown out. funny that this is the first neighborhood I have ever lived in with underground power yet we have issues here all the time.

 

[toyota]

I just figured out the problem thanks to the xbit review of the 620 watt version of my psu. http://www.xbitlabs.com/articles/cases/display/psu-600w-850w_5.html#sect0

UPS Compatibility

Like its Seasonic progenitors, the Antec Neo ECO 620C refused to be compatible with my UPS. They were stable at loads up to 370 watts when powered by the mains but couldn&#8217;t work even at 280 watts on the UPS batteries.


what must have happened that day was a slight power loss or self test on the UPS that made it use the batteries for a split second. that of course forced my pc off since it was using 325 watts. that makes since because xbit says the other similar Seasonic built psus will not work on the batteries at over 280 watts.

so just to make sure, I fired up that same benchmark and then pulled the plug from the wall. sure enough the pc turned right off. needless to say I am extremely pissed off since I have no protection if the power goes off under load.

the problem is who do I need to be pissed at? Seasonic for making the psu? Antec since it has their name on it? or Cyberpower since their UPS is useless when my pc is under load?

 

[Slugbait]

I had a Cyberpower once. Any brownout, and my machine lost power. Occurred with both my Seasonic and Topower units. I tested it with a lamp and 60W lightbulb, and it worked fine when I unplugged it from the wall, so I gave it away at a garage sale...and IT WAS A HARD SELL. The guy actually had to think about it and almost walked away from the deal. Hey, whaddya want for free? Jeez...

 

[Slugbait]

I should say that I now have an APC, and haven't had a sudden shutdown in over a year...used to get shutdowns a couple of times a week with the Cyberpower.

 

[toyota]

well I need to find out what ups was used in the xbit review. if its not a Cyberpower like mine then getting another ups may not make any difference. they certainly seem to blame Seasonic for the issue so I will do a little more research on the issue tomorrow.

 

[VirtualLarry]

This is the problem with APFC PSUs and non-pure Sinewave UPSes.

 

[toyota]

This is the problem with APFC PSUs and non-pure Sinewave UPSes.

you do see that I have a pure sine wave ups though?

 

[houe]

What is the power factor of the cumulative load on your UPS?