Jump to content
Corsair Community

HX1000 restart issues


Souls Collectiv

Recommended Posts

In preparation for an upcoming i7 build, I bought an HX1000 yesterday. Since I'm still waiting on a motherboard, I replaced the (other brand) OP850 in my current system (see profile specs) with the new PSU, and have been having issues ever since.

 

Seemingly at random, my system will restart. No error beeps, no BSOD, just restart (similar to the issues described in this thread). This can happen when POSTing, sitting idle at Windows desktop, gaming, Photoshop work, video editing, whenever. However, the strange part is, it occurs less frequently when the system is under full load - I was able to leave my system up running Prime95 Small FTTs, Linpack and FurMark simultaneously (which according to my UPS drove power consumption up to around 600W from the wall) for much longer than it usually takes before restart, and I can run 3D video apps for much longer before restart if I have Prime95 running in the background.

 

The PSU will turn on normally when using the short-the-green-wire test, and the system runs normally when using the older OP850. I've tried removing and re-seating the motherboard in the case in case of a short somewhere, but no change.

 

Anything more I should do to try and isolate the problem? These issues have been occurring ever since I replaced the PSU, and the PSU was bought only yesterday.

Link to comment
Share on other sites

  • Corsair Employees

There are a few possibilities:

 

1. Something is overheating, check your temps next time following a shutdown.

2. Possibly a memory problem, use http://www.memtest.org to make sure you are not getting any errors.

3. The motherboard or some other component is going out of spec and tripping the overcurrent/overvoltage protection circuits in the PSU. Try the PSU in a different system to see if you have the same problems.

4. There is a faulty component in the PSU.

 

I would go down the list and see if you can narrow down the possibilities. If it looks like there is a good chance that a replacement PSU will resolve the issues then, please use the On Line RMA Request Form and we will be happy to replace it. Be sure to check the box that says “I've already spoken to Technical Support and/or RAM Guy.”

Link to comment
Share on other sites

There are a few possibilities:
Thanks for the reply, appreciate the input.

1. Something is overheating, check your temps next time following a shutdown.
I don't think temps are a problem - mobo and CPU are both under water, and temps don't reach over 50 degrees under full load. It's harder to tell with VGA, but according to RivaTuner and GPU-Z, none of the component temps it has sensors for go over 80 degrees at load, which is well within spec. And even so, surely if there was a temperature problem it'd show up when being powered by the OP850?

2. Possibly a memory problem, use http://www.memtest.org to make sure you are not getting any errors.
I'll check with Memtest again, but I ran a test just a few days ago when trying for 1600 CAS-6 and was able to complete five full passes without error.

3. The motherboard or some other component is going out of spec and tripping the overcurrent/overvoltage protection circuits in the PSU. Try the PSU in a different system to see if you have the same problems.
I'll try that. I'm not sure if I'll be able to provide an appropriate test, however - the only other PC in the house is a file server, powered by a VX450, and which barely draws 120W under full load.

4. There is a faulty component in the PSU.

 

I would go down the list and see if you can narrow down the possibilities. If it looks like there is a good chance that a replacement PSU will resolve the issues then, please use the On Line RMA Request Form and we will be happy to replace it. Be sure to check the box that says “I've already spoken to Technical Support and/or RAM Guy.”

If after working through your list, and given that I'm in Australia and it'd be prohibitively expensive to send a PSU off to the USA (as I had to when RMAing my RAM), would it be possible to get an RMA number so I could then take the PSU back to the shop I bought it from and let them deal with it?

 

I've been trying to narrow down the problem all afternoon, and during that time the frequency of rebooting has increased, particularly when the system is under full load, which is the reverse of what I was experiencing earlier. System still works fine when powered by OP850.

Link to comment
Share on other sites

  • Corsair Employees
And even so, surely if there was a temperature problem it'd show up when being powered by the OP850?

 

I would agree.

 

I'll check with Memtest again, but I ran a test just a few days ago when trying for 1600 CAS-6 and was able to complete five full passes without error.

I'll try that

 

If you already ran memtest and got no errors, then the memory should be OK.

 

I'm not sure if I'll be able to provide an appropriate test, however - the only other PC in the house is a file server, powered by a VX450, and which barely draws 120W under full load.

 

Maybe a better test would be to use your VX450 along with the video card you are using with it, in your other system, and see if you can duplicate the same issues. If the issues go away, then it's either the HX1000 causing the problems, or possibly the video card could be going out of spec and causing the PSU to reboot (its possible that our HX1000 is more sensitive than your 800 watt PSU).

 

If after working through your list, and given that I'm in Australia and it'd be prohibitively expensive to send a PSU off to the USA (as I had to when RMAing my RAM), would it be possible to get an RMA number so I could then take the PSU back to the shop I bought it from and let them deal with it?

 

You can use use the On Line RMA Request Form and we will be happy to replace it. Be sure to check the box that says “I've already spoken to Technical Support and/or RAM Guy.” You would need to contact the reseller directly to find out if they can do the replacement for you, or at least help you with the RMA process.

 

I've been trying to narrow down the problem all afternoon, and during that time the frequency of rebooting has increased, particularly when the system is under full load, which is the reverse of what I was experiencing earlier. System still works fine when powered by OP850.

 

I would make sure you can still get the system to cause the same issues when running in safe mode. If you do not have any problems when running in safe mode, then there could be a problem with your drivers.

Link to comment
Share on other sites

Maybe a better test would be to use your VX450 along with the video card you are using with it, in your other system, and see if you can duplicate the same issues. If the issues go away, then it's either the HX1000 causing the problems, or possibly the video card could be going out of spec and causing the PSU to reboot (its possible that our HX1000 is more sensitive than your 800 watt PSU).
I would agree - except that the VX450 isn't capable of powering an HD4870X2 - a) it doesn't have the grunt, and b) there aren't enough PCI-E connectors.

I would make sure you can still get the system to cause the same issues when running in safe mode. If you do not have any problems when running in safe mode, then there could be a problem with your drivers.

Safe mode is problematic, because of course I can't produce meaningful system load. Furthermore, if this was a driver issue, why has it only started to show up once I swapped in the new PSU, and why don't I see the same problems when using the OP850 to power the system?

 

Further testing has narrowed down what will cause a restart, and it seems to be related to graphics. Heavy system graphics load is now guaranteed crash - eg. Furmark now reboots only a minute or so in, along with COD4/5/FEAR/Oblivion/Fallout 3/etc. Prime95 etc. seem to produce crashes no more often than with the system at idle. Idle crashes still occur, but these are reduced now that I've drastically dropped my 2D clocks to only 300MHz core/mem on both cores. I think I was seeing false patterns to start with when I said that there were fewer crashes at load, or the problem has worsened as I've been testing.

 

Given that I'm only using the hardwired PCI-E connectors, I'm inclined to think that there's either a problem with my VGA that shows up only with the HX1000 (and therefore is related to out-of-spec draw), or there's a problem with the 12V2 rails, if not the PSU generally. I'm going to go swap the hardwired connectors with the modular connectors and see if it changes things, and try and grab another VGA with a reasonable power draw to double-check tomorrow.

 

I'm a little dubious about the out-of-spec possibility, though - the OP850 is a very high-end unit, as I'm sure you're aware. If we were dealing with a no-name crappy knock-off, I'd agree that this could be a viable explanation, but as it is I'm doubtful. To check, I borrowed an HX620, shorted the green wire, and connected it to the HD4870X2, leaving the rest of my system on the OP850 - system ran fine.

 

Edit: I've just tried swapping the VGA over to the modular cables (ie.12V1). No change.

 

I've also tried using the VX450 to power the motherboard (ATX 24-pin and EPS), with the HX1000 powering everything else, including the 4870X2. Works fine. Swapped the HX1000 back to powering everything, problems start again.

 

So, we know that the graphics card isn't the issue (two separate PSUs have no problems with it, and the HX1000 is capable of powering it if there is no other significant load), nor is the motherboard or CPU (two separate PSUs have no problems powering the rest of the system, and the HX1000 is capable of powering with a minimum of reboot issues if it's not also required to power the VGA). To the best of my ability to determine, nothing is overheating, nothing is unstable due to OCing, there're no software issues, the HX1000 simply causes reboots if the power draw reaches over ~400W.

 

Dead PSU?

Link to comment
Share on other sites

  • Corsair Employees
I would agree - except that the VX450 isn't capable of powering an HD4870X2 - a) it doesn't have the grunt, and b) there aren't enough PCI-E connectors.

 

I meant to use the same video card that you were originally using with the VX450.

 

Safe mode is problematic, because of course I can't produce meaningful system load. Furthermore, if this was a driver issue, why has it only started to show up once I swapped in the new PSU, and why don't I see the same problems when using the OP850 to power the system?

 

Thats a good point, and its probably a long shot, but I wanted to rule out the possibility that the graphics driver itself was causing the card to pull more volts from the 12v rail that what is safe and tripping the overvoltage protection circuit on the PSU. The overvoltage circuit on our PSU could be more sensitive than other PSUs.

 

At this point it seems like the issue is either with the PSU or the video card, but since you have had no issues when using the other PSUs I would say the PSU is probably a little more likely. Lets get it replaced! Just use the link I posted in the last post.

Link to comment
Share on other sites

I meant to use the same video card that you were originally using with the VX450.
Ah, gotcha - but this wouldn't have been possible anyway - the file server uses an IGP.

At this point it seems like the issue is either with the PSU or the video card, but since you have had no issues when using the other PSUs I would say the PSU is probably a little more likely. Lets get it replaced! Just use the link I posted in the last post.

I took the PSU back to the point of purchase along with a printout of this thread, and based on your troubleshooting they were happy to take it back, no questions asked. System is so far completely stable with replacement PSU, no issues at all.

 

Thanks for your help!

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

×
×
  • Create New...