DeNs Posted January 30, 2006 Posted January 30, 2006 Specs: Opteron 146 CAB2E 0546FPAW DFI NF4 eXpert SLI-DR Revision R.A02 - Latest official BIOS (12/7/05) OCZ Plat EL R2 TCCD 2x 512MB XFX 7800GTX OC OCZ Powerstream 520W PSU Custom Watercooling: Swiftech MCP600, Silverprop EvoSE, BIX2 6x WD HDDs The problem I've been having happens with only one common thing - When the CPU is overclocked (and more often that not, under load), the system completely shuts off, as if I pulled the power plug straight out of the wall. It has happened twice when doing nothing to stress the system, but it mostly happens under load. It's probably happened well over a dozen times to me now. I've tried: OCZ 4000VX and my TCCD PCI Matrox Video and my XFX 7800GTX OC Almost every kind of voltage/multi and mem option available to me Today I tried to find out what triggered it quickest, using Folding@Home to stress the cpu most. These were my results. Times on the right indicate how long it took to crash. 290x10 - 2900MHz - 1.47v + 107.2% - 1.4v - 1.56v - 200 | 2.5-4-3-7-1T 2.79v - 15 Minutes into F@H 290x10 - 2900MHz - 1.47v + 107.2% - 1.4v - 1.56v - 166 | 2.5-4-4-8-2T 2.79v - 30 Minutes into F@H 290x10 - 2900MHz - 1.55v + 102.4% - 1.4v - 1.59v - 166 | 2.5-4-4-8-2T 2.79v - 1 Minute into F@H 317x9 - 2850MHz - 1.47v + 104.8% - 1.3v - 1.56v - 166 | 2.5-4-4-8-2T 2.79v - 1 Hour 50 Minutes into F@H Monitoring the 12v, 5v and 3.3v rails with a DMM shows that they're rock solid. This is with both CPU and 3D stress with the GTX, as well as anything else I could throw at it. The PSU does not seem to be at fault, and even through I can't rule it out altogether, the Mobo seems much more likely the cause. This happens with both the default BIOS and the latest, which I've been using to test for the last week or so. I would dearly love to get one of the new revision boards which i've seen perform so well, but it really pains me with these kinds of problems - going through the same damn thing that I went through with my NF3 Ultra-D last year I fear there's no hope for the board, but I can't imagine this being a suitable excuse to RMA the board, even though i'd love to be able to. If anyone could help I'd greatly appreciate it. ---dens Share this post Link to post Share on other sites More sharing options...
Slammin Posted January 30, 2006 Posted January 30, 2006 Unplug a few of your HD's and see if it still happens. As much as I preach that PSU, my bet would be that you may be asking too much from it. Share this post Link to post Share on other sites More sharing options...
DeNs Posted January 30, 2006 Posted January 30, 2006 Ok i'll give it a try now - I'll just use the 2x 120GB drives (Windows + Apps pretty much) and try 2.9GHz (makes it quicker to see if It'll crash). If it turns out to help it, then bummer, but at least i'll know what it is. I'll try anything at this stage Thanks, and i'll report back soon. ---dens Share this post Link to post Share on other sites More sharing options...
DeNs Posted January 30, 2006 Posted January 30, 2006 Nope, after about 15 minutes at 2.9GHz the system just completely shut itself off again. This is with 2x HDDs connected only. The power draw absolutely cannot be too much for the 520W. Unless there's an automatic heat shutoff thing with either the board or the PSU, I can't think of anything else to try for now. I know the LP NF4 series boards had additional connectors for general stability but the eXpert does not, apart from the one floppy connector, which as far as anything i've read tells me, this is only a connector for using SLI. At the moment this clearly looks like a heavy load issue, so is there anything else board or PSU related that could possibly help? ---dens Share this post Link to post Share on other sites More sharing options...
Slammin Posted January 30, 2006 Posted January 30, 2006 Could be something in the thermal shutdown circuit on the mobo? If you are 100% certain your CPU isn't overheating then you could try disabling thermal shutdown in the bios. Just be careful and keep an eye on things. Share this post Link to post Share on other sites More sharing options...
DeNs Posted January 30, 2006 Posted January 30, 2006 I never have thermal shutdown enabled in the BIOS, generally because it's disabled by default on any BIOS that i've used. I trust my watercooling and my mounting. However i'll boot it back up and give it a shot now. Edit: Yep, just as I thought - already disabled. ---dens Share this post Link to post Share on other sites More sharing options...
Slammin Posted January 30, 2006 Posted January 30, 2006 Maybe your PSU is going into overload protection and shutting down. Do you have another you could test with? Share this post Link to post Share on other sites More sharing options...
DeNs Posted January 30, 2006 Posted January 30, 2006 Overload protection with only Mobo, CPU, PCI Video and 1x HDD? This thing worked with a highly OC'd Venice on a DFI NF3 Ultra-D with 6 HDDs and an OC'd 6800NU. I would hope to think that it could handle this.. My mate has a Delta 600W Quad-12v Rail and an OCZ 600W Powerstream that I can test at some stage, but I don't personally have another 24/8-pin PSU that I could test it with. Thanks for keeping the ideas coming ---dens Share this post Link to post Share on other sites More sharing options...
DeNs Posted January 30, 2006 Posted January 30, 2006 Update: Well it's been going for almost 4 hours running F@H at stock 2GHz 1.375v so I think it's safe to say that I won't have the shutdown problem at stock speeds. Seems like this is mainly an OC problem. Tomorrow I might give the Silicon Image controller a go to see if there's any effect on OC when i'm using the nVidia ports which i'm using now. Update2: Silicon Image controller doesn't seem to like Non-RAID setups so I just didn't bother. ---dens Share this post Link to post Share on other sites More sharing options...
Recommended Posts
Please sign in to comment
You will be able to leave a comment after signing in
Sign In Now