Jump to content

Comp Shutdown Problem


Recommended Posts

Ok this is a common thread here on the forums, and I am sure alot of you are getting sick of having to give the same answers over and over again.

 

But this I assure you is not a common problem, I have been here and there and nobody can seem to figure it out including me.

 

Long Post for detail please read it and help me dont be scared!

 

Here is the lowdown.

 

I get random shutdowns of the PC, it does not matter if its idle, or in a game, or prime.. it is random.

 

Heat is not the issue: I am using a high end water cooling setup (swiftech h20 apex) so temps are always great, even PWMic's are cool with a fan blowing directly on them.

 

Power is not an issue: Only guess anybody had was my psu was the problem... I knew it wasnt since I had tested it vigerously with a multimeter and I am very well knowledgable with electronics and how they work.. but to eliminate that as a probem in a desperation I got the best of the best PC Power & Cooling Turbo 510 SLI so now I KNOW power is ok.

 

Memory seems to not be the issue: I had a setup prior to this one a opteron 146 Oced to 2.9ghz on air with my 1gb of OCZ ram, had it stable up to DDR600, that same ram was in this computer for a while and I had the problems, Its memtest stable overnight, prime95 stable, and game stable.. plus when I was first tweaking memory for stablity in high clocks if I did get it to an unstable value id get a restart, or a blue screen... not a shut down. So memory seems to not be an issue... right now im only running DDR500 @ 510 so im barly pushing it.

 

Factors: It happens more at higher voltages, when I had high 3ghz OC's with higher voltage (1.55v) it could shut down on me as much as 2 or 3 times a day, I noticed as I lowered my cpu voltage they happen much less often, and at stock voltage they "never" happend but latley I got 2 shutdowns in the last 2 weeks @ 2.8ghz 1.45v wich is pretty much stock... so its still happening to me even at stock voltage just not as often.

 

Things I have seen: My girlfriend actully saw this one day I had 1.7v for cpu and was running 3dmark05 for a bench score at 3.1Ghz and she saw a blue spark near the corner of the PC where my black light was... I have checked it and see no evidence of burns or sparks... but if the black light was grounding or shorting out I should see the light fail and thats not happening, and that doesnt explane why higher cpu voltage would be the X factor.

 

Other info: I never had a single shut down with the old cpu (the 146) and on air cooling.... but I also ran the cpu at stock votage 1.4v this started when I put the new cpu and water cooling in.. not sure if that has any bearing.

 

 

Leads

 

Lead1: My only leads right now are maybe the black light is shorting out, so gonna take it off when I get home and see if the issue goes away but I doubt thats the problem.

 

Lead2: I have bios 623-1 wich uses the yellow ram slots according to a post I saw here and also the manual tells me to use the yellow slots... while totally stable in all those test and games.. I just today read a post saying yellow slots have to use a 2t command rate to run correctly... this combined with my knowledge of the old memory controllers (clawhammer) that shut down the pc if running with incompatible memory setups (in the clawhammers case 4 dimms or 1gb dimms caused issues often) so maybe the cpu has an issue and shuts down the pc.. still I think id get a restart or a blue screen much like if I OC the ram too high not a total system shut down. If the black light doesnt fix, then I will try this next.

 

Besides those 2 things I am trying I am deserpate for help with the issue, any recomendations and help are very much appreciated.

Share this post


Link to post
Share on other sites

Ok this is a common thread here on the forums, and I am sure alot of you are getting sick of having to give the same answers over and over again.

 

But this I assure you is not a common problem, I have been here and there and nobody can seem to figure it out including me.

 

Long Post for detail please read it and help me dont be scared!

 

Here is the lowdown.

 

I get random shutdowns of the PC, it does not matter if its idle, or in a game, or prime.. it is random.

 

Heat is not the issue: I am using a high end water cooling setup (swiftech h20 apex) so temps are always great, even PWMic's are cool with a fan blowing directly on them.

 

Power is not an issue: Only guess anybody had was my psu was the problem... I knew it wasnt since I had tested it vigerously with a multimeter and I am very well knowledgable with electronics and how they work.. but to eliminate that as a probem in a desperation I got the best of the best PC Power & Cooling Turbo 510 SLI so now I KNOW power is ok.

 

Memory seems to not be the issue: I had a setup prior to this one a opteron 146 Oced to 2.9ghz on air with my 1gb of OCZ ram, had it stable up to DDR600, that same ram was in this computer for a while and I had the problems, Its memtest stable overnight, prime95 stable, and game stable.. plus when I was first tweaking memory for stablity in high clocks if I did get it to an unstable value id get a restart, or a blue screen... not a shut down. So memory seems to not be an issue... right now im only running DDR500 @ 510 so im barly pushing it.

 

Factors: It happens more at higher voltages, when I had high 3ghz OC's with higher voltage (1.55v) it could shut down on me as much as 2 or 3 times a day, I noticed as I lowered my cpu voltage they happen much less often, and at stock voltage they "never" happend but latley I got 2 shutdowns in the last 2 weeks @ 2.8ghz 1.45v wich is pretty much stock... so its still happening to me even at stock voltage just not as often.

 

Things I have seen: My girlfriend actully saw this one day I had 1.7v for cpu and was running 3dmark05 for a bench score at 3.1Ghz and she saw a blue spark near the corner of the PC where my black light was... I have checked it and see no evidence of burns or sparks... but if the black light was grounding or shorting out I should see the light fail and thats not happening, and that doesnt explane why higher cpu voltage would be the X factor.

 

Other info: I never had a single shut down with the old cpu (the 146) and on air cooling.... but I also ran the cpu at stock votage 1.4v this started when I put the new cpu and water cooling in.. not sure if that has any bearing.

 

 

Leads

 

Lead1: My only leads right now are maybe the black light is shorting out, so gonna take it off when I get home and see if the issue goes away but I doubt thats the problem.

 

Lead2: I have bios 623-1 wich uses the yellow ram slots according to a post I saw here and also the manual tells me to use the yellow slots... while totally stable in all those test and games.. I just today read a post saying yellow slots have to use a 2t command rate to run correctly... this combined with my knowledge of the old memory controllers (clawhammer) that shut down the pc if running with incompatible memory setups (in the clawhammers case 4 dimms or 1gb dimms caused issues often) so maybe the cpu has an issue and shuts down the pc.. still I think id get a restart or a blue screen much like if I OC the ram too high not a total system shut down. If the black light doesnt fix, then I will try this next.

 

Besides those 2 things I am trying I am deserpate for help with the issue, any recomendations and help are very much appreciated.

 

 

From my personal experience, every time I have had a "up one second, power off the next" situatiuon, it's been power related. I don't mean necessarily a bad PSU, I just mean something caused an over volt/current situation and the PSU shutdown to protect itself.

Just the other day I was messing with my CCFL lights with the machine running (yeah, I know) and one of the pins on an unused male molex somehow contacted a protruding metal piece of the case. There was a tiny blue spark and instanatenously the PSU shut down. Like I had yanked the plug from the wall.

I've even heard of bad fans causing PSUs to trip. Could also have a standoff or someting else under the mobo contacting something. Also, as far as this being related to OCing, it is possible that there is something pulling hard on one of your rails (that's not supposed to be) and OCing pushes it over the edge and causes the PSU to trip.

I know this would be a major PITA, more so even with the water cooling, but could you try running your rig outside the case for a few days?

Good luck

 

Steve

Share this post


Link to post
Share on other sites

Hmm all the wires are nice and neat, and the mobo is been remounted on the mobo tray with non conductive washers and screws so thats not it.

 

bad fan, bad neon light... that could be the case... wher ethe spark came from that one day was right where the inverter was for the black light, or the front 120mm fan i have it could have been eather one.

 

Thing is when I OC its just cpu voltage that im raising, as far as I know the rest of the system should stay totally the same.

 

sombody told me to up my northbridge voltage, said they needed theres higher to get a OC stable and had shut downs, i can look into that later.

 

also still wondering about the running 623-1 bios with yellow slots @ 1T sombody that knows about the slots/bios/memory configs can help me with that and tell me if this could be a possible issue, even tho I would expect a restart or blue screen if memory was the problem.

Share this post


Link to post
Share on other sites

  • 11 months later...

I have same issues like you my friend...

Everything you state in your post reminds me my problems with my DFI NF4 SLI DR & Opteron 165 UP...:eek2:

 

Still serching for all possibilities but i found something strnge....

My PSU is OCZ 600W Powerstream...

Look what i get when i testing my OPTERON even at Stock Vcore...:sad:

 

untitledky6.jpg

 

Bad readings ? or Reall PSU problem ?

I still dont know for sure, because my FLUKE multimeter never read such low vcore...

Share this post


Link to post
Share on other sites

Bluescreens can be caused by power supply. If you can try one from the recommended list you can see if your same issue reproduces itself. Check, re-check and check again to be sure all power connectors are plugged into your board. I had a bad connection on my 4-pin floppy connector and my board would start then shut off, or not even start. Sometimes these can be overlooked.

About the best way to troubleshoot an issue is to use the 'ol standby of process of elimination.

Share this post


Link to post
Share on other sites

got my first dfi ultra-d setup going last night and i'm having the same problem ast the op...

 

pulled these things that happened from my Hardforum post...

 

 

1) hum... well she powers up for a second .. beeps.. then just turns off...

 

 

2) well she posted and i got the lanparty spash screen .. it said cmos checksum error .. load defualts f1 to continue DEL to setup ... saw that it was a opty 146 and then just shoot off

 

3) ultra-d pencil modded to sli - first boot attempt since pencil mod

2x512 Corsaire 3500 BH5 in the orange slots

opteron 146

7600gt stock using a dvi-vga adapter in the right port of the dual dvi

audigy sound card installed

using all four power ports on the dfi board

dvdrw and hdd are connected via the two ide controllers

dvdrw is in the one to the right

hdd is in the one to the left

cmos was cleared but gives the checksum error noted above

starts, shows post, beeps, then just shuts down

running tdx, d5, bip3, maze4gpu w/ t line

don't think it's a heat issue

psu is an always steady xfinity 500w gen2

hum.....

any suggestions

 

4) so i got it stable long enough to get into the bios and i checked the pc health and it says cpu core temp 103C... then she just shut down

is that even possible?

 

5)experincing just random length shut downs .. sometimes it'll run for 5 min sometimes 30 sec... damn it

 

6)think maybe the bh5 isn't getting enough juice w/ the cmos clear?

edit: well i tried uping the vdimm a little, tried it with the drives connected and with them not... removed the sound card, down to just on 512mb stick i the top orange slot...

red led debug goes all the way down to the last one to the left .. i.e. red led to the far left stays lit... the yellow board power light is on...

i'm am out of ideas here ...

 

 

thanks for any help

Share this post


Link to post
Share on other sites

got my first dfi ultra-d setup going last night and i'm having the same problem ast the op...

 

pulled these things that happened from my Hardforum post...

 

 

1) hum... well she powers up for a second .. beeps.. then just turns off...

 

 

2) well she posted and i got the lanparty spash screen .. it said cmos checksum error .. load defualts f1 to continue DEL to setup ... saw that it was a opty 146 and then just shoot off

 

3) ultra-d pencil modded to sli - first boot attempt since pencil mod

2x512 Corsaire 3500 BH5 in the orange slots

opteron 146

7600gt stock using a dvi-vga adapter in the right port of the dual dvi

audigy sound card installed

using all four power ports on the dfi board

dvdrw and hdd are connected via the two ide controllers

dvdrw is in the one to the right

hdd is in the one to the left

cmos was cleared but gives the checksum error noted above

starts, shows post, beeps, then just shuts down

running tdx, d5, bip3, maze4gpu w/ t line

don't think it's a heat issue

psu is an always steady xfinity 500w gen2

hum.....

any suggestions

 

4) so i got it stable long enough to get into the bios and i checked the pc health and it says cpu core temp 103C... then she just shut down

is that even possible?

 

5)experincing just random length shut downs .. sometimes it'll run for 5 min sometimes 30 sec... damn it

 

6)think maybe the bh5 isn't getting enough juice w/ the cmos clear?

edit: well i tried uping the vdimm a little, tried it with the drives connected and with them not... removed the sound card, down to just on 512mb stick i the top orange slot...

red led debug goes all the way down to the last one to the left .. i.e. red led to the far left stays lit... the yellow board power light is on...

i'm am out of ideas here ...

 

 

thanks for any help

 

had a customer with a lanparty venus board with the same issue it would run for a while then just shut down and the power button did not power it back up until you unplugged it let it sit and then plugged it back in and it ended up being the power supply even thought my multimeter said it was in speck. he had a modular 600w Nspire power supply (nspire is same as Thermal take you could put them side by side with there 650w just some of the caps and coils r a little lower quality on the nspire hence the lower wattage.) so we replaced it and it has been bench stress testing for about 24 hours now no shut downs. I beleive it was a short inside the power supply and this may be the case for you as well. if the power supply detects a short it will hit a crowbar circuit to an open short to save your hardware i think this is possibly what your experiencing.

Share this post


Link to post
Share on other sites

maybe

 

hard to believe since this psu has never given me any problems before and always has been ROCK stable with other systems...

 

the things that have changed are

 

mobo

ram

cpu

vid card

 

the psu is the only thing that i have previous experience with it working properly so to point to it first as the piece that is wrong is hard to swallow...

Share this post


Link to post
Share on other sites

maybe

 

hard to believe since this psu has never given me any problems before and always has been ROCK stable with other systems...

 

the things that have changed are

 

mobo

ram

cpu

vid card

 

the psu is the only thing that i have previous experience with it working properly so to point to it first as the piece that is wrong is hard to swallow...

 

barrow a buddies power supply before u run off and buy one cuz it may not be one. but i figure if its the only thing that isnt new it may just be tired from all the hard ware ur throwing at it as well as age is a factor how old is the beast? and i just replaced my 550W with a new PSU cuz it was flaking out i was drawing to much on spontanious load and it was only 5 months old. sold herto a pal who has a lower end system and its working for him so yah psu's can act very strange to say the least

 

oh and could you please fill in your signature with your system specs so we all know what ur running under the hood bud.

Share this post


Link to post
Share on other sites

Ok guys, I was having the same problem as you, and it was driving me insane.

 

What I did to fix it, you'll probably *NOT* want to try, because it's expensive. However, I was desperate. This is what I did (this is going to be a little long):

 

Initial Hardware: 4400+, Aspire Q-pack (and that hunk of crap PSU), 6800gt, Patriot memory, SB A2 ZS plat

 

1. I started getting the shut down problems...so, I took out all of the unnecessary stuff, and it was still happening. The PSU died a week later.

 

2. Buy Enermax PSU, highly recommended, works fine, until it continues shutting off. I become confused and wonder what to do next.

 

3. I break down and buy TT armor case and DFI motherboard (SLI-D). Best change I could have done, but...it still shuts down.

 

4. At this point I've figured out what memtest is and I find that one of the DIMMs isn't working (lots of errors at really loose timings). Patriot Tech Support was really awesome, and Newegg was really fast about getting me new memory. Both sticks memtest fine, however, the problem's still there. I'm confused.

 

5. I try a buddy's graphics card in my system (7800gt) and it is seemingly stable. However when he tries mine in his, he has no problems. I RMA the graphics card anyways. Problem is still there. Yuck.

 

6. At that point, I try his CPU (a 3800+ x2). And, to no avail. I'm really confused. I've tried literally every single component in one way or the other.

 

7. Fast forward to a little closer to the future: I hear a slight buzzing coming from the Enermax, and I don't want to worry about that going bad and frying something, so I call up Coolergiant (Enermax subsidiary in the US) and the RMA is on the way. New power supply has the same problem.

 

8. At this point, I'm about ready to throw my monitor out the window. But, instead, I buy an Expert board and an Opteron 144. Build it outside the case. Not a single problem. Boots at 200x9 (primes for 99 hours for crying out loud) and I can't kill the thing. I hop right up to 289x9 and I've been happy since.

 

 

 

Now along the way, there were countless prime tests (both working, and not working), memtests at any time I changed any timings, CMOS clears (short / long / whatever). I also bought a x850xt somewhere in there for testing, but it didn't make a difference.

 

If in doubt, take everything apart. Build it from scratch with the bare minimums and see if you've got a problem. Follow the initial build guide and do it outside of the case.

 

If I had more time right now (read: no finals), I'd be swapping components to figure out what works and what doesn't, but as it stands right now, my 4400+ is sitting in my SLI-D and I haven't touched it since I got the 144 working in the Expert.

 

Sorry for such a long post, but I just want to say, don't get too frustrated, you'll eventually figure it out. :sweat:

Share this post


Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

 Share

×
×
  • Create New...