Computer Lock ups

[Chipmyster]

Already ahead for taking the side off and blowing air into it, still crashes.

I haven't tested the psu with a multimeter as such, I do however have readings from speedfan if they are of any use to you.

Vcore: 1.17v
+12v: 9.03V
AVcc: 3.25V

 

[Route44]

Hmmm, what is the recommended voltage for the vcore etc.? Have you overclocked? If so, have you tried easing back on your timings, etc.?

 

[nickc]

that +12v is terrible. a new power supply would help everything in that computer.

 

[Chipmyster]

It is nickc? What should the +12v at then? And I would need to find a power supply to test this with somehow o.o

 

[Route44]

That's what I thought too, Nickc, and I left it out in muy last response.

that 12v should be at 12v.

Craftsman makes a nice digital multimeter for $20.

Also, if I didn't already have it I'd definitely get this: http://www.zipzoomfly.com/jsp/ProductDetail.jsp?ProductCode=371324&prodlist=celebros

 

[Chipmyster]

Well I feel stupid for not noticing that one! That is however intresting, is there a reason behind that other then the psu is just not doing what it should be or could it be something more trivial such as something just not being connected as it should ( I should hope not )

 

[Route44]

Good question.

Definitely check all connections including your fans. Sometimes that is all it takes.

However, I certainly would check to see if your PSU is stable and checking your voltage output is a great way to do this. When my old P3 system began having issues I suspected my Antec 350w power supply. After testing all connectings I realized it wasn't the psu afterall but it certainly helped me narrow down my trouble shooting.

The thing with a multimeter you need the 24 pin connected to your motherboard for the psu to work but if you connect the 24 pin to this it will run fine. Here is another suggestion and it tests 6/8 pins as well:

http://www.zipzoomfly.com/jsp/ProductDetail.jsp?ProductCode=360903&prodlist=celebros

 

[nickc]

as a PSU gets older it gets where it does not supply the power it is required and maybe I missed it but u did not say how old it was, also if not a good PSU it can go bad in no time. I have seen cheap PSU go bad in a month and also there has yet to be anything built that does not ware out.

 

[Chipmyster]

I will do this the moment I have enough time to check everything, this is the first major lead I have had in the whole time I've experienced this!

I will be purchasing a multimeter, the one you suggested is cheap and cheerful and certainly worth getting in case I ever have issues again ( I now officially hate computers ) this could take a few days to get around too with a busy week ahead but as I have this thread bookmarked Ill post back when I have some results!

Edit for NickC, it's roughly 1 month or so old now. It's corsair HX1000W PSU!

 

[Route44]

Corsair makes very fine psu's but even they can crank out an unstable one now and then.

There is some great online documentation that can help you read your multimeter correctly like where to place the black and red probes to get the various voltage reads. Keep in mind not only can lower than expected voltage read outs means a problem but also higher than expected as well.

Let us know how it goes.

 

[Chipmyster]

Will do! And understandably there will always be black sheep when things are made, at least it if fails the test I can rectify the problem fairly quickly!

And thanks for all the pointers so far, I would have completely overlooked the fact the fact the voltage was too low!

 

[gbhall]

Those temperatures worry me....still a bit of concern must remain.

(a) the PSU might only misbehave under stress, so will test ok on the plug-in type tester. (Possibly the unit makes allowance for this by applying a load).

(b) the PSU might be working perfectly ok, but there is a short on the motherboard, and that enormous and sophisticated PSU is compensating for it. The PSU would then test out fine on it's own, Worse, if you tested the motherboard with a substitute PSU of much less sophistication, it might blow up the motherboard....

 

[Chipmyster]

I have no idea why the temperatures are so high but anyway...

If I were to test the motherboard I would still be using a 1000w one regardless as I do require that power so if there IS a problem with the motherboard at least it shouldn't explode.

And I can't say the computer dies under stress it really is random, I lasted all yesterday without a freeze, but today I've already had one, absolutely no pattern to it at all!

 

[insoman]

I have Vista x64 also and have found that the sound card drivers are not so compatible. (had the same problem albeit with '97) I eventually found a x64 compatible driver. Keep looking and when you think you have found the correct driver look again.

 

[Chipmyster]

I am confident that I have the correct drivers for everything! I did a huge amount of checking for awhile but I can check if any have been updated!

 

[insoman]

Right click computer go to properties, advanced system settings, advanced, performance settings, advanced, change virtual memory to 6000...(or max) unless you've tried it already

 

[gbhall]

Well, to me, having things much hotter than they should be, and a big drop on the 12v rail shouts fault ! power drain excessive on something, dragging the voltage down and raising the heat !

The obvious thing to do is use your hand - find something much hotter than it should be, and disconnect it. In fact, as I said way back near the beginning of all this, your first test should have been to strip out everything to just 1 HDD and video. Disconnect everything else - even remove your sound card if you have one. See how you go then if it still locks up, you are basically down to PSU, CPU, video card and motherboard. Otherwise you will find the faulty device just by adding them back one at a time. How hard can that be?

Please read and follow post#6

 

[Chipmyster]

I can't reply to your PM GBhall due to not having enough posts but I can respond here!

Firstly, thank you very much for the reply no matter how late!

Secondly, I have a LONG list of stuff to report back.

In order.

(a) + (b) I changed harddrives with my old computer and it still crashed. So that was XP and a working HDD.

(d) overheating - The system on the Nvidia temps is sitting on around 40C. The CPU however varies between 40-60. the GPU is around 40-60. Up to 70 if it's being strained.

I also took out 2GB of ram every time it crashed I moved the ram 1/4 slots down. And did the same for the second stick. This made the computer WORSE and the 2nd stick was moved all the way through the 4 slots in a night, whereas the first took 3 days.

So It's looking like mobo trouble.