Shutting down under load

[ketchup79]

If you haven't done so already, make sure there is no movement available where each screw attaching it to the case is. And make sure there is a screw for each hole. As I said a few days ago, it could just be that at least part of your motherboard is fried.

[cody0851]

Picture this(my hsf assembly, following the given directions.)
side of metal case
metal plate
rubber pad
screws through plate and rubber
motherboard
nonconductive washer
nut
metal bracket holding hsf
nut

Say those screws were possibly touching the metal case.. would that cause a short? Of course the psu is touching the case, too.

When I tried my spare psu, I'm not sure if it was touching the case or not. Grabbing at straws here..

[ketchup79]

Originally Posted by cody0851

When I tried my spare psu, I'm not sure if it was touching the case or not. Grabbing at straws here..

Same here dude. Some people have luck running testing outside of the case. I have never done so, but there are instructions out there.

[cody0851]

Ok, plugged my new 850 into a known working computer. It booted up, but very slow. I restarted, and it booted faster. The computer itself hasn't been used in a year, so I figure that's just going to happen.
I put that computers normal psu back in and it booted even faster.
(2 ide dvd drives, sata hd, amd brisbane 2core, onboard graphics)

So.. I've got to ask if a psu can cause a system to run slow? If yes, then the psu is bad. If no, then my m/b or cpu went bad.

And then, if my new 850 ISN'T bad, why would it not boot my computer when the other psu would?!

Makes no freaking sense!

[mlc]

did you try the working computers psu in the one you have problems with ?

[cody0851]

Yea. It worked.

To clarify- it worked with just cpu, ram, and m/b attached. It gave me a beep code, and shut off. I didn't add peripherals b/c I didn't know if something might fry a known working psu.

[cody0851]

Update -
Hooked up m/b, cpu, 1 stick of RAM outside of case with 850 psu. It actually booted to 1 long, 3 short beep code. Added video card. It posted normally.

Must have been a short in my case. Looking into new cases now.

[mlc]

glad to hear ...

but you don't really need a new case .. you just need to ensure when you remount the board into the case.. that the standoffs match up properly with the motherboard mounting holes..

also make sure the i/o plate is lining up properly with the mobo when you reinsert.. Same problems will existing with a new case if you don't take these precautions

good luck
Mike

[cody0851]

Thanks. I hear you, mlc.
A new case would give me more room to work, and more air flow. This case has no m/b tray, no cable management holes, and only 1 side comes off. The front fan is a 90mm, which even using one known to be super quiet, was far too loud for me. With the Xigmatek cooler, I couldn't even remove the psu without taking out the m/b.
I'm looking at these:
Antec 302 70$
Corsair 300r 90$
Corsair 200r 50$
NZXT Source 210 50$
Xigmatek ASGARD II 36$

[mlc]

ah... ok.. makes sense.. .

FWIW, i have 2 antecs .. the 300 and 900 models.. both are great in terms of airflow, especially if you get one with the blow hole on top..

[cody0851]

Update-
Finally received the Corsair 400r. Installed everything. Powered on. Excellent.
Ran CPUIDHW. Temps at idle in 40s. Excellent.
Ran IBT. It passed all 10 tests. Temps in the 50s. Tests were each 2-3 seconds faster than before. Excellent.

Did some catch up work on the net. Facebook, etc. Turned on youtube, angry birds. Adobe flash crashed repeatedly. Not so good.

Rebooted. Ran IBT and CPUIDHW again twice. Temps hit 62c, IBT stopped on test 2, and said my cpu is unstable. Facepalm.

Annnd I'm not sure what to do now. Perhaps my cooler needs break in time?

[Smoove910]

Perhaps you need to reinstall your heatsink... ensure you only use minimal thermal paste (size of a grain of rice). Set heatsink onto CPU and twist slightly, then ensure it is latched and a tight fit.

Also, how is the airflow in your case? Are you getting negative pressure somewhere? Should be pulling air in the front, out the rear, and up through the top if you have a fan up there.

[cody0851]

I should have plenty of air moving. 2 120mm intakes up front(only 1 hd behind them.) 1 120mm out the back. Cpu has a 120mm pushing air through the heatsink toward the rear fan. And lastly the psu is sucking air up into the bottom.
If that's not enough, we are under a tornado watch.. if I need that much wind, I'm in trouble.

[ketchup79]

Well you temps aren't bad. Even if they were the CPU would just throttle anyway. Could still be that other stick of RAM. It could be the CPU as the program you mentioned suggests, but that would be extremely rare.

[Smoove910]

By chance is your CPU voltage set to 'auto'? If so, I would recommend setting it to default of what your CPU requires. If on 'auto', it could be getting more juice than it requires, which causes temps to get out of control... resulting it unstable system.

Also, I see you have DDR2, what is your voltage set to on the ram? I'm thinking something like 1.85 ~ 2.1v is standard on that ram, but not sure since I don't know what you have.

EDIT - Saw the kit you linked... I would ensure it has 1.8 ~ 1.85v going to it and the timings are set to 5-5-5-18 per Corsair.

[cody0851]

I ran memtest on each stick, with no errors.
CPU and ram are set to auto. Memory timings are already 5-5-5-18.

I ran prime95 for an hour, with no problems. IBT tests are taking 20seconds longer that normal, and fail at the beginning of test 3.

I've never oc-ed, so all the settings have been in place for years.

[Smoove910]

try taking voltages off auto and put them at what they should be (rather than have the BIOS determine it)

[buggers]

um..all the things well go well.