Post prior to working on a pc..

Rastalovich

New member
B4 I go ahead and get sticky wrist off an ABit KV7, hosting an AthlonFX, lets see if any1 recognizes this behaviour.

Out of the blue the pc will shut down, not from doing anything in particular, just randomly just turn off.

The shutdown is similar to a "Fail At" shutdown, in that to get it to immediately turn back on you need to press the front case on/off button several to a dozen times b4 the BIOS attempts a last-state turn on.

Of course the pc turns on and the bios is scrubbed. Leaving no alternative but to clear the cmos and readjust the bios setting b4 you boot into OS (Win XP Pro).

Pc can be fine from anything between 5 seconds and 24 hours and the cycle will repeat.

You don`t even need to do anything with the pc for it to do it, you can just be in the same room and hear all the fans turn off - O o

Rough Specs:

ABit Kv7

AthlonFX 2600+

5200FX

512m

2 drives 1x ATA 1xSATA

300w psu

Not a gaming pc, just plays DVDs/AVIs to the TV and the odd DVD burn/convert - nothing too strenuous, pretty lazy really.

I`m thinking BIOS alongside it`s CMOS power etc. Maybe even psu failiure, that`s as old as the Kv7 >.< I did however see the voltages fluctuating on the BIOS Health section when brought up. The 12v and 5v rail was dropping from something like 11.98 to 11.75, and similarly the 5v, and noticed the cpu fan rpms dropped and rose to suit. mm we`ll see.
 
It`s a Hiper, actually states 350w next to the zimmer-frame. SF-350TS

The model in this test: http://www.nordichardware.com/Reviews/?page=15&skrivelse=366

However it`s got to be maybe 5 years old, and very rarely been turned off during this period. It`s wholely possible the inside of the chasis is crammed.

I don`t comprehend the BIOS going if it`s just the psu failing during operation. Normally if you physically rip a psu out of a pc, the battery takes care of the cmos storage. Could be that in turn the battery has fried somehow.
 
name='Rastalovich' said:
Of course the pc turns on and the bios is scrubbed.

Usually a flat CMOS battery is indicative of the problem mentioned above. I had the same issues with an Asus P4 board of mine quite a long time ago.

Out of the blue the pc will shut down, not from doing anything in particular, just randomly just turn off.
I'm thinking along the lines of WC's post above. You say the PSU may be around five years old, there's a good chance that it's clogged with accumulated dust and heating up...tripping the thermal protection. Do you have another PSU that you could hook up to see if the Hiper is causing the issue? Errors caused by heat within your PC are notoriously difficult to troubleshoot because usually once the component has cooled down everything's fine again. I'm putting my bet on the PSU!

Keep us posted on your progress mate :)
 
Yeah, I couldn`t look at it yesterday due to some1 elses pc probs, but I think you guyz may`ve hit the problem on the head.

I do have a number of psu about the place, think they`re mainly of the freebee 275w variety tho, but if the Hiper is nicely screwed together and not freakin rivetted, I may well `jet wash` the inside of it as one of the first options.

Thanx guyz.

-----------------

Ok, before I do anything else, I`m going to get a new psu in. I`ve un-crammed this psu and first turn on lasted seconds. I disconnect some power and it stays up allowing me to see the pc health bios screen.

cpu temp starts @ 45 and rises slowly, is currently 49. (it`s flicking 49/50 as I close the post)

cpu fan is flippin between 3276rpm to 3245. (which seems low to me, 3308 on closing)

cpu core flicks from 1.61 to 1.63/4 regularly

3.3 = 3.29 & 3.31

5 = 4.99 % 5.02

12 = 11.79 & 11.85

vddq = 1.53

ddr = 2.65

batt 3v = 2.97 & 2.99

batt 5v = 4.8

But it`s all flicking and changing pretty regularly, doesn`t seem right.

Gonna get a psu and a flowery cpu hsf.

Thnx.
 
Rastalovich said:
3.3 = 3.29 & 3.31

5 = 4.99 % 5.02

12 = 11.79 & 11.85

vddq = 1.53

ddr = 2.65

batt 3v = 2.97 & 2.99

batt 5v = 4.8

dont trust those numbers, they are often wrong.
 
Id say your motherboard is dieing my msi did same thing it would power off at random eventualy after a day of doing it. Mine started when i powerd down one day to clean video card and blow out dust when i put it back together problems started then. It would no longer post and i tested mutiple psu's i had to see if it was problem and none of them would get it to post.
 
Multimeter verifies the 12v 5v and the 5v 3v batts to be what the bios reports, fluctuating too. The settled reading is/was merely an average.

I`ll check the schem of the connector to see where the best place to jam probes to test the 3.3 vddq and ddr.

It`s still on btw, but that doesn`t mean much - if it`s on next week that`ll be a different matter.

Shopping anyway.

Don`t electrical components offer a life expectancy of something like 50000 hours ? (under perfect conditions)

365 x 5 x 24 = well over 43000 hours, that`s under hot smokey conditions, well past it`s lifetime :P

(which can be said for both the mobo and the psu sadly.. eek and the cpu)
 
If you dont want to buy another PSU then run teh machine off two working ones. One connected to the Mobo (ATX20/24 pin + CPU 4 pin) & one running drives and fans. Like this:

26-01-06_0010.jpg
 
I meant to get back to ya`ll - the problem was the psu, it wasn`t the case that the bios was blanking - more to the fact it was me that was making it blank.

The pc I found could, after shutting down by itself, be turned off @ the psu i/o button - then back on - and if you have "last state" set on the bios power failure setting, presto you get the boot into last state. And on it would stay for a random amount of time - i.e. until the psu wasn`t happy, then rinse & repeat.

Thanx for the input.
 
Back
Top