Interesting Problem With ASUS K8V Board

[unkn0wnvariable]

Here's a good one for you.

If the ambient temperature in the room is below approx 18C the system fails POST on "System Failed CPU Test". Switching off and back on again usually resolves the problem, if not the first time then the second.

If the ambient temperature is above approx 18C the system passes POST with no problems.

Nothing else is wrong with the PC.


I've done everything I can think of to try and resolve the problem but nothing seems to make any difference. All seems to point of a dodgy motherboard and a quick search of the net suggests the same - there seem to be a fair number of people having this problem which as been resolved after replacing the mobo.

The only problem there is that the board is more than 12 months old now and although ASUS offer a 3 year warranty, I can't really spare the time for it to be sent back to ASUS for analysys/repairs (Open University work to get done).


Suggestions? Apart from heating the room before turning the PC on. :lol:

[Taxation]

Your computer likes heat?!?!?!?!?
Just keep the radiators on :wink:

[Danno]

i have seen this problem a few times before and allot of the time it has been caused by a faulty power supply, however once or twice its been the motherboard.

when its cold and you switch it on a little power goes through the psu and motherboard and it warms up the components, if you switch it off and on again a little more power goes through it warming it up a little more. after a while the components are warm enough and the system will work.

if you have not tried one already i would recommend a trying a new power supply as there is a good chance that this could solve the problem.

[unkn0wnvariable]

My current PSU is a 600W Enermax so seem unlikely there should be anything wrong with it. Unfortunately the only spare I have is a cheapo thing that came with a case so I'm not sure if it'll be upto the job of powering the PC. I keep meaning to try it but I have to put it in and then leave the PC overnight or during the day in the hope that the ambient temerature (and the internal temperature of the machine) will drop to below 18C, which at the moment it isn't. :lol:

[beardyman]

sounds like a dodgy connection or dry joint on the board - like it needs the (minute) expansion of a track or component to make a decent contact.

[unkn0wnvariable]

Actually thinking about it, I have the BIOS set so that the PC doesn't auto power on when I connect the mains. I thought it might be the PSU a while ago so I tried leaving the system for a bit after connecting the mains before I switched on and it didn't make any difference. Therefore it does seem unlikely to be a power supply issue.

I similarly suspect a dry joint or fractured pcb track on the mobo somewhere, hence the working when warm but not when cold.

Anyway I've swapped my PSU for a 400w Macron item for now, I'll see what happens next time it's cold. Haven't used this PSU for so long I'd forgotten I had it, it didn't half stink when I turned the system on!! Unfortunately the room temp was about 18C before I got it all back together... so it hasn't prooved anything yet. :lol:



Your computer likes heat?!?!?!?!?
Just keep the radiators on :wink:
And there's me thinking they prefered to run cool...!! :lol: :lol:

[Chewie]

You say the board is a year old or so, but when did you fit the 600w enermax?

Just which enermax is it?

Its possible your PSU is a bad one, just because its a brand name doesn't mean a bad one hasn't slipped through QC, but not likely. Not as if its a Q-Tec :wink:

If its only occured since you installed the Enermax
1stly, recheck all your connections in case one is loose.
if it boots fine, problem solved, if not go on.

2ndly barebone it, justhave your mobo, psu, ram vga and 1 drive connected, see if she starts first time then. EVEN with the windows open to cool the room a bit to try and get the fault as normal.
If it boots fine after this add more of your equipment 1 item ata time, untill it fails to boot. This will tell you if you osu can't deliver all the power it needs to.

If this fails, then try a new PSU/Mobo, whichever is easier.

1 last thing.

Google for COld boot problems, someone may just have the same as you ;)

[unkn0wnvariable]

I've nearly always had the Enermax in it, it's a EG701AX-VE(W)SFMA. I installed it about 2 weeks after building the PC because the original Nexus PSU disagreed with the Asus board and my Seagate harddrives. Apparently this is a know problem with Nexus PSU's. :roll:

Also it's a 600W one, so I'd be impressed if the PC needed more juice than it can supply. :lol:

It's hard to say when the problem started since it only does it when the ambient temperature is cold, so obviously it first showed itself at the start of the winter. :lol:

I have already searched Google for the problem, there's a fair few people having the same problem with similar Asus boards, many have resolved it by getting the board replaced. Mine is over a year old now though and has since been discontinued so I suspect it would have to go back to Asus for repairs, even if not there isn't a suitable replacement for it available at the moment.


Anyway it's not the PSU, it did it again this morning with the Macron PSU installed but it was slightly more helpful today. Rather than the usual blank screen it powered up with the GPU BIOS at the top and an error saying BIOS Corrupt, then started cycling round the DVD and Floppy drives looking for a rescue file. Switch off and back on and it's fine as usual.

So I've taken the BIOS chip out and cleaned it, I got a suprising amount of muck off the pins and the socket!! They were all somewhat dull looking so after a quick solvent clean they're all shiney again now. See what happens. :)

[unkn0wnvariable]

Update.... lol

That didn't seem to solve the problem as it's happened once or twice again since but it's been too warm since to do any more troubleshooting.

However, now I seem to have lost the use of the centre RAM slot on the mobo, whenever I put RAM in it I get errors in the same address range regardless of which of my RAM slices I put in there. :roll: :lol:

So I'm assuming it's the mobo that's gradually dying... Is there any way I can test it's the mobo and not the CPU? Bearing in mind that I don't have any parts to swap them with. :lol: