Super User is a question and answer site for computer enthusiasts and power users. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

I have a computer with 2 SATA HDD and an abit KN9 motherboard.

If I set the CMOS jumper to the clear position, and if I disconnect one or both of the HDD, then my computer powers on --- at least it doesn't power off, I get no beeps or BIOS screen etc, but the lights and fans stay on.

If both of the HDD are connected, the position of the CMOS jumper doesn't matter, the computer lights and fans power off within 1-2 seconds.

If just one HDD is connected, then when the jumper is in the clear position: no boot, but lights and fans remain on. With the same single drive connected, putting the jumper to the normal position kills the power on within 1-2 seconds. This situation is true also when it is the other HDD solely connected.

If no HDD are connected, we have lights and fans indefinitely (but no BIOS or OS) when jumper is set to clear position, and power off within 1-2 seconds when CMOS jumper is in normal position.

With no HDDs and no RAM I get beeeeep beeeeep beeeeep (my BIOS is Phoenix AwardBIOS) and then shutdown the first time, but sometimes I get no beeps, other times I get an endless number of long beeps, so it seems somewhat variable.

Any ideas what the issue could be? I've tried the obvious things like loose cables, re-seating the RAM and CPU. I'm thinking perhaps the PSU needs replacing?

EDIT: So I now think the CMOS jumper setting was just coincidental; it seems that the behaviour alternates between power ons. The first time I might get fans and lights (no POST) indefinitely, the next time I will get fans and lights and then dead after 1-2 seconds and so on....

I also checked for bulging capacitors --- there aren't any.

share|improve this question
Replace the motherboard. You shouldn't need to clear the CMOS to boot the pc. – Ramhound Jun 10 '13 at 21:23
I now think the CMOS being on clear/normal was just a coincidence, as the above behaviour seems to occur with repeat switch on/offs regardless of changing the CMOS; that is to say, one time it the fans and lights will die with in 1-2 seconds, the next power on the fans and lights will go indefinitely, the next 1-2 seconds and so on.... – fpghost Jun 10 '13 at 22:51
It couldn't be the PSU? – fpghost Jun 10 '13 at 22:51
It could be but it's unlikely based on the behavior you describe. – Ramhound Jun 11 '13 at 1:32
Isn't the fact that it gets to POST stage and starts beeping when the RAM is removed kind of interesting? – fpghost Jun 11 '13 at 8:33

It's dead Jim.

Get a new motherboard.
That behavior makes no sense what so ever unless something is seriously fried.

Did you have a thunderstorm lately ? Lightning strikes in the vicinity can do this sort of thing.
(It doesn't have to be a direct hit on your house. Anywhere in the same local power-grid.)
It can be several days later before the computer starts acting weirdly. Computers are more sensitive to this sort of thing than e.g. tv's or radios.

If it is lightning damage consider replacing the entire computer.
Chances are very big that after replacing one part some weeks later the next one starts misbehaving.

share|improve this answer
No lightening I am aware of, but who knows. Does the fact that the computer at least will POST/beep when memory and HDD are removed not tell me something? Although the amount of beeps before it again dies seems somewhat random. – fpghost Jun 10 '13 at 22:49
@fpghost As long as most of the CPU works you will get the POST beep. The fact that the number of beeps is varying is what worries me. That usually means that either the on-board voltage regulators are damaged or the motherbaord chipset. Either of these would also affect SATA. It could be just a flaky power-supply (see wjeemba's answer below as well), but then the behavior is usually more consistent. – Tonny Jun 11 '13 at 10:21

It could possibly be a fried power supply- if you have a spare one with enough wattage capacity sitting around, you can try testing that way. Though I have my doubts because the computer would likely not power on at all if it were bad.

Otherwise, I am more inclined to think that it is indeed the motherboard that needs to go to the great recycling center in the sky. A new low-end motherboard in the US costs $40-50 dollars, but if you have the extra $30, something in the $70-$80 range will be more robust and of better quality overall. Just make sure you get one that will work with your current CPU/RAM, or think about putting the money towards a new computer.

share|improve this answer
So, the original asker said in the below post that my suggestion was the one that fixed the problem. :D It doesn't look like they know about how to select it as the answer to the question- is there a moderator process for going about that? – wajeemba Sep 6 '13 at 21:34

I had the local computer shop swap the PSU, and it turns out it was this at fault!

share|improve this answer

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .