Is my PC Fried? plz help!

i finally got off my *** to install a new psu into a pc that had been totally unplugged for about a month and i run into these problems:

first, a few seconds after pressing the power button, i hear this long, 'crackling' beep (which doesn't appear to be listed here: http://www.bioscentral.com/beepcodes/awardbeep.htm ) -which i first attributed to having no ide drives connected- that stops when i hit the del key to enter bios setup. problem is, the pc never enters setup; no matter what keys i hit, i'm stuck on the initial screen reporting no ide drives found.

the no ide drives found message is expected: i disconnected all drives previous to starting up the pc for the first time.

here's the situation; this pc had been totally unplugged for more than a month following an unsuccessful transplant of a ****ty backup psu -during that occasion, i got one long, unending beep, with nothing showing up on the screen at all. i tried this about three times before tossing the cheap piece of crap aside.

the backup was necessary because my previous psu was quickly expiring due to bad caps; it had in fact corrupted an entire hdd before i finally turned the thing off for good. sadly, my backup was one of those cheapo, no name, freebie psus that come with some cheapo cases.


so far, with my brand new psu, all i've really tried to do is clear the cmos and try again. the third and second to last times i tried this however, i couldn't get anything to show up on the screen and the pc made no beeps at all. the final time, i was finally able to get the single, long, crackling beep back with frozen screen (and no visible glitches other than apparently not responding to my keypresses; i've tried using two different keyboards (neither are usb), both of which light up and one of which i'm using right now, so it can't be the keyboard...).

what do you think, all? is my rig salvagable? could it be a simple problem like requiring a new mobo battery? is the cpu fried or is it the mobo? is my ram ok (sadly, i neglected to pull out both sticks to test the mobo)?

specs:
psu: antec EA380 (no voltage select switch, could this a problem? someone mentioned my problems could be due to an overvolt... also, this is a 24 pin psu currently residing in a 20 pin mobo, with the extra four pins pushed aside; are those wires (yellow, orange, red and black) supposed to plug into the mobo rather than the four pin, double black and double yellow (12v?) plug?)
mobo: abit nf7s v2 w custom bios (i think it's merlin's d25 cpc off)
cpu: amd xpm 2500
ram: 2x 512mb pc3000 kingston hyper x
gpu: 64mb radeon 9500

HELP!!!

This would indicate the CPU heatsync is not on properly, resulting in a huge overheat. Possible broken CPU. Try taking the CPU out completely and see if you get the same error.

A new battery will not resolve this issue. The battery only recalls the BIOS settings, (E.G. date/time, how many drives/size of drives, boot order, etc.) A new battery will not help with the issue described.

If your PSU does not have a voltage switch, it means that you bought one with an active PF* (I can't recall the three letters, PFC maybe?) It's not a problem. When you plug it into the wall, it looks for 110 or 220 volts and acts accordingly.

When using a 24 pin PSU into a 20 pin MB, the four extra pins will hang out into the air. This is correct. The connector will still only fit one way so you can't let the wrong four pins hang out. Unless you used a hammer to install it.

Look at the caps on the MB carefully. Are any of those bad? Also, I've seen a lot of eMachines where the PSU took out the south bridge. Look for bubbling on the south bridge.

Put the RAM into another machine and run memtest86+.

Good luck.

well, i guess this is as good a time as any to ask this:

spurred on by a post from these very forums, i tried to make my previous application of arctic silver as thin as humanly possible (i recall the message stated it should be 'almost transparent'); it was so thin that i could only see how evenly i was applying it by the difference in gloss when the cpu was tilted.

surprisingly, this netted a significant increase in temps that remained quite high well past the burn in time.

so my question is this: are there any HIGH RES pics demonstrating how much of this stuff should be applied and what a cpu die should look like before and after applying arctic silver/htm?

i've tried the grain of rice rule and discovered that only about 20% stays on the die and the rest remains on the application tool. is this right?


@ jayfella: do you mean try starting the pc with no cpu installed? i haven't heard of this before -and i remember seeing the cpu specs listed in the boot screen, but i'll try that later tonight; i REALLY hope this is just a simple issue of there not being enough htm.

the reason i suspected the battery is because i got a cmos checksum error each time i started up, despite resetting the cmos.

The correct application of thermal paste is a drop in the middle of the CPU large enough that when you place the heat sink on it it spreads to cover the entire CPU. Some run-off is fine.

The only applicator that I use is the tube itself. I just squeeze what I need directly onto the CPU.

I used to spread it on thin but I started having heat issues with my CPUs. Mind you that I never made it transparent. That's too thin even for me in the day. I covered the CPU with just enough for it NOT to be transparent. Then I spread just a minor amount on the heat sink. Again, this was too little.

When my HTPC 2.8 P4 with HT could not fold on both "cores", I found that it was getting too hot and crashing. I took my heat sink off, cleaned it, then put a large dalup in the center. I put on the heat sink and it's been folding fine ever since.

I have since done this to all of my other machine and everything is cooler than before.

If your computer starts up then your CPU is fine. Its just overheatig bigtime - beeping like that is a surefire way of making sure you know something is not right.

If you are using the stock cooler - the grain of rice idea is a generalised one at best. No 2 computers are the same, nor CPU, nor heatsync. Yours might be warped ever so slightly. This is natural. In your case, more would be better. I would put maybe 3 grains or rice if 1 grain only covered 20%. As long as your temperatures are reasonable, apart from the odd degree or maybe 3 - your not going to get much better, unless you have done something drastically wrong like forgot to put a push-pin in or something like that.

after applying some more arctic silver, i noticed something new; the mobo only makes the crackling beeping sound if a keyboard is plugged in (so far only tested with the ps/2 slot)! this has to mean something, right?

so far, i've only tested with the ps/2 socket. i'd use a usb keyboard but...


...i also made a huge mistake; i accidentally knocked off a little rectangular metallic piece on the front of the mobo, next to the cpu, while trying to unscrew the hsf backplate (so i could get a good look at the back).

the bit looks like a little, silver extruded square with a large tan area in the middle, and is located around here:
i still have the piece, fortunately. unfortunately, i also scratched off the name, which i gather from looking at other, similar bits ought to be 'CT(some number).' so i have some new follow up questions; can this little bit be safely soldered back onto the board? does the orientation matter for this part (if so, i'm screwed since it has no markings aside from the tan part on the body)? also, ps/2 error aside, can i safely use this mobo without soldering this piece back on?

finally, what might have caused the original problem if it wasn't due to overheating? oddly enough, on the first startup with the new application of as5, the crackling beep was significantly shorter, less than a second long, but with subsequent startups the beep obtained its original length of about 5 seconds. as far as i can tell, it ONLY ever beeps if a keyboard is plugged in to the keyboard ps/2 slot.


someone else informed that i could have something causing an overvolt somewhere or that i might have fried the southbridge. -the latter observation is particularly meaningful because i used to use the zalman northbridge cooler i've got superglued to the southbridge as a kind of handle for the mobo. i've been doing that for years without incident. someone please tell me that couldn't have caused this.

help?

hang on. Does it sound like you have a key stuck on your keyboard? It might not like that keyboard for some reason.

As for the bit of motherboard - well... They wouldnt put it on there for no reason huh. But having said that - it might be for something your not using - like the 4th PCI slot or something - or it could also be something seriously important - so i think its hit and miss tbh - sorry - i know that doesnt help much.

When you say "superglued" - you mean you took the old cooler off and superglued the new one on - or you just glued it on top of the existing cooler?

What you knocked off sounds like a resister. The board can not be considered stable with a component knocked off. Also, with the name scratched off, were any traces also scratched?

At this point, you could try to get some one to solder that bit back on. I don't think it matters which direction. However, I would consider the board toast and move on.

knocking things off a mobo cant be good lol, if it cant be fixed mate, id agree ^^ get a new mobo to be safe

ive broke resistors of a mobo a few times with clumsy screwdriver slips
when i soldered them back on it seemed to have worked

thanks for your replies, all!

@jayfella: sadly, i've gotten the same result with two different keyboards. my next step, after soldering the bit back on, is to try using a usb keyboard and see if that nets a better result. if not... well... damnit i just don't want to spend upwards of $100 getting a brand new nf7sv2! especially not when i still don't know what happened to this one.


the nf7sv2 doesn't come with a southbridge cooler (if it does, it's been totally replaced, and the southbridge area cleaned off); the nb cooler is seperate and superglued directly onto the southbridge; i think i applied some as5 to the middle and put superglue around the edges.

either way, i've been using the superglued sb cooler as a handle to mount the mobo; can someone please verify that this is safe (because otherwise it's rather difficult to lower a full size mobo into a medium sized tower, especially without removing all the optical drives and such)?

Well its not advised. You should just put the back plugs on the mobo (speaker connection, usb, etc..) in first, then lower the other side down. Its quite simple once you get the plugs into the backplate. I wouldnt put any force on a cooler at all. Your just going to make things worse, with ZERO chance of making it better. I wouldnt see why you would need a cooler on the SB anyhow ??