The most annoying computer problem ever

drewp29

Distinguished
Jan 7, 2009
66
0
18,630
So a buddy of mine asked me to troubleshoot his computer and see why it wasn't turning on. My first thoughts were - power supply, the first thing to check is always the power supply. I hooked the machine up and pushed the power button. Nothing happened. I got one of my old power supplies and hooked it up. Pushed the power button, the fans kicked over for a second and then nothing. I jumped the PS to make sure it would turn on, and it did. Thinking maybe my PS was still bad (low current/voltage etc.) I got my BRAND NEW OCZ 1kW PS out of its box, hooked it up and the same thing happened. So my next thought was motherboard. Just to reinforce this thought I tried both sticks of memory one at a time, in all four slots. Nothing.

So I went online and found a motherboard. This was a Socket AM2 CPU, so my options were limited. I found a cheap used one from Ascendtech. The motherboard arrived, I built the system, powered it on and it did the same exact thing. So I thought, well, so much for that idea. I got a CPU from my brother, and two more sticks of DDR2 memory from my dad and tried those in the new Foxconn (used but verified working motherboard). Same situation, except now the board would power on for 5 seconds and then turn off. I assumed the new motherboard was bad since it is rare to have two bad CPUs and four bad sticks of memory. So I sent the motherboard back to Ascendtech. They tested it and said it was good. I had them send it back and built the system again. Same result (definition of insanity anyone?). So I went to Newegg and found an ASRock motherboard that is backwards compatible with the AM2 socket, as well as DDR2 memory, thinking that if all else failed we could always buy an AM3+ CPU and some DDR3 memory and have an entirely new system.

I built the system and now it is behaving strangely. One - no video output from the onboard video or a card. Two - the system powers on when the switch on the power supply is turned on, regardless of whether the case power button is connected to the MB. Three - The system does not beep any error codes if I do not have memory in, and yes, I hooked up the piezo speaker. Four - the system is obviously not reaching POST. I tried both CPUs again, and all four sticks of memory, two graphics cards, three power supplies, and the only theory I have is the case is being haunted by a spirit and once the hardware comes near the case it is forever cursed!

Please help, I am starting to think I don't know anything about computers and should swear off technology completely because it obviously is too complex for my feeble mind to comprehend.

BTW, I have built 24 PCs to date, and diagnosed and fixed more than 50 at least. This isn't my first time at the dance, but it is the first time I have apparently spilled punch all over the front of my collared shirt, and pissed myself while everyone was laughing at me . . .
 
I believe you have a case that's been improperly labeled - meaning, the reset/power/HDD and the other connectors that usually come as a bundle to plug into the mobo have been mislabeled. Another possibility would be if the power/reset buttons are malfunctioning/shorted out.
Suggestion: before going nuts over this, try to build the system outside the case (what we know as breadboarding). The power switch can be "replicated" by shorting the pins on the mobo, in order to turn the system on. Try that first (make sure you don't short any other things on the mobo while turning it on) and get back to us after that.
Good luck!
 

drewp29

Distinguished
Jan 7, 2009
66
0
18,630
Thanks for the reply. I bread boarded the system thinking perhaps the case was shorting out the motherboard and got the same result. The odd thing is no matter whether the case power button is hooked up or not, the system turns on when the power supply switch is turned on. The power button will turn the system off, and the system stays off provided you don't switch the power supply off and back on again. The system will turn on and off with the power button, as well as by shorting the header pins with a screwdriver.

I am still not getting any video output, and its not reaching POST, with either CPU+Memory+Graphics Card combo.
 
If the power button does the right thing (turn the system on and off) I would not be concerned too much about it.
Now for the not reaching POST thing, that is something weird. You sure the graphic card is hooked up and all the power cable(s) are also hooked to it? I suppose so, but it doesn't hurt to check it. Monitor is OK, plugged in and everything?
 

drewp29

Distinguished
Jan 7, 2009
66
0
18,630
Yeah, I hooked up the monitor to the onboard video as well. The card I am using currently for testing is an HD4350, passively cooled, and does not require a PCI-E power connector. My only other thought is that both CPUs are bad, but that seems like a stretch. The CPU I got from my brother is an X2 4000+ dual core, and the previous CPU was an X2 3800+ dual core. The 4000+ was working when my brother pulled it from his system for an upgrade. Monitor is working, I use it for my system downstairs.
 
It's common for the motherboard to receive power when you turn on the Power Supply and spin the fans slightly, but NOT actually initiate the POST/Boot procedure.

As for NOT reaching POST, I'm not sure if you can tell with no video output.

Are you certain the MONITOR is working?

It's hard to give much advice aside from swapping parts except to use the MINIMUM configuration required:

- motherboard
- one stick of RAM (see motherboard for one-stick location)
- CPU (proper heatsink component)
- CPU heatsink + fan
- Power Supply
- monitor

Keep this basic combination and swap one part at a time, including the monitor. If every single part except the motherboard swaps then your motherboard is bad. "Luckily", you appear to have another motherboard to use in troubleshooting.
 
UPDATE:
- okay your monitor appears to work

*It seems your troubleshooting is fine, however since you get no CPU BEEPS it seems unlikely that that is the issue.

Sorry, but aside from swapping parts it's hard to provide more information and it sounds like you have that covered.
 

drewp29

Distinguished
Jan 7, 2009
66
0
18,630
See that's the thing, I have swapped everything at this point. I do not have another working AM2 system to put the CPU into to verify that one or the other is actually working. The only system I can verify RAM with is my main system hooked to my entertainment system, so I might pull the side off it tomorrow just to verify at least one stick of RAM out of the four is good.

The list of components you listed are what is currently hooked up. I have swapped RAM into the available slots one at a time and with different sticks. The most confusing aspect at this point (other than the POST issue) is the initiation of power when the power supply switch is turned on. I do not understand how the motherboard is being prompted to turn on without the power button being pushed, or why the power supply would turn on by itself. That is the weirdest thing about the issue.

Thanks for helping me troubleshoot, any other ideas outside the ordinary test procedures? I think I have covered the basics.
 
You said you have onboard video. Try to remove the add-on graphic card, plug the monitor in the onboard, reset the BIOS by removing it's battery (btw, is the battery still OK?) and replacing it after a few minutes, and give it a try.
 

drewp29

Distinguished
Jan 7, 2009
66
0
18,630
I set the machine up initially using the on board video since the default monitor output is most likely the on board and not the pci-e. I only put the graphics card in to see if perhaps the on board graphics was not outputting a signal. At this point I am going to pull the memory from my main computer and test at least one stick of the memory I have been using for this build in my main computer. After verifying that at least one stick is good my only other conclusion is that BOTH CPUs are indeed bad. I do not have another AM2 system to try a CPU in. All my other systems in the computer graveyard are Intel based systems.
 
I re-read your first post. If I understood correctly, you first got the CPUs, built the system with first mobo, then tried both CPUs with the second one (after returning back and forth the first mobo).
If that's correct, then I would go on a limb here and explain this: the first mobo did something tho the CPUs, and that issue is now happening on the second system, with the new mobo (since both CPUs are now damaged). I know this is a very long shot in the dark, but logically I do not see any other reason for this weird thing to happen.
I know this is not helping the current situation any, but at least offers an explanation (other than the "ghost in the machine"). Sorry. If, at any point in time, you get your hands on a new CPU, you can test this hypothesis on this mobo (if you'll still have it by then).
Looks like we have covered pretty much all the bases with this issue. Good luck, mate. If you come up with something please let us know.
 

drewp29

Distinguished
Jan 7, 2009
66
0
18,630
That's about the only explanation I have for the issue as well. The second CPU was not placed in the first motherboard, only in the second and third motherboard, but your logic still may prove true. I am still going to test the memory just to make certain it isn't a problem. I'll probably buy a 955BE or another cheaper AM3/AM3+ chip and some DDR3 memory since it is cheap right now, and try to build a new system instead of harvesting old parts. I was trying to save money for my friend since money is tight right now, but if I have to I'll spot him on this. If this were my machine I already would have sledge-hammered the parts and bought new . . .
 

But your second mobo was the one you got back from testing, right? Was the same mobo you sent them and they sent back to you because it was "OK".

Anyhow, I see your reasoning here with this not being your system and all. I would advise him to save some money, get some parts and when all available, put them together.