Disconnect power to start pc....uh?

[Ithinkso]

Well folks even if you can't solve this one at least now you'll be able to say youv'e heard of it.
I've had this pc - asus A7N8XE-deluxe , AMD 2100 for a couple of years now and apart from the original Maxtor SATA drive failing a couple of months back (my son was downloading pics from his camera -freeze- restarted and that terminal metallic tap tap tap said it was dead) I've had no problems except....About eight months ago it started to take several tries on the power button before it would start, within a week it failed to start at all.I decided to check all connections and noticed that when the power lead was pulled the fan spun up briefly....to cut a long story short to power up my pc you need to pull the power lead off completely press the start button and reconnect the power lead in that order.
Everyone who's seen it just scratches thier heads.PSU seems favourite but a techie at work loaned me his test unit and it passed, all caps on the board appear to be okay,new power lead,all connections checked ,still the same.For some reason the power has to be disconnected for switching to occur..?

 

[Lexluethar]

You could use a key to short the connection between the two power pins - creating the connection that would start your PC as well

Are you positive you have all of the power connectors right from the case to the MB? Kinda sounds like maybe some of the wires are switched.

 

[advancedgeek]

does this mean unplugging and plugging back in the power cable to turn your pc on? if that's the case the power switch might be faulty on your case and either way you should get a home made switch so it will short the pins that are connected to the power to turn on the pc and it will also be convenient as you don't have to reach all the way down to your computer to turn it on. I seen those like http://www.logicsupply.com/images/photos/cables/PowerSwitch_big.jpg where you just press that button to turn it on.

 

[Ithinkso]

Thanks for your response guys.
The power lead I mentioned is the lead from the mains outlet on the wall to the pc,sorry if you thought I meant the cable from the power switch to the motherboard.
Switching the power off at the wall socket, pressing the power switch on the case then quickly switching on again at the socket has the same effect...the pc starts.
I did try shorting the two pins, but without any luck I'm afraid......Thanks again for taking the time to reply.

 

[NetCablesPlus]

I still think that you simply have a bad on/off switch on your box. I have seen quite a few of them wear out and they all get little quirks about how you have to click them just right, with the moon in the proper phase, and the birds singing, etc., to get the machine to boot. This sounds too much like that to not be the switch, itself.

 

[S0ULphIRE]

I've had this problem MANY times, all with older computers tho. Haven't had one for a while now (I've kinda moved on to newer pcs ) but as far as I recall it was an automatic protection issue (as in mebbe short somewhere, too high temp, etc.).
And to get the pc to start again, I had to completely reset it's 'memory'.

I'll see if I cn rustle up one of the old pc's again and duplicate the fault, see if I cn narrow it down and find the problem

I still think that you simply have a bad on/off switch on your box.

how could it still be that if he totally bypassed the switch and shorted the connection? oy

 

[Nitestick]

trying another psu is the best means to rule out the psu as the culprit. don't trust a psu tester on this one.

 

[Ithinkso]

Once again thanks for your suggestions folks.
Yes it isn't cutting edge but it's perfectly adequate for my needs...The Zalman cooling kit would seem to rule out the overheating route (remember this is starting from cold) and a look at the figures confirms this.
I'm inclined to go with Nitestick's suspicion of the psu , would a pocket sized tester reproduce the startup load on the psu or simply confirm there's output ?.....it may take me a while to get my hands on a decent psu but as soon as I have I'll get back.
All the best..........

 

[Nitestick]

you're correct in doubting whether a PSU tester can replicate actual usage conditions. if i had to guess i'd say there's probably something wrong with the PWR_GOOD signal.