Solved: My PC decides when to startup...not me!

[jimlarkey]

Looking for ideas to resolve an outstanding issue I have with my PC; it decides when when to start, not me. I typically press the soft-start button on the front panel, and it may be minutes or days before the PC starts. Nothing, no fan, no lights, no HD led, no nothing for as few as minutes, and as much as days. Yeah, the hard-power switch is on, and the unit is plugged in.

Some observations:

* ASUS A8NSLI-B motherboard, WinXP-Home ed.
* BIOS... Phoenix - AwardBIOS v6.00PG
* Power Supply Unit checks out OK with ULTRA PSU checker; MB connector OK, 4-pin Molex OK, 6-pin PCI-E OK, P4+12v OK.
* CPU temp 100 deg F, MB temp 105 deg F
* PC will almost always (~80%) start up from hibernate OK, but fails to predictably startup from shut down.
* Last time to startup, I pressed and HELD the soft-powerup button, and toggled the hard power switch off and then on...fired up immediately.
* BIOS was set to restore PC on press of space key....space key failed to activate, except at random times.
* Disabling Hibernate does not improve startup prob.
* Takes longer to fire up approaching a full moon, then a new moon

Other suggestion was to try WinXP's repair option on the orig WinXP CD, but concerned that "repair" might upset too many other setup options config'd over the years.

Yeah, this is driving me nuts. It's not the soft-powerup switch, because I can almost everytime startup from hibernation OK. I'm thinking it's some low-level function like BIOS. Also wonder if it's a flaw in the motherboard? What do you think? Looking for ideas, experience.

Thanks, and Cheers!
Jim

 

[6rgwa]

It is not probably your case, but have you tried disconnecting case(I mean cables) and turning on computer without it. Have you tried different PSU?

 

[Aussie_Dundee]

It is not probably your case, but have you tried disconnecting case(I mean cables) and turning on computer without it. Have you tried different PSU?

Sounds like good advice. I would recommend bypassing the case and try jump starting the comp directly from the mobo.

You do this by following the wires from the mobo to your case where the soft power switch is. The short the wires with each other instead of letting the switch do it for you.

 

[jimlarkey]

BAD PSU.....
The PSU is 1yr old, and I've tested it with an ULTRA 7-in-1 PSU tester...tests OK...how would it be power supply?

BAD SOFT POWER-ON SWITCH.....
If you're thinking that the soft power-on switch is bad, as I originally did, then why does the space key on the keyboard behave the same way as the soft power-on switch....sometimes works, sometimes doesn't?

Thanks for your suggestions and interest. Let me know why you think the PSU and Switch would be bad.

Cheers,
Jim

 

[Aussie_Dundee]

Wear and tear are most probably the reasons for both problems. Altough keyboard and power have nothing to do with each other whatsoever. Dont even try and find a paralel betweem them.

try my suggestin to rule out the possible defect button. then try prying the space button off the keybaord and cleaning underneath it.

 

[midnight51]

Bad motherboard my idea

 

[jimlarkey]

Passed the paperclip test as described at:
http://forums.techguy.org/hardware/583336-solved-computer-dead-there-simple.html

Failed to start when shorting across terminal of soft power-on switch.
Failed to start when shorting across pwr terminal on mobo.

My problem is similar to the poster in the above link, except my PC WILL fire up, randomly. Sometimes minutes, hours, or days. Doubt that it is thermal. Doubt that it is PSU. Doubt that it is contacts.

Replacing the mobo is like building a new PC. Easier solution would be to never shut down the PC......the heck with a green PC....it draws only 105 watts at idle -> $7 / mo.

Anyone have any thoughts on the OS (WinXP-Home ed)? WHen I disabled the hibernate feature, I was more likely able to fire up the PC, but I can't quantify it. Is it possible that there is something wrong in the OS?

Still looking for ideas.

Cheers,
Jim

 

[Aussie_Dundee]

Nothing software wise can affect you powering up since you have to get the juice going before anythign in the computer can start processing software affected configurations. Its impossible for the mobo or anything else to get info off the hdd before you pressed the power button... So no. it cant be OS imo. Possibly a BIOS problem but i highly doubt that as well.

i am thinking its the mobo or psu tbh... Are you sure you did the mobo shortening properly?
If the PSU works which obviosuly it does. (even if only sometimes) it still SHOULD have started. If you get the same symptoms from shortening the mobo directly bypassing the case button then its time for a new mobo tbh. they arent expensive. 40 punds will get you a model similar of what i am guessing you have.

 

[jimlarkey]

I'm certain the test was done correctly. I'm a registered elec engr so I do know how to short contacts. Believe me.

Soooooo......what do you suppose contributes to the variable time factor for startup?

If mobo was bad, seems like it would be bad or good......not bad for 5 minutes, bad for 7 hours, or bad for 2 days.

I will run PC continuous rather than gamble on a new mobo. Not convinced it is mobo yet.

Cheers, and thanks once again. Appreciate the quick responses and interest.
Jim

 

[Aussie_Dundee]

Nice to know you got that as a background. Why not try to measure current levels when pressing the soft power button to see if anything registers. try error searching as you know best. It definately sounds like a hardware fault. I have some electrical background. but not even close to yours. But i still would get out a voltameter and make sure everything is working...

to 99% i judge it isnt a software problem thats about all the help i think i can give atm.

keep us updated.

 

[midnight51]

I have heard if there is a faulty solder connection somewhere... maybe it isn't complete or it got chipped away so it doesnt complete the circuit that it could lead to problems like this. Maybe it is sending current through the board and it can't complete and as it sends current it gradually expands the solder to make the connection? I could be completely out of my element but it sounds good in my head

 

[Aussie_Dundee]

Its definately a possibility but a far reached one though. The small amount of volume something like that would increase even after being very overheated would mean that its more likely never to complete the circuit rather than sometimes. the variables needed to make this occur naturally have such a low percentil that its nearly ruled out without checking.
But yes,. it is a possibility... Nice idea

 

[badders]

Have you got any Bad Caps on the board?

 

[jimlarkey]

Sorry, but it seems that solder connections, and bad capacitors, and other integrated mobo elements are like looking for a needle in a hay stack. How does one go about checking all the integrated elements on a mobo?

Thanks,
Jim

 

[PuppyLinux]

Bad caps are pretty easy to spot; they'll be leaking electrolytic goo onto the motherboard (usually from the bottom of the cap), or the top of the cap will be split or bulging, obviously deformed...

You can generally shotgun the caps if you have quality replacements available, with a good soldering station...

Check this website for the gruesome details about the industrial espionage that began the saga of leaky / bulging caps, and some good photos to show you what to look for:

http://www.badcaps.net

Hope that information helps you...