Computer turns on with the reset button

[Dan886]

Hi,

I've had this computer since 2010 working fine. Since some weeks ago, sometimes (specially leaving it fully turned off for the whole evening, then coming back to it the next day) I can't turn it on using the power button, it will only turn on with the reset button. I haven't checked the front panel connectors but I'm certain I haven't tampered with them since the installation was done. I can hear the power button clicking all the way in so I'm not sure if it's the button itself. What could be going bad?

Thanks for any help.

 

[Paperdoc]

Maybe the power button on the front is wearing out and fails to make proper contact when you push it. You could replace it. Even if you can;t get an exact replacement part, it is only a momentary-contact pushbutton type similar to a doorbell pushbutton. So any switch of that type could replace it.

Alternatively you could open your case, pull off the leads from the Power Switch at the mobo Front Panel header, and transfer to those header pins the leads from the front panel Reset button. Then the Reset button would become your Power button permanently, and you'd have no Reset button to use.

 

[Dan886]

Well I think I fixed it.

I disassembled the front panel and the switch, then tested the switch itself with a tester after cleaning it and reassmbling it and it gave continuity all times, so it had to be good. then I reconnected it after loosening the power switch conenctor as it was tightened by whoever did the install, just to make sure it wasn't make it lose contact with the motherboard. So far so good. Will test again tomorrow after leaving it turned off for the evening. I asked because I thought something on the power supply could be going bad since it was weird that the RESET SW was turning on the PC (as it's installed on the reset pins not in the power ones!). Thank you Paperdoc

 

[Paperdoc]

Actually, I've never tested what happens when you go to the computer in the "Off" state and push the Reset button. Maybe it "Resets" then, too, but of course that would look just like a "Turn On".

Glad you were able to clean the switch and its wiring contacts and get it working again. Good solution! And thanks for Best solution here.

 

[Dan886]

Whoops, no, it's not solved yet! I tried turning it on after the evening and it wasn't turning on using the front power button, so I wondered what would happen if i turned the power switch off, then on again (i didn't use the one in the power supply, instead i used the one in the wall plugs because it was easier to reach) and THEN it turned on with the front power button. And it worked, so it's definitely not the power button. ???????

 

[Paperdoc]

This could be any of several things, and take a while to figure out. Maybe an odd setting in BIOS Setup. Maybe some setting in Windows about how it shuts down or uses "sleep mode".

I had a similar situation some years ago, but I don't want to alarm you. I'm not saying this is your problem - just a possibility to watch for. My machine started doing something like that. Sometimes it would not start from the front pushbutton, I could always get it to start by turning off the switch on the back of the PSU (or unplugging) and waiting for 10 minutes, then restoring power. It seemed to me that somehow when it shut down, the small part of backgound software running (to wait for a start-up signal) was locked in a loop and could not function. Could not figure that out. Later it began spontaneously shutting down during normal use, and I'm not a heavy user / gamer. I tried cleaning out dust in case there was a cooling problem, but that made no difference. Later the spontaneous shut-downs and need to re-start became more frequent, so utlimately I pulled out my PSU module and opened it to inspect - it was well past warranty coverage. Inside I found a couple of large capacitors with "bulged tops". If you are not familiar with that situation, there was a period about 2005-2009 where a lot of PSU's (and many mobos) failed after a few years due to faulty capacitors in power supply circuits that developed internal leakage causing overheating and bulging tops.) Replacing the PSU was my permanent solution. But it took nearly a year for the symptoms to progress to shut-downs in 15 minutes of use, so that I could recognize what to look for.

 

[Dan886]

This could be any of several things, and take a while to figure out. Maybe an odd setting in BIOS Setup. Maybe some setting in Windows about how it shuts down or uses "sleep mode".

I had a similar situation some years ago, but I don't want to alarm you. I'm not saying this is your problem - just a possibility to watch for. My machine started doing something like that. Sometimes it would not start from the front pushbutton, I could always get it to start by turning off the switch on the back of the PSU (or unplugging) and waiting for 10 minutes, then restoring power. It seemed to me that somehow when it shut down, the small part of backgound software running (to wait for a start-up signal) was locked in a loop and could not function. Could not figure that out. Later it began spontaneously shutting down during normal use, and I'm not a heavy user / gamer. I tried cleaning out dust in case there was a cooling problem, but that made no difference. Later the spontaneous shut-downs and need to re-start became more frequent, so utlimately I pulled out my PSU module and opened it to inspect - it was well past warranty coverage. Inside I found a couple of large capacitors with "bulged tops". If you are not familiar with that situation, there was a period about 2005-2009 where a lot of PSU's (and many mobos) failed after a few years due to faulty capacitors in power supply circuits that developed internal leakage causing overheating and bulging tops.) Replacing the PSU was my permanent solution. But it took nearly a year for the symptoms to progress to shut-downs in 15 minutes of use, so that I could recognize what to look for.

I see. Well I've never checked this PSU's capacitors but I definitely should check it out later or something. In my BIOS I have the Sleep Mode set to [Auto], not S1 or S3. This issue seems to happen regardless of me running Linux or Windows (i dual boot).

Edit: I've changed the Sleep Mode from Auto to S1 and i'll see what it does now.

 

[Dan886]

I opened the PSU after taking it out. Turns out there are 2 1000nf 16v capacitors that are bulgy. Might try replacing them with capacitors of other PSUs I have where they don't look bulgy.

 

[Paperdoc]

You sound knowledgeable about repairs to electronics, but just let me echo a common caution. Be VERY careful when opening a PSU and working inside. They do have large capacitors that can store a big charge for quite a while, and they do have high voltages in some places. So I suggest a thorough session with an insulated wire to go through the interior and ground every component you find.

 

[Dan886]

Thanks for all your help again. I do have at least some basics in electronics stuff but this is my first serious computer that I've had so I was not quite sure on where to shoot at, whether it was the motherboard or the PSU or the damn button itself.

I turned it on again today and the power button worked, I'm not sure if I fixed it by cleaning the insides of the PSU (it was VERY dusty at least) or not but at least I can be safe to assume that it was the interior of the PSU going bad. (it's an OCZ StealthXStream 500W btw) If I can't find 2 of those capacitors (I have seen one in an old PSU but I would need 2) I'll just buy another one, since the 500W ones seem cheap these days, and my PC draws up to 390W according to the Power Supply Calculator sites.

Also, the fan of the Power Supply wasn't working, so just for testing I've cut the cable of the fan and tried wiring its 12V and ground lines to a molex connector that i had unused and this worked, it's running again. I remember this went bad for more than a year, so this could maybe have contributed to its heavy dust buildup and malfunctioning.

 

[Paperdoc]

The heavy dust inside the PSU probably was not caused by the failed fan. But the failure could have been related to poor cooling. Maybe not, however - overheated and bulged capacitors usually was related to faulty components inside the cap and internal current leakage causing internal heat generation, not to external heat. Interesting, though, that the fan now works. Do you think it was not getting power, or was it jammed by dust?

 

[Dan886]

The heavy dust inside the PSU probably was not caused by the failed fan. But the failure could have been related to poor cooling. Maybe not, however - overheated and bulged capacitors usually was related to faulty components inside the cap and internal current leakage causing internal heat generation, not to external heat. Interesting, though, that the fan now works. Do you think it was not getting power, or was it jammed by dust?

I'm not sure about whether the dust was acting up or not. I did a bunch of things - changed the BIOS Sleep Mode from Auto to S1, changed the qfan setting to Silent (it was fucking noisy before), cleaned up the PSU from all the dust I could without getting shocked or using a can of compressed air (which I didn't have at that particular moment) cleaned up the front panel button with a cleaning solution, and reconnected the POWER SW cable after loosening it a bit. What seemed to have stopped it from failing was the very last step (removing all that dust inside the PSU), because as I said before on the thread right after replugging the POWER SW and trying again it was still failing, but in these 2 days I've tried leaving the PC turned off as much as I could (every evening 8-9 hours, and along the weekend I left it turned off for 4-5 hours) and it's just not failing anymore, the front panel button turns on every single time. There are 2 capacitors that indeed seem bulgy but since I don't have a capacitance tester I'm not going to tamper with that because I can't confirm whether they're bad or not.

 

[Paperdoc]

If it keeps working without fail, it's probably best to leave it at that. By the way, a capacitance tester would not help. I doubt that the "problem" capacitors had any measurable deficiency in their capacitance. Instead it was their internal leakage, possibly measurable by the resistance across the unit. But unless you have specs for what is normal and what is excessive resistance, whatever you measure is useless.