Solved Why won't my PC power on after being off for a few years?

November 29, 2015 at 19:40:24
Specs: Windows 7
A few years ago I built a PC with a fast, water-cooled CPU, lots of ram, and a big hard drive. Basically, I made the best desktop/development computer you could for around $1200 at the time. Some things happened since then, and it's not been used or even turned on in a couple years. Now I want to use it again & when I hooked it up & pushed the power button, it came on for a second, then turned off. It just won't power up anymore.

I haven't been inside a computer since I built this one, but I'm no stranger to them. I'm 37 now & built my first one when I was 13. Anyway, so I've done everything I can think of to troubleshoot this problem: blew out dust, reconnected power cables, disconnected all peripherals (even hard drives, CD, etc.), reseated RAM, removed RAM completely, removed CMOS battery, nothing has worked so far. Obviously, searching online hasn't provided any suggestions that have fixed my problem either.

At one point I was able to get it to boot up & it got to the BIOS & started reading the hard drive & Windows said it had detected a failed startup and it was going to try to fix the problem. It said it might restart itself multiple times during the process. I left it for some time while it claimed to be fixing it. I left the room & when I came back, it was off. I powered it back on & the same thing happened, Windows detected startup problem & was going to fix it. I left the room again, for just a few seconds this time & when I returned, it was off again. I tried again, but from this point forward it won't power on at all.

The hardware is a few years old, but it's all still in very like-new condition. It only had only been used for a few months before being put in storage. I would be extremely surprised if any of the hardware components (PSU, CPU, RAM, Fans, etc.) have failed causing this problem. I don't have any extra components (spare CPU's, PSU's, etc.) to swap out to test, but as I said, they are quite new and it is very unlikely that they have failed already.

What should I do to get this thing running again?


See More: Why wont my PC power on after being off for a few years?

Report •


✔ Best Answer
December 1, 2015 at 23:47:41
I am interested that you still remain convinced that it is not an overheating problem when your IT guy seems to have proved pretty conclusively that it is. I think your best bet is to let him have a go at fixing it, as you suggest.


#1
November 29, 2015 at 19:59:18
First, you probably have oxide on the RAM contacts and sockets so the first thing I would do would be to remove the RAM and clean off the contacts with a soft pencil eraser, then pop them in and out of the sockets 5 or 6 times to burnish the sockets as well. Let us know how that goes.

Since it powered up partially a couple of times the power supply, motherboard, and CPU are probably good so do not concern yourself with them for the time being (at least until we can eliminate other things). Your CMOS battery is probably dead if the machine was left unplugged all this time and should be replaced, but that would not prevent it from booting up.

You have to be a little bit crazy to keep you from going insane.


Report •

#2
November 29, 2015 at 20:03:31
I've tried removing the RAM completely and AFAIK, it should power on without it, and error-out right away. But, it behaves the same, so I don't believe the RAM is the culprit. Just for completeness, I will try your suggestion anyway.

Report •

#3
November 29, 2015 at 22:56:16
I'm wondering whether, after two years sitting idle, the water pump may have failed. Or there could be a lot of air in the system.

Report •

Related Solutions

#4
November 29, 2015 at 23:11:38
The water pump is working. And there is some air in the system (it's a closed system and there has always been some), that I can hear when the pump is running. I know this is an issue, but a separate one.

But I'm at my wits end with it not powering up. I found a way to verify the power supply is definitely working properly. And I've removed the motherboard from the chassis and tried powering up with the same result, so it's not a problem with mounting either. I'm thinking about trying to replace the motherboard or CPU somehow, maybe take it to work & see if they have any there I can test with. I'm pretty sure the problem is either the CPU or motherboard at this point.


Report •

#5
November 30, 2015 at 00:44:17
Enough air in the system would prevent the cooler from cooling the CPU, so it would power on momentarily and then power itself off. Do you have a conventional heatsink that you could fit to test it?

Report •

#6
November 30, 2015 at 00:56:42
I don't have another heatsink. But I'm pretty sure the CPU wouldn't heat itself to the point of shutting off in less than a second even without a heatsink though. The time until it shuts off is not consistent either. Most of the time it won't try to turn on at all, sometimes it'll try for a second or less, a few times it's been on 10-15 seconds. A couple times it lasted minutes. However, whether it's hot or not before trying doesn't appear have an effect on how long it will stay on.

Report •

#7
November 30, 2015 at 01:10:38
Without a heatsink the processor will power off in a very few seconds at the best. If the cooling is working intermittently results will vary.

I've found, when troubleshooting computers, that it pays to test every possibility however remote it may seem. As you've said that you already know that there is air in the system, and that was the case before it sat idle for two years, I think the possibility of insufficient cooling is worth considering.


Report •

#8
November 30, 2015 at 03:02:55
Hello,

If you are good at opening the pc then remove motherboard from the casing and see that there any dust or foreign matters that causing short circuit to the Mother board, then try to hear beeps without RAM which is the first thing of knowing that your mother board is good.


Report •

#9
November 30, 2015 at 03:05:14
Hi Des,

maybe a silly question, but are *ALL* fans starting, including the one in the psu?

Good Luck - Keep us posted.


Report •

#10
November 30, 2015 at 05:37:56
I'm wondering if the thermal compound has deteriorated over time?

If the rig has been sitting for a few years untouched it could be that the integrity of the bond between the CPU and heat sink has been diminished enough that the bond was broken when the computer was moved.


Report •

#11
November 30, 2015 at 12:52:12
Could be motherboard trouble. Electrolytic capacitor dielectrics can fail if there has been no power around for years.

I would still use that pencil eraser on the RAM and Video Card edge connectors just in case. Same applies to SATA signal and power connectors.

Are you getting the right voltages out of the PSU?

Always pop back and let us know the outcome - thanks


Report •

#12
November 30, 2015 at 13:13:00
To check the PSU, try the paperclip test.

http://www.seasonic.ru/tech-support...


Report •

#13
November 30, 2015 at 20:07:15
Will not power up without RAM but should give you Beep tone of missing RAM if you have a chassis beeper attached. Try cleaning the RAM.

Air in the water lines will cause you this issue. If there is an air bubble within the CPU 'block' then the time on will be a second or two or less. If there is water in the 'Block' then it will remain on until the water has moved out of the 'block' or has boiled off, then it will shut down to prevent overheating.
With an air cooled system which has some mass to the heat sink, the system will stay on longer without the fan because the heat sink will soak up quite a bit of heat from the CPU but with the water cooled system there is very little thermal mass to absorb the heat without the water. So yes, this is a real possibility. Try an inexpensive heat sink with fan to see if this is the issue and/or see if there is a way to 'refill' the 'closed' system you have (Google the question and see if there are any videos out there with a procedure).

I did not realize that Electrolytic capacitors could dry out so that is a possibility, and I guess that thermal compound could have become dry depending on which type you originally used, but there is no reason that a CPU sitting unused could have developed an issue.

Other suggestions I see above also have merit like the paperclip test, so do not give up yet on it.

You have to be a little bit crazy to keep you from going insane.


Report •

#14
November 30, 2015 at 20:11:17
You said in # 4 you verified the power supply was OK. How did you do that? While stored was it 'off limits' to everyone? If it has a 110/220 voltage switch on the PSU make sure it's in the correct position. Don't guess and only switch it when the power cord is disconnected.

If you haven't already done so, temporarily remove all add-on cards and disconnect all the drives, including anything connected via USB. Then see if it powers up.

For how may years was it stored? And while stored I assume it wasn't in some damp basement? I'm having a hard time believing the electronics could have gone to pot in just a few years of non usage. A few decades, maybe, but give some credit to all the Chinese workers who make our electronics. . .

message edited by DAVEINCAPS


Report •

#15
December 1, 2015 at 21:15:46
I took it to work today & the IT guy tested it & it powered up almost every time for him, though he found it to be overheating. It overheated a few times & shut off on him, but when he took an external fan and kept it blowing on the CPU it stayed on for him. Good news, right? I just need a new heat-sink & CPU fan... So after a 3-4 hour debacle trying to get a new heat-sink/fan from Fry's after work, I finally got a new one installed. I turn on the power switch.... The system powers up, the fans inside start spinning, for 1/2 second & then it all shut off again. Then I try again, and no power, no nothing. The behavior is exactly the same as before. The new cooling system did exactly nothing to fix that problem. I kinda didn't think it would fix it, since for me, it either doesn't come on at all or doesn't stay on long enough to get hot.

To try to answer all your questions:
- The computer was kept in storage in a walk-in closet on the floor under my clothes for about 2-3 years. It was kept at room-temperature and normal (dry) humidity the whole time.
- It was extremely dusty inside when I got it out and my first inclination when it failed to power on was that there must be a short somewhere from all the dust. I have removed almost all the dust, and verified that it is not the problem.
- Before it was suggested here, I verified the PS operation using the paperclip test. The PS stayed on while the appropriate pins were shorted.
- I had also tried removing every component, every non-critical wire, from the motherboard (including hard drives, RAM, CD-rom, audio, etc.) in order to see if a short in some connected wire was the problem. It was not.
- I also removed the motherboard from the chassis and tried to power it on while it wasn't touching anything conductive, to see if it were a problem with the mounting. Again it was not.
- The motherboard has solid-state capacitors, so them drying out is also not a problem.
- The success the IT guy at work had with starting it up proves that none of the critical electronic components have failed. The CPU, motherboard, power supply, hard drives, RAM, all worked normally for him once he prevented the overheating.

I guess I'll have to take it back to him tomorrow & see what he says about it now. :( I'm going to have to buy him several drinks if he can figure out how to get this thing working reliably.

message edited by destenson


Report •

#16
December 1, 2015 at 23:47:41
✔ Best Answer
I am interested that you still remain convinced that it is not an overheating problem when your IT guy seems to have proved pretty conclusively that it is. I think your best bet is to let him have a go at fixing it, as you suggest.

Report •

#17
December 2, 2015 at 00:01:06
It may be the heatsink/fan assembly isn't adequate or not correctly installed.

Another thing to try is to take off the side panel and point a regular room box fan at the motherboard and turn it on. Of course that's not a permanent fix but if that arrangement works then overheating somewhere must be the problem.


Report •

#18
December 2, 2015 at 00:02:17
My experience (as an embedded software engineer) with these things says that for this particular problem (not powering on) it's exhibiting an electrical/electronics problem, not a thermal one, however my experience could be tainting my opinion.

I'm not saying there isn't an overheating problem.There is definitely *also* an overheating problem. There always has been with this computer. The water-cooling system never worked that well. It's my first experience with more than a heat-sink and a fan & it could have been installed better. And maybe I did a bad job installing the new heat-sink too. I'll ask him to re-install it if necessary.


Report •

#19
December 2, 2015 at 00:03:35
Dave, that's how the IT guy was able to keep it on. He put a small but powerful desk fan pointing directly at the cpu with the case off.

Report •

#20
December 2, 2015 at 01:12:22
OK, yeah I missed that. Possibly something else is overheating or the cpu itself may have been previously damaged and is more sensitive to heat now.

Report •

#21
December 2, 2015 at 01:55:14
Not powering on at all is very different from powering on for a second or so and then powering down. A second is plenty enough time for a CPU to get hot enough to trigger thermal cutout.

Report •

#22
December 2, 2015 at 01:56:47
So apparently, I don't know how to install thermal grease. After (d'oh) I installed the new heat-sink/fan assembly, I looked up how to properly apply thermal compound. I used way too much, probably both times (when I first built it & now). So, I removed the new heat-sink and cleaned off all the thermal grease as much as I could from both the cpu & heat-sink with isopropyl alcohol (only 70%, 90% would've been better) and q-tips. Then reinstalled it using only a small dab of thermal grease, between the size of a grain of rice & a pea. Long story short, that appears to have been "the" problem. I haven't tested extensively, and powering up is still a little quirky, but it appears to power on & boot up & stay on now, like it did for the IT guy. Progress!!

Thanks for everyone's suggestions. I didn't want to believe what was staring me in the face (that the prior minor heating problem was now a critical problem).

While I've built several computers in the past, this is my first one since a Pentium-II 66Mhz that lack of heat dissipation has been a big problem. I didn't know a cold i7-3.4GHz would only run for less than a second with improper heat dissipation.

Thanks again.


Report •

#23
December 2, 2015 at 02:18:52
Good news!

It's one of the rare cases where more isn't better. It is essential to have complete even coverage though.

message edited by btk1w1


Report •

#24
December 2, 2015 at 11:41:43
Solid state capacitors, sounds like it was quite advanced for its time.

Thanks for the feedback and glad to hear about your progress.

Always pop back and let us know the outcome - thanks


Report •


Ask Question