Back from vacay, no POST. Next steps?

cbliss1000

Limp Gawd
Joined
Jul 15, 2009
Messages
361
Hello! My beloved PC (specs in sig) is facing some existential issues and I'm concerned--any help is appreciated. Recently, I gently moved the case about 6 feet along a carpet to accommodate a guest, then left for a long weekend on some much needed R&R. The computer lives in my basement, which is humid'ish. When I returned, I went to boot it up to enjoy some Civ 6; no success. The PSU, GPU, and the fans all turn on for half a second, then power off, then on again, then off again, then on, off, on, etc. It's a hopeless cycle of not turning on.

I've tried clearing CMOS and reseating the mobo power cable to no effect. Can anyone suggest my next move? Likely culprits? What could have happened?

Possibly relevant: this computer has always had a single cycle of on/off before booting. During normal operation, it would spin up once, stop, then boot normally. I never thought much of it.

Thank you in advance, heroes.
 
Last edited:
Your description Sounds like a short fault in one of your components.
I'll assume, for the moment, it is your sig rig. I'd bet it's the motherboard due to the "possibly relevant" comment above, but you have to go about it methodically. (Step 1) Start by going with a minimum config meaning only essential parts to boot. Remove all other peripherals. For example one stick memory no optical or backup drives no lighting or expansion devices etc. (BARE MINIMUM). If you still have the problem go to step 2 if it boots fine, add your stuff back in one at a time booting after each new device until you find the culprit. (Step 2) Try all your significant components in another system if the CPU, GPU, memory, and PSU work fine there then you know for sure it is the motherboard. If you don't have another PC of the same ilk ask a friend who has a similar spec system to swap out their parts for yours. No friend willing? Make friends at your local PC repair shop and have them test your junk. Either way, I'd start with the PSU, but the easiest would be the GPU, memory, and CPU. The repair shop should be able to test the PSU easily without having to install in an alternate system.
 
Your description Sounds like a short fault in one of your components.
I'll assume, for the moment, it is your sig rig. I'd bet it's the motherboard due to the "possibly relevant" comment above, but you have to go about it methodically. (Step 1) Start by going with a minimum config meaning only essential parts to boot. Remove all other peripherals. For example one stick memory no optical or backup drives no lighting or expansion devices etc. (BARE MINIMUM). If you still have the problem go to step 2 if it boots fine, add your stuff back in one at a time booting after each new device until you find the culprit. (Step 2) Try all your significant components in another system if the CPU, GPU, memory, and PSU work fine there then you know for sure it is the motherboard. If you don't have another PC of the same ilk ask a friend who has a similar spec system to swap out their parts for yours. No friend willing? Make friends at your local PC repair shop and have them test your junk. Either way, I'd start with the PSU, but the easiest would be the GPU, memory, and CPU. The repair shop should be able to test the PSU easily without having to install in an alternate system.
Super helpful, thanks. That's what I was fearing I'd have to do. Sigh. I'll post back here once I've had a sec to do this. Fingers crossed it's easily resolvable!
 
Seems like the troubleshooting steps have been covered above, that is frustrating to come back to. I hope you figure it out quickly. Another option is to try to clear CMOS due to your on/off cycle that happens on initial post. I don't see how moving a desktop could cause that.
 
I have seen this symptom many times when working on computers in the last 20+ years. In order of most often to least often, here are the possible causes that I have seen.

1. PSU
2. motherboard
3. expansion card
3. fan (3-4 times)
4. HDD (once)
5. keyboard (once)

You will also probably want to check the outlet the computer is plugged into to make sure you are getting proper voltage.
 
make sure everything is seated correctly. test the cmos battery. ive seen dying ones cause all sorts of weird issues. then try another psu. start there. if no luck, rip it out of the case and get it down to the basics as suggested by Ow.
 
I forgot to mention I had a motherboard once that was shorting out on the motherboard tray by an errant screw that got between there. Assuming this may have happened in your case and the short hasn't produced irrevicable damage you could try the motherboard on the benchtop before committing to replacing it but heres to hoping it's something more simple. GL
 
Thanks all. Took advantage of a sick day to take the recommended steps and I'm thinking it's probably the mobo. Here's what I did:
  • Removed & reseated CMOS battery
  • Tried a different PSU
  • Disconnected GPU, HDDs, and all peripherals, one by one
  • Removed all but 1 stick of RAM
  • Removed the mobo from case and attempted w/ only CPU and PSU
No luck. Every reboot attempt ended the same way (power cycling).

Assuming I'm not missing anything, I guess that means it's either the motherboard or CPU? Unfortunately, I don't have a separate X99 chipset or CPU, so I can't swap them out to test. I guess I could see if local repair shops could test them... but based on the comments above, it sounds like it's almost certainly the mobo.

Does anyone have any additional advice, or should I just suck it up and buy a new motherboard? Any sense in taking this to a shop (I've never done that before)?

Thanks.
 
Last edited:
I did not--don't have a multimeter. Good reminder though. I could probably borrow one from someone, or buy a new battery.
ive seen dying batteries cause all sorts of weird issues. borrow one from another mobo if you can or hit the dollar store for a new one.
 
I saw a weird issue like this recently on a Dell system that had a bent i3 in it. I replaced the i3 with an i5 that was known working and this time didn't fully tighten down the heatsink and it booted and is working as I type.

I know you don't have a way to test the mb, but getting a cheap celeron/pentium that is compatible for $15-20 will be a really easy way to sort out which is bad.

Is anything still under warranty?
 
Take out all the ram sticks but one and try to boot. Then move the ram stick to the other sockets and boot and repeat until all sockets and all ram sticks have been tried. I had one socket go bad wasted hella amount of time troubleshooting. My setup at the time was 4x16gb sticks and I took 3 out. Of course the one stick I left in, was in socket A1 as per the manual. When I moved the same stick to another socket, it booted right up. Swap sticks into socket A1 and no go for any of them. Tried the same thing in any of the other sockets and all the sticks booted right up.

So bad A1 ram socket. I had never seen / had that happen before but goes to show you that weird shit happens sometimes.
 
Take out all the ram sticks but one and try to boot. Then move the ram stick to the other sockets and boot and repeat until all sockets and all ram sticks have been tried. I had one socket go bad wasted hella amount of time troubleshooting. My setup at the time was 4x16gb sticks and I took 3 out. Of course the one stick I left in, was in socket A1 as per the manual. When I moved the same stick to another socket, it booted right up. Swap sticks into socket A1 and no go for any of them. Tried the same thing in any of the other sockets and all the sticks booted right up.

So bad A1 ram socket. I had never seen / had that happen before but goes to show you that weird shit happens sometimes.
only had that happen once but is a good tip for troubleshooting.
 
Take out all the ram sticks but one and try to boot. Then move the ram stick to the other sockets and boot and repeat until all sockets and all ram sticks have been tried. I had one socket go bad wasted hella amount of time troubleshooting. My setup at the time was 4x16gb sticks and I took 3 out. Of course the one stick I left in, was in socket A1 as per the manual. When I moved the same stick to another socket, it booted right up. Swap sticks into socket A1 and no go for any of them. Tried the same thing in any of the other sockets and all the sticks booted right up.

So bad A1 ram socket. I had never seen / had that happen before but goes to show you that weird shit happens sometimes.
Just tried this, no luck. Good tip though.
I saw a weird issue like this recently on a Dell system that had a bent i3 in it. I replaced the i3 with an i5 that was known working and this time didn't fully tighten down the heatsink and it booted and is working as I type.

I know you don't have a way to test the mb, but getting a cheap celeron/pentium that is compatible for $15-20 will be a really easy way to sort out which is bad.

Is anything still under warranty?
Not a bad shout, though I'm not sure there's anything particularly cheap that's compatible w/ X99 (could be wrong though, as it has been a few years since I thought about this stuff). And unfortunately I'm about 5 months past warranty, though I might shoot ASRock an email (and link to this thread) to see if they can help.
 
Holy crap those the x99 cpus are expensive, so scratch that idea. It will be cheaper to get another motherboard for testing the cpu--plus that handles if the mb is bad too.
 
Take out all the ram sticks but one and try to boot. Then move the ram stick to the other sockets and boot and repeat until all sockets and all ram sticks have been tried. I had one socket go bad wasted hella amount of time troubleshooting. My setup at the time was 4x16gb sticks and I took 3 out. Of course the one stick I left in, was in socket A1 as per the manual. When I moved the same stick to another socket, it booted right up. Swap sticks into socket A1 and no go for any of them. Tried the same thing in any of the other sockets and all the sticks booted right up.

So bad A1 ram socket. I had never seen / had that happen before but goes to show you that weird shit happens sometimes.

This.. infinite bootloop cycle for me in all my life have always been related to RAM in any way of bad socket, bad stick or (which may be this case) a overtighten cooler messing with the motherboard PCB RAM lanes..

so definitively trying different RAM sticks on different sockets may help, also loosening a bit the cooler worth the shoot.
 
Thanks all. I reseated the CPU per your advice, no luck, and tried cold-starting the mobo with a screwdriver, in case the power switch was on the fritz. No luck with anything--still just power cycling. I'll try the different RAM in different slots thing again (tried once, but there are more combinations I could attempt). But other than that, sure looks like a fried motherboard to me. I guess these things just happen sometimes?
 
Thanks all. I reseated the CPU per your advice, no luck, and tried cold-starting the mobo with a screwdriver, in case the power switch was on the fritz. No luck with anything--still just power cycling. I'll try the different RAM in different slots thing again (tried once, but there are more combinations I could attempt). But other than that, sure looks like a fried motherboard to me. I guess these things just happen sometimes?

Unfortunately yes they do, ultimately everything breaks down. Hope you can pickup a cheap replacement
 
I'd try removing all the ram and see if it does the same thing. If it beeps or gives a different error, you still have a shot at it being a bad memory module/slot. (y)
 
Check for loose cables and try the bare minimum hw config to get PC to POST.
 
Just as an update to this--I was finally able to replace the motherboard and all is well in the world. Computer posts no problem. So, the last one must have shorted out or been stressed in some unknown way. C'est la vie. Thanks again to everyone for your helpful troubleshooting advice.
My only remaining unknown issue is that this new motherboard--which is the exact same model as the old one (ASRock x99 Extreme4)--exhibits some of the same turn-on/turn-off/POST behavior that the last build did (about every 4th power on, it spins on briefly, powers down, then tries again and posts). The last motherboard did this every single power on attempt (and, as a reminder, when it died it would just cycle turning on/off, fruitlessly). Not sure what to make of that, what causes it, and if whatever the root cause is led to the last one's death.
 
I would be curious how this new motherboard would work outside of the case--sounds like something somewhere is shorting something that eventually deteriorates.
 
I would be curious how this new motherboard would work outside of the case--sounds like something somewhere is shorting something that eventually deteriorates.
Yeah that seems plausible. I did have the new board out of the case and booting normally, with just the CPU, but haven't tried it outside the case with all peripherals. Perhaps this is a good excuse to just buy a new case--I'm running a heavily modded Antec 300 from, I dunno, 2007?
 
Yeah that seems plausible. I did have the new board out of the case and booting normally, with just the CPU, but haven't tried it outside the case with all peripherals. Perhaps this is a good excuse to just buy a new case--I'm running a heavily modded Antec 300 from, I dunno, 2007?
I would definitely try it outside the case with just basic peripherals attached and see how it does. If it doesn't have any issues, I'd carefully examine the case and then if you can't find anything wrong, either put a piece of paper under the motherboard and see if that helps or just replace the case completely. You definitely don't want to fry another board.
 
Back
Top