Question Old Gaming PC Restarts within seconds and doesn't get to post

Nov 16, 2024
5
0
10
Hi everyone,

Specs:
16 GB RAM (DDR3?)
Intel i5
Geforce 970
Asus Maximus Ranger VII board

I have an older gaming PC that recycled to use as a desktop for my fiancee. It's an older model but it works fine until a few days ago.

When you press the power button, some lights start blinking, all fans start to spin but after 2-3 seconds the PC shuts off again and restarts, endlessly rebooting. There is no Post and no beep sounds. The system restarts by itself before it can even get there.

Short video (no sound) of the issue: View: https://imgur.com/a/XAu6ZPM


Here is a list of things that I have tried:
  • Reseat/start without RAM
  • Reseat/start without graphics card
  • Unplug all drives
  • Reset CMOS
  • Get a new battery
  • Bought a brand new PSU and tried with that
  • Unplugged power button and reset button and booted using a screw driver
  • Search this and other forums for answers but nothing I found has helped yet.
I am not a hardware person so I really have no idea what I am doing. The only two that I can think of before bringing the PC to a repair shop is removing it from the case? I haven't done that yet because I have little experience with these things and it would also require me to undo all the cable fasteners and probably remove several other components as well which sounds daring and like hours of work. I am grateful for any insights or further things I could try.
 
Last edited:
brand and model of the psu?
The current PSU is a fancy but very old Nesteq Xzero and I bought a brand new BeQuiet System Power 9 to test if it's the PSU (700W both) but in both cases the error persist. For me that means that it is not directly related to the PSU and the old PSU works just fine.
 
take the gpu out and test with igpu?
I already tried that. Same with RAM and drives. Maybe my description is misleading.

I have updated the topic to better explain what I tried.

Later today I will attempt to remove the board from the case and assemble the computer on the kitchen table.
 
Last edited:
Hi everyone,

Specs:
16 GB RAM (DDR3?)
Intel i5
Geforce 970
Asus Maximus Ranger VII board

I have an older gaming PC that recycled to use as a desktop for my fiancee. It's an older model but it works fine until a few days ago.

When you press the power button, some lights start blinking, all fans start to spin but after 2-3 seconds the PC shuts off again and restarts, endlessly rebooting. There is no Post and no beep sounds. The system restarts by itself before it can even get there.


Here is a list of things that I have tried:
  • Reseat/start without RAM
  • Reseat/start without graphics card
  • Unplug all drives
  • Reset CMOS
  • Get a new battery
  • Bought a brand new PSU and tried with that
  • Unplugged power button and reset button and booted using a screw driver
  • Search this and other forums for answers but nothing I found has helped yet.
I am not a hardware person so I really have no idea what I am doing. The only two that I can think of before bringing the PC to a repair shop is removing it from the case? I haven't done that yet because I have little experience with these things and it would also require me to undo all the cable fasteners and probably remove several other components as well which sounds daring and like hours of work. I am grateful for any insights or further things I could try.

This is often a problem with the PSU, the reason I will not go fully into. If the PSU is trying to draw too much power, the computer will display your problem. Another issue could be the CPU thermal paste. Most thermal paste has to be reapplied every year, with a few every 4 years and Noctua NT-H2 paste is 5 years. Suggest you remove the CPU heat sink and take a look at the TIM (paste), if it is dry or looks dry it needs replaced. Also, TIM thins out over time leaving uncovered areas, this causes heat buildup.

There are other things that can cause the problem, however, checking the TIM is important on an old computer. A closed electrical circuit can also cause this. I have found that issue in different areas, such as the motherboard touching the case, a bare wire touching the case, even an old fan that had gone bad.
 
This is often a problem with the PSU, the reason I will not go fully into. If the PSU is trying to draw too much power, the computer will display your problem. Another issue could be the CPU thermal paste. Most thermal paste has to be reapplied every year, with a few every 4 years and Noctua NT-H2 paste is 5 years. Suggest you remove the CPU heat sink and take a look at the TIM (paste), if it is dry or looks dry it needs replaced. Also, TIM thins out over time leaving uncovered areas, this causes heat buildup.

There are other things that can cause the problem, however, checking the TIM is important on an old computer. A closed electrical circuit can also cause this. I have found that issue in different areas, such as the motherboard touching the case, a bare wire touching the case, even an old fan that had gone bad.
The thermalpaste is probably 8+ years old. Would a completely wrecked old thermal paste explain why the PC ran for a few months and then no longer start because the thermal paste finally gave up?
How long does it take for the heat safety to kick in after start? Could that be the observed 1-2 seconds right after power on?
 
The thermalpaste is probably 8+ years old. Would a completely wrecked old thermal paste explain why the PC ran for a few months and then no longer start because the thermal paste finally gave up?
How long does it take for the heat safety to kick in after start? Could that be the observed 1-2 seconds right after power on?

Yes, short duration ON then OFF can be caused by this.
 
Yes, short duration ON then OFF can be caused by this.
I understand that it could go really quickly, but would you say that a completely cold i5 would go from power on to heat shutdown within 2 seconds before the system beeps?
 
That narrows it down to a few dozen CPUs.

Next thing I'd do is to replace the CPU. Anything that fits and is compatible will do. Could be an i3. Just to prove if it's the CPU or maybe the motherboard.

You might even find that simply reseating the CPU fixes the problem.

When the CPU is out, check for tarnished or dirty gold-plated contacts on the underside of the processor and look for bent pins in the CPU socket.

Replace the CPU carefully back into the socket. If you drop the CPU on one edge, you could bend the delicate pins in the socket. A magnifying glass or a closeup photo of the contacts might reveal one or more out of place. Remember to clean off the old paste and then apply new paste.

I have loads of old LGA1150 boards so I'd just swap two processors over. Failing that, I'd buy a used CPU on eBay. Most of the time they work fine. $10 to $15 on another CPU might show your i5 has died.

The only time I replace thermal compound is when I'm fitting a new cooler or replacing a CPU. I've never seen any problems when replacing coolers that have been working for years.

You only need the thinnest of layers for good heat conduction. Too much paste is just as bad as too little. A small pea-sized blob is enough on most CPUs (apart, possibly, from LGA1700 and larger Xeons). I might even try some Graphene pads one day.
https://www.howtogeek.com/what-is-a-graphene-thermal-pad/

On ancient PCs, I use a tube of white Aluminium Oxide paste from RS which is more than 25 years old. For newer builds, it tends to be Noctua NT-H2. You might have problems with Arctic MX-4, which apparently starts to separate in tube after a few months.

When you press the power button, some lights start blinking, all fans start to spin but after 2-3 seconds the PC shuts off again and restarts, endlessly rebooting.
During POST, the BIOS is looking for a compatible CPU, RAM, GPU and then it searches for bootable drives. If your BIOS is still OK, then it's failing to find a vital component. Hence the restarts.

Have you tried booting with only one DIMM? Have you tried any other DIMMs. If you have another compatible computer, transfer the RAM over and run MemTest86 from bootable USB.

At the moment I'm thinking it's either CPU or mobo.
 
  • Like
Reactions: BlacKcuD