[SOLVED] How to fix Overheated PC???

ThemegaHG

Prominent
Mar 26, 2019
57
3
535
Hey, My PC specs are at bottom. Thanks in advance! I have had this issue alot for more resons on another thread, but my computer has overheated and BSOD about 10 times over its lifespan. I went upstairs and came down 10 mins later and found my computer commited suicide by killing MSI afterburner for no reson while heroes and generals was running. It supposedly died and when I came down and went to touch the PC, it was 90F~ for the case. I went to restart it with a simple click of the power button, yet alas, nothing. Nothing at all. No fan weeee or LED in sight. I'd assume its a power issue, but I cant be certan becasue this is completely new to me. Can you give my any suggestions to get it functioning again? Is it a power issue or Mobo or anything? I am just really worried as im broke as hell and just bought this with the last bit of my money to relax and now its jacked. I've tried power cord circuit breaker different power socket, not sure if it's the psu still or what thank you so much.

Please help me figure witch component ( if any ) Is damaged and needs replacing, or if its some automatic shutdown and I can necro it to life, pleassssse advise.

Ive checked other threads but nothing close to this specific problem.

PC SPECS:
Eclipse p300 Temperd Glass.
3x case fans
XPG SSD 256gb
XFX rx 580
Ryzen 1600
Team group Vulcan 16gb RAM
MSI B450M pro
Corsair CX550m PSU (bronze I think)
 
  • Like
Reactions: CallMeFriend
Solution
It lights up, so I guess it's not dead, but perhaps the psu throws up the white flag once you plug it in.
So to be sure, you'd need to test with another gpu, or try the gpu in another system.
Hi,
  • What do you mean with the case been 90F~?
  • Also before it never turn on again thing, did you see GPU or CPU temps reported by MSI afterburner?
  • Are you sure the CPU cooler is well seated and the screws are tightened enough in an X pattern?
  • Did the cooler have any sticker on its base?, Maybe you forgot to remove it ? (sotck cooler as far as I remember doesn't come with a sticker for any Ryzen CPU).
  • If you turn the PC and it boot, have you tried to go into the BIOS and see what the CPu temps looks like in there?
  • Whats your room temp ?
- There are some overheating safeguards, but as soon as the temps go down, it should turn on again (if this was the issue in your system).
 

ThemegaHG

Prominent
Mar 26, 2019
57
3
535
Hi,
  • What do you mean with the case been 90F~?
  • Also before it never turn on again thing, did you see GPU or CPU temps reported by MSI afterburner?
  • Are you sure the CPU cooler is well seated and the screws are tightened enough in an X pattern?
  • Did the cooler have any sticker on its base?, Maybe you forgot to remove it ? (sotck cooler as far as I remember doesn't come with a sticker for any Ryzen CPU).
  • If you turn the PC and it boot, have you tried to go into the BIOS and see what the CPu temps looks like in there?
  • Whats your room temp ?
- There are some overheating safeguards, but as soon as the temps go down, it should turn on again (if this was the issue in your system).

Hey! thank you for replying to this!

So when I came back downstairs I noticed the PC was just off. No small blue light on my RX 580 or red case fan LED. Just blank. I Just assumed it overheated because MSI does this a lot for no reason (which I asked in another still unanswered thread) and It will just turn off randomly. But I was running a graphically intensive game at the time while I was gone and it must have just skyrocketed heat and then mabey kept on heating up because, to be honest with all the times that this has happened, the first the audio started doing the VERY loud audio loop and the screen froze, it gave me the dreaded Blue Screen of Death, and then I manually do a hard shutdown by holding the power for 3-4 seconds. Though I've never just left it at the BSoD and let it stay really hot, Mabey that's why it was damaged but I head mobos have heat failsafe so Im pretty clueless on what to do right now.

Sorry, When I went to touch it to check the heat before trying to see if it was secretly on so I could check MSI, It was, to my touch, about 85-90 Degrees Farenheight. (I use ~ to mean give or take a few, kind of like the word ish)

My computer sadly was already off before I could check MSI as the temps were fine and MSI was on before I went upstairs.

I have used for about 8~ months and the CPU stays nice the few times I've checked it under load, barely touching 60 sometimes in physics simulators, so I don't know why the RX refuses to run a bit cooler and stop BSoDing.

The screws to the CPU heatsink are fastened nicely and I don't see any problems with them, the room temp when this occurred yesterday was 79 on the thermostat.

I have not been able to boot since It shut off yesterday.
Hope this helps!!!
 

ThemegaHG

Prominent
Mar 26, 2019
57
3
535
The earlier Ryzens were real picky about the ram you could put in, and I'm not too confident on how reliable that Team ram is, but that might have been your BSOD issue.

Psu is probably dead. While the RX 580 is a strong 1080p card, it's not power efficient.

Hey, thank you! I actually took my PSU out and did the supposed test where you put a paperclip in the 24 pin fan header and turn it on and the fan started up in the PSU unlike when it was connected to the Mobo. I checked w/ an Ohm meter too while it was "on" and it did have current but I will be sure to try to start the PC with another PSU to see if the PSU is stll the issue.
 

ThemegaHG

Prominent
Mar 26, 2019
57
3
535
Update: I took the PSU halfway out and check the power cord that connects to the PSU with an Ohm meter, and it was fine. I then put a paper clip in the PSU's 24 pins to turn it on, and supposedly if the fan spins when you do this it's working. And surprisingly the fan spun up silently. So if the PSU seems fine, what could cause this to not even power up? is there a severed line somewhere so a current cant even flow? I don't really know what to check because I don't know how the overheating could have busted the Motherboard. Is it a RAM issue? What should I try to replace to get it to function? Sorry I have done a lot of forums and I get 2~ answers and then get buried and nothing works.
 

gamerbrehdy

Honorable
Jun 15, 2018
320
29
10,790
Check your room temps. If it's really hot in the room whre your computer is located... go to the beach or something :D.

It's probably the best idea to run the pc with open case until you found a permanent solution.
Maybe pointing a desk fan at your open pc would help.

This sounds utterly ridiculous to do, but I do this every summer evening that I'm gaming.. It helps decently well.

I wouldn't be worried about noise if you're using headphones.
 
  • Like
Reactions: CallMeFriend
If the case was at 90ish° F to touch, thats only 32° C and its a pretty normal temp for a case while gaming, its not even hot. The idea of a metal case, is that the metal also help to disiapate heat.

Also most GPUs and CPUs are designed to work at 80° C without issues.

If you have 3 fans on the case, make sure atleast 2 of them are working as intake fans in the front of the case (to bring fresh air in from the outside) and the last one is at the back as an exhaust fan (taking the hot air out of the case). Also make sure the case has enough room around it, in the front, back and on the sides, for example this is really a horrible desing for a computer table: https://rukminim1.flixcart.com/imag...mal-black-original-imaf78g7cneyrdyp.jpeg?q=70.

Cheers
 
  • Like
Reactions: Phaaze88

ThemegaHG

Prominent
Mar 26, 2019
57
3
535
Woah thanks for all the replies.

Sorry about assuming the temp.

So first, my GPU fans run loud when I play games. Sometimes I will catch the fans quiet and quickly open MSI. My computer likes to close MSI afterburner and return the GPU fan curve to terrible. So every once in a while I will open it and then see the temperatures at max 92 degrees Celcius, and then quickly fall to 65c. Sometimes It will freeze up and I quickly shut it off and cursing myself for not checking MSI.

So I can ASSUME that at the time of it dying within the span of me going upstairs and coming back down, it must have reached high temps, something shutoff and caused the whole system to shutoff or some sort of damage. Or may it's not a heating issue, and something randomly out of nowhere quit working. Either way, thank you all so much for the suggestions and help.

But My PC still isn't even powering on so I can't test out most of the comments.

When I get home though (at school rn) I will surely check my RAM first, and then other things I think could work, (Cmos battery, different ram sticks, ect)

I will be back with results soon I hope. (stay posted please!) :)
 

ThemegaHG

Prominent
Mar 26, 2019
57
3
535
Are you running with only the single included 120mm fan that came with your case?
No, I am using a Thermaltake riing red LED fan. 2 intakes running at 60% power and the exaust running at 60% also. The top isn't working because I connect it to a motherboard fan header and it draws too much power so I can only use the 3. Hope this helps.
 
Nov 20, 2019
1
0
10
It's sad, especially since it was your last money - it must end well :)
I came here for advice, but it could be a matter of power supply and ram IMO
My computer only overheated while playing games with considerable hardware requirements. For example, when I play Solitaire, everything works well. But as soon as I start Escape from Tarkov and try to buy some weapons and items, then my computer begins to take temperature and slowly overheats. What could I do in such situations to be able to buy weapons and items in EFT? Buying EFT Items is a priority when it comes to gfrania in Escape From Tarkov. Where can I buy EFT items? In general, I highly recommend and I am a very satisfied buyer since I buy my items on Odealo https://odealo.com/games/escape-from-tarkov/items
cheap, safe, what do you need more?
 
Last edited:

Phaaze88

Titan
Ambassador
My computer likes to close MSI afterburner...
Using any 3rd party antivirus?

The top isn't working because I connect it to a motherboard fan header and it draws too much power so I can only use the 3.
What does this mean?
Unless the top fan is a Delta 5000rpm fan or something, the 'draws too much power' thing doesn't add up, as most mobo headers are rated for 1 amp of power draw - some do 2A though.
 
  • Like
Reactions: RodroX

rubix_1011

Contributing Writer
Moderator
Before you do ANYTHING ELSE:

Airflow / Fan / Cooling / Overheating - How to test, steps to resolve

This could be caused by a few different things, please don't automatically assume 'the cooler is not working' without also checking if the case airflow is sufficient.

Remove the side panel of the PC case. Orient a house fan (desk or box style fan) to blow air into the case, directly over components at the highest setting.

This will represent a case with the best possible airflow possible. For reference, the fans I am providing as examples would look like the items below (just to clarify for anyone who might want reference)

nI6vx5v.jpg
2GBempv.jpg


Re-test as you have normally done - play games, run benchmarks, etc. to get to where temperatures were normally seen to be higher than they should. Normal room temperature is usually between 20-24C or 68-75F. Please note that every air or liquid cooler operates as a product of delta-T over ambient, meaning that if the PC is operational (simply turned on), it is impossible for the CPU to display a temperature below ambient room temperatures. If it is, this is likely a bug in software temperature reporting either from the desktop UI or the BIOS reading it incorrectly.

With the fan running at full speed, if temperatures drop by 5-7C or more, case airflow is one major issue to contend with. You will need additional fans or better fans for your setup in order to optimize air in and out of the chassis. This might even require consideration for a new PC case or leaving the side panel partially open during sessions of heavier computing until these items are corrected.

If your temperatures remain relatively the same (difference less than 1-2C), then you likely have an issue with the cooler in question (if CPU is hot, CPU cooler, if GPU is hot, GPU cooler). It would be good to then approach the next steps by thoroughly cleaning the cooler with compressed or canned air and ensuring there are not large blockages in cooling fins or on fans, etc. This might require the cooling fans to be removed from the heatsink or radiator to ensure there is not a buildup of pet hair, dust or even carpet fibers which can trap additional debris. Please ensure the PC is turned off and unplugged during this process to prevent unwanted startup to keep fingers safe from fan blades or accidental shorting if you happen to drop a screw onto other components during fan removal.

Removal of the cooler and re-application of thermal paste & re-seating the cooler can also be beneficial once cleaning of the cooler is ruled out by retesting the steps above.
 

ThemegaHG

Prominent
Mar 26, 2019
57
3
535
Using any 3rd party antivirus?


What does this mean?
Unless the top fan is a Delta 5000rpm fan or something, the 'draws too much power' thing doesn't add up, as most mobo headers are rated for 1 amp of power draw - some do 2A though.

No, but I can double-check once I get it working again.

I actually tried to solve this in another post here,
https://forums.tomshardware.com/thr...e-led-fans-while-having-enough-power.3497819/
If you scroll down you see Karadjgne talking about the wire resistance and fan header and fancy stuff. I wanted them all to be fan curveable but I just gave up and left one (the topmost fan) unconnected. I may just plug a few into the PSU for 1 constant airspeed. At least then I would have some more exaust. I also don't fully understand how I cant connect all 4 LED fans to 1 PWM Fan splitter and have them all running and able to be fan curved.
 

ThemegaHG

Prominent
Mar 26, 2019
57
3
535
Before you do ANYTHING ELSE:

Airflow / Fan / Cooling / Overheating - How to test, steps to resolve

This could be caused by a few different things, please don't automatically assume 'the cooler is not working' without also checking if the case airflow is sufficient.

Remove the side panel of the PC case. Orient a house fan (desk or box style fan) to blow air into the case, directly over components at the highest setting.

This will represent a case with the best possible airflow possible. For reference, the fans I am providing as examples would look like the items below (just to clarify for anyone who might want reference)

nI6vx5v.jpg
2GBempv.jpg


Re-test as you have normally done - play games, run benchmarks, etc. to get to where temperatures were normally seen to be higher than they should. Normal room temperature is usually between 20-24C or 68-75F. Please note that every air or liquid cooler operates as a product of delta-T over ambient, meaning that if the PC is operational (simply turned on), it is impossible for the CPU to display a temperature below ambient room temperatures. If it is, this is likely a bug in software temperature reporting either from the desktop UI or the BIOS reading it incorrectly.

With the fan running at full speed, if temperatures drop by 5-7C or more, case airflow is one major issue to contend with. You will need additional fans or better fans for your setup in order to optimize air in and out of the chassis. This might even require consideration for a new PC case or leaving the side panel partially open during sessions of heavier computing until these items are corrected.

If your temperatures remain relatively the same (difference less than 1-2C), then you likely have an issue with the cooler in question (if CPU is hot, CPU cooler, if GPU is hot, GPU cooler). It would be good to then approach the next steps by thoroughly cleaning the cooler with compressed or canned air and ensuring there are not large blockages in cooling fins or on fans, etc. This might require the cooling fans to be removed from the heatsink or radiator to ensure there is not a buildup of pet hair, dust or even carpet fibers which can trap additional debris. Please ensure the PC is turned off and unplugged during this process to prevent unwanted startup to keep fingers safe from fan blades or accidental shorting if you happen to drop a screw onto other components during fan removal.

Removal of the cooler and re-application of thermal paste & re-seating the cooler can also be beneficial once cleaning of the cooler is ruled out by retesting the steps above.

Thank you very much for this amazing article. I will 100% try when I get my computer operation again.
 

rubix_1011

Contributing Writer
Moderator
Not an article, I write that so often due to questions like this on the forum, I simply have a copy/paste of text rather than re-writing the same thing over and over and over again.

It really is just the same question asked 1000 different ways.
 

gamerbrehdy

Honorable
Jun 15, 2018
320
29
10,790
Before you do ANYTHING ELSE:

Airflow / Fan / Cooling / Overheating - How to test, steps to resolve

This could be caused by a few different things, please don't automatically assume 'the cooler is not working' without also checking if the case airflow is sufficient.

Remove the side panel of the PC case. Orient a house fan (desk or box style fan) to blow air into the case, directly over components at the highest setting.

This will represent a case with the best possible airflow possible. For reference, the fans I am providing as examples would look like the items below (just to clarify for anyone who might want reference)

nI6vx5v.jpg
2GBempv.jpg


Re-test as you have normally done - play games, run benchmarks, etc. to get to where temperatures were normally seen to be higher than they should. Normal room temperature is usually between 20-24C or 68-75F. Please note that every air or liquid cooler operates as a product of delta-T over ambient, meaning that if the PC is operational (simply turned on), it is impossible for the CPU to display a temperature below ambient room temperatures. If it is, this is likely a bug in software temperature reporting either from the desktop UI or the BIOS reading it incorrectly.

With the fan running at full speed, if temperatures drop by 5-7C or more, case airflow is one major issue to contend with. You will need additional fans or better fans for your setup in order to optimize air in and out of the chassis. This might even require consideration for a new PC case or leaving the side panel partially open during sessions of heavier computing until these items are corrected.

If your temperatures remain relatively the same (difference less than 1-2C), then you likely have an issue with the cooler in question (if CPU is hot, CPU cooler, if GPU is hot, GPU cooler). It would be good to then approach the next steps by thoroughly cleaning the cooler with compressed or canned air and ensuring there are not large blockages in cooling fins or on fans, etc. This might require the cooling fans to be removed from the heatsink or radiator to ensure there is not a buildup of pet hair, dust or even carpet fibers which can trap additional debris. Please ensure the PC is turned off and unplugged during this process to prevent unwanted startup to keep fingers safe from fan blades or accidental shorting if you happen to drop a screw onto other components during fan removal.

Removal of the cooler and re-application of thermal paste & re-seating the cooler can also be beneficial once cleaning of the cooler is ruled out by retesting the steps above.

I commented the desk fan method allready.

It is indeed the golden standard for litterally any sweaty pc 😂
 

Phaaze88

Titan
Ambassador
No, but I can double-check once I get it working again.

I actually tried to solve this in another post here,
https://forums.tomshardware.com/thr...e-led-fans-while-having-enough-power.3497819/
If you scroll down you see Karadjgne talking about the wire resistance and fan header and fancy stuff. I wanted them all to be fan curveable but I just gave up and left one (the topmost fan) unconnected. I may just plug a few into the PSU for 1 constant airspeed. At least then I would have some more exaust. I also don't fully understand how I cant connect all 4 LED fans to 1 PWM Fan splitter and have them all running and able to be fan curved.
With the way the fan splitters are designed, only one fan gets to communicate with the motherboard, the PWM controller is purposely missing on the others:
02-1-1030x791.jpg


The set curve on the 'main fan' is automatically transmitted to the other fans.
So, it's impossible to set differing curves on each fan on a splitter; that requires another mobo header. But there's only so many on these boards...
 
  • Like
Reactions: RodroX

ThemegaHG

Prominent
Mar 26, 2019
57
3
535
Huge update,

I took the CMOS battery out, waited put it back in, reinstalled RX580, plugged it in, CPU fan came on and I got a CPU led!!! Going to reinstall all connections and update you again.

Thank you all so much again for the replies to help me, it means a lot.
 

ThemegaHG

Prominent
Mar 26, 2019
57
3
535
With the way the fan splitters are designed, only one fan gets to communicate with the motherboard, the PWM controller is purposely missing on the others:
02-1-1030x791.jpg


The set curve on the 'main fan' is automatically transmitted to the other fans.
So, it's impossible to set differing curves on each fan on a splitter; that requires another mobo header. But there's only so many on these boards...

So I actually hooked my PC up but I unplugged the GPU and
With the way the fan splitters are designed, only one fan gets to communicate with the motherboard, the PWM controller is purposely missing on the others:
02-1-1030x791.jpg


The set curve on the 'main fan' is automatically transmitted to the other fans.
So, it's impossible to set differing curves on each fan on a splitter; that requires another mobo header. But there's only so many on these boards...

So I actually unplugged and resituated the GPU and the fans and LED's on the motherboards started up to life witch is amazing, though now the GPU isn't plugged in with the 6+2 cable.

The GPU is on the mobo, and has a RED light witch means not a proper power connection. Although when I plug the 6+2 into the GPU and try to turn the computer on, nothing will happen. Is the GPU dead? Please tell me it's not. the LED on it is coming on right?

Is it RAM?