• Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.
  • The forums have been upgraded with support for dark mode. By default it will follow the setting on your system/browser. You may override it by scrolling to the end of the page and clicking the gears icon.

PSU causing PC to completely shut down at high usage moments.

Joined
Jan 17, 2025
Messages
4 (0.02/day)
System Name Windows 10
Processor Intel Core i7-7700
Motherboard H1110M PRO-D
Cooling Fans
Memory 12 GB
Video Card(s) NVIDIA GeForce GTX 1050
Storage 2 TB SSD
Display(s) -
Case -
Audio Device(s) Blue Yeti + NVIDIA High Definition Audio
Power Supply ASUS TUF Gaming 650w
Mouse [Not relevant]
Keyboard [Not relevant]
VR HMD -
Benchmark Scores -
Hi,


Ever since I got a new CPU: Intel Core i7-7700 my PC crashed when the power usage was high enough.
I've checked the Windows Event Viewer and found the cause: Kernel-Power Error in the category of 47.
I naturally figured out it was because of my old PSU (VS450), so I bought a new one, ASUS TUF Gaming 650W but the issue persists.
Then I tried updating my BIOS in case the CPU wasn't supported in some way, but that didn't fix the issue either.

I was under the impression that 650W can easily handle i7s power usage. What am I missing?
I did try all the usual "automatic windows repair" stuff.

To clarify, it just shuts down completely - then reboots. I can't find anything beyond those logs below. Any assistance would be much appreciated, I'm out of ideas.



The full log:

System

-Provider
[ Name] Microsoft-Windows-Kernel-Processor-Power
[ Guid] {0f67e49f-fe51-4e9f-b490-6f2948cc6027}

EventID55

Version0

Level4

Task47

Opcode0

Keywords0x8000000000000000

-TimeCreated
[ SystemTime] 2025-01-17T10:18:49.8810047Z

EventRecordID140492

Correlation

-Execution
[ ProcessID] 4
[ ThreadID] 420

ChannelSystem

ComputerPC-Computer

-Security
[ UserID] S-1-5-18

-EventData

Group0

Number3

IdleStateCount2

IdleImplementation1

NominalFrequency3600

MaximumPerformancePercent708

MinimumPerformancePercent30

MinimumThrottlePercent2

PerformanceImplementation3
---
NIC compliance error:

-Provider
[ Name] Microsoft-Windows-Kernel-Power
[ Guid] {331c3b3a-2005-44c2-ac5e-77220c37d6b4}

EventID172

Version0

Level4

Task203

Opcode0

Keywords0x8000000000000404

-TimeCreated
[ SystemTime] 2025-01-17T10:18:48.1551381Z

EventRecordID140480

Correlation

-Execution
[ ProcessID] 4
[ ThreadID] 424

ChannelSystem

ComputerPC-Computer

-Security
[ UserID] S-1-5-18

-EventData

State2

Reason6
 
You need to tell us your full system specs. Motherboard, GPU and so on. And when you say "crash" does it just shut down or reboot? If so it can also be the motherboard is going bad.
 
I was under the impression that 650W can easily handle i7s power usage
It can - as long as you don't then overload it with a huge graphics card, several drives, several fans, water cooling and more. So as Frick said, you need to tell us about your hardware. If this is your main computer, you can just fill out your TPU System Specs and reference that.

And when you say "crash" does it just shut down or reboot?
Note he said,
To clarify, it just shuts down completely - then reboots.

@Ekrell - since you have the same results with 2 power supplies, and assuming (for now) that 650W supply is big enough, then most likely the problem is somewhere else. It could be just about anything. Heat is also suspect since you said this happens when the system is tasked. So you need to make sure the interior is clean of heat trapping dust and that you have a good supply of cool air flowing through the case.

When you mounted the new CPU, did you thoroughly clean the cooler of old TIM (thermal interface material) and apply a fresh, new layer?
 
Motherboard: H110M PRO-D (Updated: 2018-02-27) [MSI]
GPU: NVIDIA GeForce GTX 1050
The operating system is Windows 10
12 GBs of RAM

I cleaned everything both when I installed the new CPU and when I installed the new PSU. I re-applied a new paste too. I suspected the CPU to be overheating initially, but the general area around it doesn't seem to be hot at all, even seconds after the crash when I checked physically with my hand. Unless that doesn't mean anything.

It reboots whenever the issue occurs.

If there's some way to tell if it's the CPU overheating, I'm open for ideas. I installed a diagnostic programs but I'm unable to get a clear result since I can't predict when the reboot occurs exactly.
 
@Ekrell You can try limiting the TDP in BIOS, see if that fixes your problem. Since you already replaced the PSU, it may be something on your mobo that went belly-up and doesn't supply enough juice to the CPU anymore.
And like other have said, please fill in your system specs.
 
what was your previous CPU? what is your motherboard & RAM settings? the i7-7700 non-K doesn't draw that much power by itself so what where you doing when "the power usage was high enough"? If it was gaming, what is your GPU?
 
My previous CPU was Intel Pentium
The crashes happen most often when I play modern games, the most recent notable one is Helldivers 2.

I updated my specs to whatever extent I could. I'm assuming the casing, mouse, and keyboard are irrelevant.
 
why do you think that it is the PSU?
 
My previous CPU was Intel Pentium
The crashes happen most often when I play modern games, the most recent notable one is Helldivers 2.

I updated my specs to whatever extent I could. I'm assuming the casing, mouse, and keyboard are irrelevant.
650W is plenty for that system. I still suspect the mobo, see my previous post.
 
I'm assuming the casing, mouse, and keyboard are irrelevant.
Mouse and keyboard are not important for this problem, but the case is a critical component in any computer system because the case is what defines (and limits) the cooling capabilities of the entire system. So it at least would be good to know your case fan configurations, and your temps.

I think you have an extra "1" in your motherboard model. I am assuming it should be H110M PRO-D.

12GB of RAM is an unusual amount. How is it configure? I assume 1 x 8GB plus 1 x 4GB. Try running with one stick at a time. I would start with the 8GB stick.
 
I'd start with the usual
  • Update BIOS
  • Run Windows Memory Diagnostic
 
I suspect capacitors on motherboard has reached it's end of life, granted that new PSU is not second hand. I had same issue with running R9 290X on Chinese 1000W PSU. So, i have to remove both of them and install old cooler master PSU that was sitting around. My issue was related to PSU. However, i did played games and ran blender renderings on dedicated GPU with full load but never ever witnessed rebooting issue ever again. As per thermal goes my system runs pretty hot as ambient here is around 34 Celsius with humidity ranging from 60% - 90%.
Note, that all components on my system were brand new when i purchased them.
 
Run Windows Memory Diagnostic
but RAM was fine previously

if it was my PC under theses circumstances, I would agree with @bug and look at the mobo. Maybe try to undervolt the CPU
 
And they updated the BIOS (missed that on first reading)
 
I suspect capacitors on motherboard has reached it's end of life
I've re-capped a number of much older mobos (circa 2010) with swollen electrolytics, but the H110M PRO-D has solid capacitors, apart from four "gold" audio electrolytics. I have yet to find a mobo in my collection where its solid capacitors have gone bad and need replacing. Perhaps it's only a matter of time.

1024.png

Old style "wet" electrolytics, good and bad.

iu



12GB of RAM is an unusual amount. How is it configure? I assume 1 x 8GB plus 1 x 4GB. Try running with one stick at a time. I would start with the 8GB stick.
Agreed. Mismatched RAM is not a recipe for stability. Run a full MemTest86 on each stick separately (several hours). Any errors, ditch the RAM and buy a new kit 2 x 8GB.
https://www.memtest86.com/

You could run SFC and DISM to check your OS, but I doubt it'll make any difference.
https://www.howtogeek.com/222532/ho...-system-files-with-the-sfc-and-dism-commands/

When I get desperate, I grab a cheap 128GB SATA SSD, disconnect all existing drives and perform a fresh Windows install. It gets rid of any nagging doubts about possible OS corruption. If the fault persists, it's your hardware or the wrong BIOS setting. I assume you're not trying to overclock your mismatched "pair" of RAM (8GB + 4GB) using XMP?
 
I never tried overclocking anything. The thing about the RAM and Motherboard is that those issues started occurring after I got the new CPU. If it was a fault of the components I already had in, wouldn't the crashes appear earlier?

Also, I'm attaching all the settings I could think of from my BIOS. I do not see anything related to PSU or CPU beyond those. If anyone has any recommendations as to how to configure that, shoot.
 

Attachments

  • 4.jpg
    4.jpg
    146.5 KB · Views: 59
  • 3.jpg
    3.jpg
    120.9 KB · Views: 62
  • 2.jpg
    2.jpg
    150.9 KB · Views: 63
  • 1.jpg
    1.jpg
    119.7 KB · Views: 70
I never tried overclocking anything. The thing about the RAM and Motherboard is that those issues started occurring after I got the new CPU. If it was a fault of the components I already had in, wouldn't the crashes appear earlier?

Also, I'm attaching all the settings I could think of from my BIOS. I do not see anything related to PSU or CPU beyond those. If anyone has any recommendations as to how to configure that, shoot.
Don't be so dismissive, people are using their experience to try to help you.
"Worked before" tells very little. It's possible your setup could handle the needs of the old CPU, but can't handle the needs of the new one. Could be a weak component got pushed over the edge by the additional demands. Maybe you shorted something out when handling the motherboard. And that's just otoh.
 
I never tried overclocking anything.
The i7-7700 does not have a 'K' suffix (i.e. i7-7700K) so that rules out normal overclocking available on K-series processors that have an unlocked multiplier. Most people who want an easy time overclocking use a K-series Intel CPU, without this limitation. I'd forget any notion if overclocking a non-K CPU unless you want to complicate matters. Your non-K i7-7700 will automatically boost the frequency when appropriate. CPU-World indicates the maximum operating temperature for the i7-7700 is 100°C.
https://www.cpu-world.com/CPUs/Core_i7/Intel-Core i7 i7-7700.html

The i7-7700 has a nominal TDP rating of only 65W, which makes it easy to cool. You are unlikely to see circa 65W dissipation outside benchmarking programs such as Prime 95. Most of the time I'd expect your CPU to be dissipating significantly lower powers (a few tens of Watts).

If there's some way to tell if it's the CPU overheating,
To monitor the exact power being consumed by your CPU and measure its temperature, try Aida64 or HWMonitor. Other utilities exist.
https://www.cpuid.com/softwares/hwmonitor.html
https://www.aida64.com/downloads

I suspected the CPU to be overheating initially, but the general area around it doesn't seem to be hot at all
You're unlikely to detect high temperatures in the vicinity of a CPU, even when the internal Silicon die is running at 80 to 100°C. When I touch the heatpipes on an air cooled CPU running at well over 100W, they feel barely warm. It's the sign of a good cooler. If you were to remove the heatsink and touch the bare top of the CPU, you'd burn your finger, just before the thermal trip killed power.

I do not see anything related to PSU or CPU beyond those.
I have never seen BIOS settings relating to a PSU. Most ATX PSUs do not have control features, so there's no way for the BIOS to "adjust" the PSU's output. The PSU is a self-contained unit. It just generates power up to the point where built-in overload functions trip when a fault occurs.

As for CPU control, some BIOS have options to change maximum power/current limits on CPUs, but I leave these well alone. Similarly, you can adjust the voltage fed to the CPU, but if you get it wrong and apply too much voltage, the CPU dies. The standard CPU current/power limits do a pretty good job, unless you're going for an extreme overclock or using Liquid Nitrogen cooling, where you need more current than the standard (safe) limits allow.

The BIOS settings you've posted are merely those for fan control. I suggest enabling the warnings that inform you when the CPU fan stops working. A big air cooler heatsink will continue to protect your CPU if the fan fails, but it's a good idea to replace a faulty fan as soon as possible.

My previous CPU was Intel Pentium
The name Pentium covers hundreds of CPUs dating back decades, although most of them won't fit your motherboard. Can you tell us which model of Pentium you were using?
https://cpu-socket.com/cpu-list-lga-1151/

12 GBs of RAM
As already mentioned, mismatched RAM can cause instability. Until you check your RAM by booting from USB with MemTest86, we can't rule this out as the culprit. Try a matched pair, e.g. 2 x 8GB.
https://www.memtest86.com/

The thing about the RAM and Motherboard is that those issues started occurring after I got the new CPU. If it was a fault of the components I already had in, wouldn't the crashes appear earlier?
Your old Pentium might have been running your RAM at a lower frequency than your new i7-7700.

Some LGA1151 Pentiums operate down as low as DDR4-1866. The i7-7700 is rated up to DDR4-2400. If your Pentium was running at 1866MT/s and your i7-7700 is running at 2400MT/s on RAM rated a 2133MT/s, there's an obvious difference.

Are you overclocking your RAM with XMP settings. Unless you're using a matched pair of identical DIMMs, this can cause instability. Disable XMP.
 
I never tried overclocking anything. The thing about the RAM and Motherboard is that those issues started occurring after I got the new CPU. If it was a fault of the components I already had in, wouldn't the crashes appear earlier?

My tenth generation i5 runs RAM at a slower speed than an i7 can.
 
Clean your case, fans, heatsink, apply fresh thermal paste to cpu. Also reinsert all atx power connectors into every component in the case.

Last but not least remove the gpu and use igp and see if that changes anything
 
Motherboard: H110M PRO-D (Updated: 2018-02-27) [MSI]
GPU: NVIDIA GeForce GTX 1050
The operating system is Windows 10
12 GBs of RAM

I cleaned everything both when I installed the new CPU and when I installed the new PSU. I re-applied a new paste too. I suspected the CPU to be overheating initially, but the general area around it doesn't seem to be hot at all, even seconds after the crash when I checked physically with my hand. Unless that doesn't mean anything.

It reboots whenever the issue occurs.

If there's some way to tell if it's the CPU overheating, I'm open for ideas. I installed a diagnostic programs but I'm unable to get a clear result since I can't predict when the reboot occurs exactly.
Whats the deal?
 
Check video card wires plug,
 
My tenth generation i5 runs RAM at a slower speed than an i7 can.

However, this difference is not true for the 7th generation.
 
Last edited:
@Ekrell
did you reset bios after cpu switch and loaded defaults?
 
ok. my 2 cents.
you have new cpu. it's probably a "used" one. maybe, it's instability is going from well-known "internal thermal paste gate", related to that Intel has used poop-paste under the 7-th gen cpus. I bought i7-7700 brand new and after realized what sh*t I've bought, I've sold it and got new 6700T. Yeah, lower power, but SUPER COOL (you could put AIO on i7 7th gen and still it won't be effectively cooled unless "scalped", but I am against this BS better get proper CPU without issues, cmon, it's not C2D era when we have experimented the "scalping" lol).

and come on people, I've used a BUNCH of RAM MIX CONFIGS like 8+4 or 16+8 etc., and DDR4 is FKN STABLE and won't have such issues like "ram conflicts". Conflicts could persist with DDR4 on INTEL systems only from cr*ppy MOBOS. But, OP's "common" H110M PRO-D is SUPER STABLE, hence because it's SUPER BASIC.

Also, try to reinstall Windows from scratch (a reinstall from USB, not any "cloning" or "restoring image" whatever BS), and see if the BSODs will be again.
 
Back
Top