How to identify the reason behind unexpected shutdowns?
How to identify the reason behind unexpected shutdowns?
The problem started without any obvious recent hardware modifications. The most recent action was swapping out an old HDD for a SK hynix drive, and the issue emerged weeks later. The computer shuts down completely during heavy use—such as when playing BF6 beta or 2042—and maintains a stable frame rate around 30 frames per second. Upon crashing, the case fans, CPU cooler, and GPU lights turn off, while the motherboard and RAM LEDs stay lit, along with the power button light. This suggests the problem isn’t related to the power supply unit.
System details:
- Motherboard: ASUS TUF-GAMING Z590-PLUS
- CPU: Intel i7-11700k (no overclocking)
- RAM: Corsair Vengeance Pro RGB 2x16GB (XMP enabled)
- GPU: MSI GeForce RTX 3070Ti Gaming X Trio 8GB (no overclocking)
- NVMe: Samsung 980, Intel 665p Series, SK hynix BC501
- SATA: Samsung 870, Crucial CT240M, Sandisk External USB-C
- PSU: Corsair CX750M
Diagnostic results from EventViewer:
Code:
ACPI thermal zone _TZ.TZ00 has been enumerated.
_PSV = 0K
_TC1 = 0
_TC2 = 0
_TSP = 0ms
_AC0 = 373K
_AC1 = 328K
_AC2 = 323K
_AC3 = 318K
_AC4 = 313K
_AC5 = 0K
_AC6 = 0K
_AC7 = 0K
_AC8 = 0K
_AC9 = 0K
_CRT = 378K
_HOT = 0K
minimum throttle = 0
_CR3 = 0K
I received a HWInfo64 log showing temperatures before the crash, but nothing critical appears. I ran Memtest64—clean results. Tried a different power outlet, ensured all cables are properly connected, and even performed a slight undervolting on the GPU.
Given my limited budget for repairs, I can’t just guess and buy replacement parts. Based on what I have, I suspect the motherboard might be the issue.
Welcome to the forums, newcomer!
PSU: Corsair CX750M
Which colored label do you have—green or grey? The PSU is still available in those variants. How old is the PSU?
GPU: MSI GeForce RTX 3070Ti Gaming X Trio 8GB (no OC)
The RTX 3000 series was known for its high transient load spikes. Users were recommended to overprovision their PSU’s capacity, especially outside of purchasing well-built units, to handle the problem. This led people to consider an 850W unit for that card.
I don’t think the issue lies with the PSU.
+
The PC will shut down under heavy load.
Source (borrowed, not bought) a higher wattage, reliably built unit from a friend or neighbor and check if the problem continues.
You didn’t mention your case’s make and model. How is your processor being cooled?
Grey label, reaching up to around 4-5 years old on the PSU.
CPU cooling uses a 3-fan radiator AIO from Corsair, though I don't recall the exact model or have the documentation; it's working well. The case is a Corsair 5000x Airflow, featuring 10 total fans—6 intake and 4 exhaust.
For the GPU, I haven't experienced any issues, but I did notice 300W+ spikes on HWInfo before the crash.
I plan to acquire a PSU and share an update later.
Grey label, reaching up to around 4-5 years old on the PSU.
Is it 4 or 5? If it's 5, then your PSU is nearing the end of its warranty, which is listed as 5 years;
https://www.corsair.com/us/en/p/psu...
If you had a higher wattage unit, you could keep the PSU longer.
Hey there,
I’d only recommend using a CX PSU in an office setup. They really don’t perform well. I wouldn’t choose them for a gaming rig. EDIT: The CV is the main issue, the CX is just average.
The issues you’re seeing are likely due to the PSU. Consider replacing it with something better, such as an RMx.
A 750w or 850w would work fine.
Yeah, I'm not entirely confident the PSU is the issue, but upgrading it could be a smart move for this GPU. It might work unless the expense is too high for you. If it solves the problem, that's great; if not, you'll have a better PSU to rely on for years. You don't need an expensive one—just something functional.
I don't have access to the specific BIOS version of the motherboard. You can check this by consulting the manufacturer's documentation or using the system settings.
I'm considering upgrading my Corsair RM850e with an ACPI thermal zone enumeration log prior to a crash in Event Viewer. Are you certain this points to the power supply unit? I'm wondering if there might be another component involved, and I'd appreciate any suggestions on how to investigate further. If I had to estimate, it could be around five or more possibilities. I'll bring another unit for testing, and thank you for your help—you weren't aware PSUs were failing that quickly.