Problem with hard crashes – screen stuck, no reply, very high VBAT heat
Problem with hard crashes – screen stuck, no reply, very high VBAT heat
Hello!
Over the last few days, I've experienced multiple hard crashes. Initially just one, then another, and then several in quick succession. They occur without warning, even under light load, about 30 seconds after the PC starts up with no activity. I've seen three BSODs, all different types—kmod exception not handled, IRQL not less, and one more that I didn't notice.
I've reinstalled the GPU driver 20 times through various methods: Nvidia GeForce Experience, the latest official version, and the December release. I also used DDU twice for a clean setup. The memory diagnostic tool showed no issues. The PC runs smoothly under load for long periods and can crash even when Discord is open.
No GPU acceleration is enabled in any program. Windows self-diagnostics say everything is fine. I've disassembled the whole system, cleaned out dust, checked cables, re-seated RAM, SSDs, etc. Now, using "open hardware monitor," the temperature peaks at 120°C (rising from 28 to 60 to over 100°C in seconds), though the battery itself feels cool.
The voltage readings are normal—3.168V—and other components show sub-40°C temperatures. I haven't used this monitoring tool before, so I'm unsure what it was doing before. In Reliability Monitor and Event Viewer, nothing unusual stands out. It simply reports an unexpected shutdown as if it happened without me.
Could there be any cause for concern—hardware, software, drivers, or something else?
For reference, I haven't updated anything recently, changed parts, or used the GPU for extended periods. The GPU is three years old, and other components are two years old. This has never happened before.
Since reassembly, it's been working perfectly for about 45 minutes now.
Details of specs:
- Asus PRIME Z690-P
- MSI 3060 Ventus 2X
- i5 13600K
- G.Skill 32GB 5600J2834F16G
- 750W Corsair PSU (functional)
- Windows 10 22H2
Temperature 5 equals 115 degrees. It might be useful to discuss its significance.
Can't reference the temperatures; I understand that certain board parts can become extremely warm, but those readings seem too high. In the meantime, a straightforward action you can take is perform a memory test with memtest86 (or even the standard Windows built-in tool does a satisfactory job, which is a bit surprising). It's always wise to run one when you begin noticing BSODs, as it clearly indicates problems and suggests replacing the RAM.
I ran it successfully, no errors encountered.
A minor update: I experienced more crashes when no GPU driver was installed.
I attempted a Windows reinstall, but my PC crashed during the process at 1%, and it didn’t even boot.
Regarding temperature, as mentioned, the VBAT (Voltage-Controlled Battery) is supposed to be around 5 volts, which is the small CR2032 battery. I suspect this sensor might be faulty, as it rapidly fluctuates between 25 and 120 degrees Celsius. It seems physically unlikely for such unstable readings to occur in a PC.
It appears the MOBO could be the source of the problem.
If the RAM is functioning well yet you're experiencing crashes without a GPU driver or during a Windows installation, it suggests:
A failing GPU
Motherboard issues
CPU problems
The latter is uncommon, but if the GPU were malfunctioning without a driver, other components would likely be affected. I’d lean toward the motherboard.
Keep in mind that not all temperature sensors actually measure heat, but they provide voltage/current data that should correlate with temperature. Excessive fluctuations could point to a defective sensor or a more serious problem.
Currently, I’d assume the motherboard is the culprit, but it might be easier to verify by swapping with another GPU for testing.
Hello!
After checking each part separately, it seems my CPU stopped working. I didn’t overclock it, didn’t make any changes, and followed the updates from the ASUS center. It’s unclear if this is due to typical problems of that generation or just a random issue.
If anyone has encountered this problem with similar specs, try the CPU first if possible.
The technician at the service shop went through everything until it finally failed on the CPU.
In Mr. gamerk316’s comments, something interesting happened during repair: with my 3060 and a standard Windows 11 Pro setup, the installation froze at 1% after just one second. Even with a weaker test GPU, it failed before the first boot. Switching to generic-looking DDR5 RAM didn’t help much.
We also tested it on a brand new MSI Z790 Tomahawk with all other parts replaced, but the same issue occurred during Windows installation.
Additionally, the VBAT temperature stayed steady at 3.168V without any change. It probably points to the processor itself.
I’ll update once my new CPU is installed. Everything else looks fine.
Thanks for your advice!