My PC has an ASRock H510M-HDV/M.2 motherboard, a 256GB KLEVV C710 NVMe SSD that is about four to five years old, a 480GB KLEVV N400 SATA SSD that is around three years old, an Antec NeoECO 550W Bronze PSU, and a 1TB hard drive.
After the computer had been running for about four hours, I played Modern Warfare for roughly 40 minutes. The system then froze, and after waiting several minutes the audio changed from a loop into a buzzing sound. I powered it off, but on reboot the BIOS could not detect the NVMe drive. Turning off the PSU and reseating the drive did not immediately help.
At a repair shop, the computer eventually reached the Windows login screen but then froze and reported that Windows was corrupted. Another shop reinstalled Windows, but the system initially produced a BSOD before later working normally.
The drive-health results are confusing. CrystalDiskInfo reports 97% health for the SATA SSD, while Hard Disk Sentinel reports 24% health, 40 bad sectors, and 16TB written. The NVMe drive reports around 89% in both programs. Windows was installed on the NVMe drive, even though the SATA drive appears to have more serious errors according to Hard Disk Sentinel.
The NVMe drive has sometimes disappeared from the BIOS after the crash, so I am unsure whether the problem is the SSD, the motherboard's M.2 slot, or possibly an electrical or power issue. I currently need at least 512GB of reliable SSD storage for Windows and games.
Should I continue using the NVMe drive and reserve the SATA SSD for replaceable game files, replace the NVMe with a new 512GB model, replace both drives, or investigate the motherboard and PSU first?
4 Answers
I would not keep the current NVMe as the Windows drive just because it reports 89% health. The crash, corrupted installation, and failure to appear in the BIOS are enough reason to treat it as unreliable. Install Windows on a new SSD and keep the old NVMe disconnected or use it only temporarily for testing. The SATA SSD can be used for downloads or replaceable games, but its bad-sector count means it may fail or corrupt files without much warning.
Before buying a motherboard, fully remove power from the computer, take the M.2 drive out, inspect the connector and screw, and reinstall it carefully. A poorly seated M.2 drive can disappear intermittently. If possible, test the NVMe in another computer or use a known-good drive in this motherboard's M.2 slot. That will distinguish a bad drive from a bad slot.
The SATA drive should not be trusted for important data. CrystalDiskInfo mainly displays the values reported by the drive, while Hard Disk Sentinel applies its own health calculation and can reduce the rating when it detects bad sectors or other problems. A 24% rating with 40 bad sectors is a serious warning, even if another program says 97%.
The NVMe percentage is mostly a wear indicator and does not prove that the drive is healthy. An SSD disappearing from the BIOS after a crash is especially concerning. Back up anything important immediately, then replace the questionable drives rather than using their reported percentages as a guarantee. Also check whether the NVMe has a firmware update and monitor its temperature.
A failing PSU is possible, but it is not the first conclusion from these symptoms. A bad M.2 drive, poor connection, overheating, firmware problem, or motherboard slot can all cause the drive to vanish. Check the motherboard's power connections and temperatures, update the SSD and motherboard firmware if available, and test with a known-good SSD. If multiple drives fail in the same slot or the system shows other unexplained instability, then investigate the motherboard and PSU further.
That makes sense. I only reseated it once while the system was still being investigated, so I will test it properly with the power cable removed and compare it with another drive if possible.