System abruptly stopped. Insufficient power supply?
System abruptly stopped. Insufficient power supply?
Hello. My computer has been shutting down unexpectedly, stopping without a reboot. The operating system is Windows 11 Pro 64-bit, BIOS version 7B07v3E, running on an MSI A320m Pro VH Plus motherboard. The CPU is a Ryzen 5 2600 with the stock cooler. The GPU is a Zotac RTX 3060 Twin Edge OC. RAM includes two 8GB modules from Kingston. Storage consists of a 4TB Seagate HDD, a 1TB WD HDD, and a 120GB Sandisk SSD. Fans are installed front and rear: three Arctic P12 RGB fans daisy-chained to the JLED header on the board, connected via a 3-way splitter to the SYS_FAN header. The top has two Phanteks MG 140MM fans, also daisy-chained for the same purpose. The PSU is Seasonic M12ii EVO 620W (80 Plus Bronze).
This issue started three times recently, with the last occurrence two hours ago at 9 AM GMT+8. I recently purchased the RTX 3060 and previously used a 1660 Super without such problems. I performed a clean installation of the latest NVIDIA drivers.
During testing, I ran Furmark and OCCT for CPU and GPU stress tests for 15 minutes. Temperatures stayed low (91°C CPU, 75°C GPU) with no crashes. Benchmarks like Heaven and Valley showed similar temps and no issues. After running sfc/scannow, restarting, and performing dism restorehealth, the problem reappeared earlier.
System diagnostics indicated 100% on both HDDs, SSD at 73%, and no errors found with HDTune. Applications were running while I was troubleshooting: installing a game from a specific source, downloading torrents, and using a game installer. The 4TB HDD is being used, but I’m unsure what else to try. Possible causes include the fan configuration, power supply capacity, GPU load, Windows corruption, or needing a thermal paste refill and a dedicated fan hub during cleaning.
Unless the PSU's wattage rating doesn't reflect its real capacity... or sudden power spikes can disrupt it... I don't understand how that might occur. But I'm not the most knowledgeable about power supplies, only my general understanding.
It's a bit worrying, but the PSU wattage seems sufficient for the setup. The specific model appears to be group-regulated, which might limit its ability to handle sudden power spikes from 30 series devices. However, you do have plenty of headroom in terms of power consumption.
Initially, focus on the CPU temperatures which are connected to the VRM temps. On that board, this could cause the VRM protection to activate—leading to a hard shutdown—and then a slow CPU cooler fan might start after the next boot (common on older MSI AM3 boards, uncertain about AM4). Lower the CPU and VRM temperatures. In reality, upgrading the motherboard would be a better solution.
I assume I'll check if things get better this weekend after cleaning the PC and reapplying thermal paste. During gaming, CPU temperatures stay between 76-80°C. It's summer, so outside temps are also higher—about 29°C according to the forecasts. I have an AC, but I tend to avoid using it because electricity prices are high.
Updated BIOS and turned TPM back on in BIOS since it was disabled despite using Windows 11. Completed all downloads first before installing fg repack; installer limit was enabled. No crashes occurred. Will reapply thermal paste and install fan hub this weekend. Will share an update if it happens again or not. UPDATE 2: It happened again while I was reading manga, probably due to issues with SSD, motherboard, or PSU.
Updated 3: Experimented with reducing GPU voltage/clock speed. Ran benchmark five times, performed FURMARK for 35 minutes. No crashes occurred. System appeared stable. Attempted FG repack installation, but PC crashed afterward. Monitored temperatures with HWMonitor and GPU-Z; highest CPU reached 62°C, GPU 47°C. Likely power spikes from SATA cables are the issue. Swapped the 4TB HDD cable and ran dskchk /f /r before shutdown. Woke up PC still operational. Successfully reinstalled FG repack. Tried downloading another repack concurrently, no failures. Suspected storage cables caused problems. Replaced other drives and added backup supplies. Cleaned PC again and reapplied thermal paste. OCCT CPU stress tests now peak at 85°C, gaming temps 62-68°C (about 50+ to 60°C with AC).