After the power supply fails, the system restarts repeatedly.
After the power supply fails, the system restarts repeatedly.
hello. 3 days ago my old psu (antec 900 watt bronze) blew up. after opening it up i noticed a mosfet has shorted the high voltage line to ground explaining the smoke and bang. i installed a old corsair 650 watt unit for the time being and after some stressfull 15 minutes of plugging it in the system stil posted... it seemed fine at first glance but atm i'm experiencing weird problems. every so often the system would just randomly shut off and reboot. then after full boot shut off again. then boot perfectly fine again. f or some reason the os logs blames "google chrome Helper" for this.. but the system shuts down so fast it seems not os related (instant all screens go black and 5 secs later bios splash arrives) i have loaded the cpu and gpu with a high load to see if it shuts down under load which did not happen. so i left it for the time being thinking it was a random event and gamed while watching youtube. about 2 hours later it just randomly did it again making me lose 2 hours of work designing a vehicle. i checked the bios and my bios apparently does not have a logging section for random shutdowns. my motherboard is a msi z270 gaming m7 with a core i7 7700k 32 gb corsair whatever model that is a gtx 1070 a bunch of leds and like 6 hdd's the os is installed on a 256 gb nvme ssd. could this possibly be board damage? according to the internet 650 watts should be enough for this rig but i feel like its close to the psu limit. (especially at 5ghz and the planned 5.6 ghz after custom cooling) and if a small power dip accours the psu cannot compensate and handle it thus shuts off? when it crashed i held my hand behind the psu to feel if there happens to be too much heat but the psu is cool and the fan is stil in stop mode. (the psu is corsair 650 model from 2017.. not sure wich exact model it is but its fully modular) thermals for the cpu would not be the case as i'm using a noctua nh d15 with a delidded and liquid metalled cpu keeping it in moderate range with custom fan profiles. any tips/recommendations to what i could try? thank you in advance please mind my grammer. its 4:38 am right now and i should've been asleep 4 hours ago..
High processor or VRM temperatures may lead to shutdowns. (Assuming no blue screen, it simply powers off if that’s the case) Or the power supply could have damaged the motherboard during a failure. This is something that occurs, which would be undesirable. If the problem stems from an operating system, program, or app issue, you might try using a separate partition or drive and reinstalling the OS. Since Chrome helper appears in logs, removing this program could help. Usually you see a blue screen first, but I didn’t explicitly discuss that here, so I focused on the above points.
Tomorrow I plan to install Windows 10 on a spare drive and run additional benchmarks to see how it performs. Maybe with some luck, the motherboard could still be under warranty. I still have the CPU cover in case there’s damage to the board or CPU. I checked the BIOS and saw voltage surge protection; it looks like MSI disabled it by default. That’s frustrating. My breaker tripped instantly, so the PSU must have had very little time to supply power.
I'm really concerned about these old Antec power supplies I've been using for years. They're around 10 years and their warranties are over. Most shutdowns seem to be hardware issues, but a fresh OS might help with diagnostics. I'm hoping you'll be okay. Make sure to check the VRM temperatures carefully. Using HWinfo64 will give you all the info you need for monitoring. You can run it alone or alongside any other monitor software. It would be helpful to know the specs of your PSU, especially since yours is a Corsair model. A 650-watt unit should suffice, but if not, it might overheat. Avoid running the system under heavy load and feel how hot it gets on the back side. If it's too warm, it could be throttling and shutting down.
this the model incase ur wondering and the explody part. so far i reinstalled windows and the problems are gone. only one is intel gpu keeps dying.. just keep getting BSOD every 2 to 3 hours on random and the code is directly linked to intels igpu drivers (i run 1 screen of the igpu)
Yeah, I still have it. I use the wires for projects, especially adding PCIe power to systems without any issues. It's really useful because it has a wide range of connectors. The 4x8 pins on the 2x8 strands cover many things like SATA and Molex. It’s sad it stopped working but it was great!