My Windows 11 system has an ASUS PRIME B760-PLUS D4 motherboard and is suffering from random shutdowns during idle or light use, even though it remains stable during demanding benchmarks. The screen goes black first, audio may continue briefly, and then the computer loses power. Afterward, the behavior varies: the power button may not respond until the AC power is disconnected, the system may loop at the ASUS logo, or it may fail to boot into Windows. I have tested a different PSU and GPU, tried the RAM sticks individually in multiple slots, updated the BIOS, cleared CMOS several times, and confirmed that temperatures are normal. Could a faulty motherboard power circuit, CPU socket, or connection be responsible, or is there another component—such as a storage drive—that I should test before replacing or RMAing the motherboard?
3 Answers
There is also a temperature-dependent pattern worth watching. After one update, the machine reportedly powered off within five seconds of startup and would not work again for roughly ten minutes. That kind of recovery delay could point to a protection circuit, board fault, or another component that only behaves normally after cooling or discharging. Try starting with only the CPU, one RAM module, system drive, and integrated graphics if available, while checking all motherboard and CPU power connections.
A failing SSD or NVMe drive can cause crashes, endless boot cycles, and strange behavior during startup. Try disconnecting every nonessential drive and booting with only the system drive attached. If you have more than one drive, test them separately. If the issue disappears, check the drive’s health or replace it. Otherwise, the motherboard becomes a more likely suspect.
The fact that it fails at idle but survives heavy load makes a simple overheating or load-related PSU problem less likely, especially after testing replacement parts. The motherboard’s power delivery, a CPU socket contact, or a marginal trace could still be responsible. If the board can be returned or RMAed, that is a reasonable next step. Testing the system while it is cold and failing may also help document the problem before it changes state.
I’ll disconnect and test the drives separately.