guide · static export
Hardware

Diagnose a PC That Restarts or Shuts Down by Itself

A machine that dies without a blue screen is usually losing power, not crashing. The event log, the pattern and four swaps tell you which part is at fault.

by@guidox updated:10/31/26 time:45 min level:Intermediate tested on:Windows 11 24H2 works on:Windows 10, Windows 11 tools:Event Viewer, Spare PSU optional
TL;DR

Note **when** it dies, read the **Event Viewer** for a Kernel-Power 41 entry, turn off **automatic restart** to see the bug check, then test **power delivery, cooling and RAM** in that order.

Before you start

Requirements

  • Event Viewer
  • A known-good power supply if possible
  • An hour
Step by step

Instructions

01

Write down the pattern

Under load, at idle, after ten minutes, only in one game, only after a cold start. A shutdown under load points at power or cooling; a shutdown at idle points at firmware or a failing board.

02

Read the event log before you open the case

Kernel-Power 41 means the machine lost power without a clean shutdown. It is a symptom, not a diagnosis, and the entry above it in the list often names the real cause.

eventvwr.msc > Windows Logs > System > Filter: Critical
03

Turn off the automatic restart

With the restart disabled, a real crash stops on the blue screen and shows a bug check code. That code separates a driver problem from a hardware fault in one reboot.

System Properties > Advanced > Startup and Recovery > uncheck Automatically restart
04

Suspect the power path first

Separate PCIe cables instead of one daisy chain, a different wall socket, and no cheap extension lead. A power supply that trips its protection cuts the machine with no log entry at all.

⚠ warningReusing a ten year old unit with a new graphics card is the most common cause of exactly this symptom.
05

Stress one subsystem at a time

Ten minutes of CPU load, then ten of GPU load, then both together. Whichever run reproduces the shutdown narrows the fault to that part or to the power that feeds it.

tipIf only the combined run fails, the power supply is the first thing to swap, not the graphics card.
06

Strip it back to the minimum

One RAM stick, no discrete GPU if the CPU has integrated graphics, one drive. A machine that stays stable in that configuration is a parts problem, and adding them back finds it.

Troubleshooting

Common issues

It dies instantly with no log entry

Power. Either the supply trips or the board loses standby voltage. Swap the unit before anything else.

It restarts only in one game

A transient load spike, or a driver resetting the GPU. Check the graphics driver and lower the power limit as a test.

It shuts down at idle, not under load

A firmware power state or a memory profile that is unstable at low load. Load the default firmware settings and retest.

A blue screen appears after the change

Now you have a code. Look it up before you change hardware again; it may be a single driver.

FAQ

Frequently asked

Is Kernel-Power 41 the cause?

No. It records that the previous shutdown was not clean. The cause sits above it or outside the log.

Can a bad wall socket do this?

Yes, and a failing extension lead with a switch does it more often than the socket itself.

How to undo this

Re-enable **Automatically restart** once the machine is stable again and reinstall any part you removed. Nothing on the disk is modified by the diagnosis.

Changelog · 1 entry
  1. 2026-10-31
    First version
Interactive version: checklist, feedback, related guides →

Free · no account needed · tracks your progress