A PC that shuts off without warning — mid-game, mid-browsing, or seemingly at random — is one of the most frustrating faults to chase down, because it can stem from overheating, a failing power supply, faulty drivers, or even a Windows setting. This guide gives you a structured process to narrow down the cause, starting with the checks that take minutes and moving through to the more involved hardware tests.

Common Causes at a Glance

  • Overheating — a CPU or GPU hitting its thermal limit will trigger a protective shutdown.
  • Failing or underpowered PSU — unable to maintain stable voltage under load, especially during demanding tasks.
  • Faulty RAM — can cause both crashes to a blue screen and, less commonly, sudden power-offs.
  • Windows power settings — fast startup, sleep, or an overly aggressive power plan misbehaving.
  • Driver issues, particularly graphics drivers, causing a crash that looks like a shutdown.
  • A failing surge protector or extension lead cutting power intermittently.
  • Dust and poor airflow — indirectly causing the overheating issue above.

Safety warning: If you smell burning, see smoke, or notice the PC is unusually hot to the touch on the case exterior, stop using it immediately, unplug it, and do not attempt to power it on again until it has been inspected — this can indicate a serious electrical fault.

Step 1: Note the Pattern

Before diagnosing anything, spend a day or two paying attention to when shutdowns happen — the pattern narrows the likely cause considerably.

  • Only during gaming or heavy tasks (video export, benchmarking): Points strongly towards heat or PSU load issues.
  • Completely random, even at idle: More likely to be a Windows setting, a driver conflict, or a faulty power connection/PSU issue not related to load.
  • Always after a specific amount of time (e.g. always after 20 minutes): Strongly suggests a thermal issue, as components take time to heat up to their limit.
  • Immediately on waking from sleep: Points to a Windows power management or driver issue rather than hardware.

Step 2: Check Temperatures

  1. Download a monitoring tool such as HWMonitor or HWiNFO, which display real-time CPU and GPU temperatures alongside fan speeds.
  2. Run the PC under the same conditions that trigger the shutdown (e.g. play the game that crashes) while watching temperatures.
  3. Compare readings against typical safe limits: most modern CPUs throttle or shut down in the 95–105°C range, and most GPUs in the 90–95°C range, though exact limits vary by model — check your specific CPU/GPU's documented maximum.
  4. If temperatures climb rapidly and shutdown coincides with hitting a very high figure, heat is very likely your cause.

If overheating is confirmed

  1. Clean out dust. Use a can of compressed air (holding fan blades still to avoid over-spinning them) to clear dust from heatsinks, fans, and case filters. This alone resolves a large proportion of heat-related shutdown cases.
  2. Check case airflow. Ensure intake and exhaust fans are working and not blocked, and that cable clutter isn't restricting airflow inside the case.
  3. Reapply thermal paste if the CPU cooler hasn't been serviced in several years — dried-out paste significantly reduces heat transfer.
  4. Check cooler mounting. A cooler that's come loose or was fitted with uneven pressure won't make full contact with the CPU.
  5. Consider ambient conditions. A PC in a poorly ventilated cupboard or exposed to direct summer sun will run hotter than one in open air.

Step 3: Rule Out the Power Supply

A PSU that can no longer maintain stable output under load is one of the most common causes of shutdowns during demanding tasks, particularly as PSUs age or if the unit was undersized for the components it powers.

  1. Check your PSU's wattage against your components' combined power draw, particularly if you've upgraded the graphics card or CPU since the PSU was bought without reassessing capacity.
  2. Test with a PSU tester (an inexpensive tool that checks each voltage rail is within tolerance) or, if available, swap in a known-good spare PSU temporarily to see if shutdowns stop.
  3. Inspect cables for any damage, and ensure all power connectors (24-pin, CPU 4/8-pin, GPU power) are fully seated — a marginal connection can fail specifically under load when current draw increases.
  4. Check the wall connection. Try a different socket and a different power cable, and avoid running the PC from an overloaded extension lead shared with other high-draw appliances.

Pro tip: If shutdowns coincide specifically with GPU-intensive moments (loading a new game area, a demanding cutscene), this is a strong hint towards a PSU that's struggling with a sudden power draw spike, known as a transient load, even if its average wattage rating looks sufficient on paper.

Step 4: Check for Faulty RAM

  1. Run Windows Memory Diagnostic by searching for it in the Start menu, or use the more thorough third-party tool MemTest86, booted from a USB drive, left running for several passes (ideally overnight).
  2. Errors reported by either tool point to a faulty module — identify which stick by testing them individually in the same slot.

Step 5: Rule Out Software and Driver Causes

  1. Check Windows power settings. Go to Settings > System > Power & battery > Screen and sleep, and also check Control Panel > Power Options > Change plan settings > Change advanced power settings for anything unusual, such as an aggressively low processor power state or a misconfigured sleep timer.
  2. Disable Fast Startup, which can occasionally cause issues resembling shutdowns on wake: Control Panel > Power Options > Choose what the power buttons do > Change settings that are currently unavailable, then untick "Turn on fast startup."
  3. Update or roll back graphics drivers. A crashed graphics driver can sometimes present as a full shutdown rather than a simple crash-and-recover. Update via the manufacturer's site, or roll back to a previous version if the issue started after a recent update.
  4. Check Event Viewer for clues. Search for and open Event Viewer, then check Windows Logs > System around the time of the last shutdown for critical errors — particularly Kernel-Power Event ID 41, which indicates the PC lost power unexpectedly rather than shutting down cleanly, pointing back towards a hardware or power cause rather than software.

Troubleshooting Branches

  • Shuts down only under GPU load, temperatures look fine: Suspect the PSU rather than heat — test with a PSU tester or known-good spare.
  • Shuts down at idle with no clear pattern: Check Event Viewer for Kernel-Power errors, disable Fast Startup, and check for a loose power cable.
  • Shuts down consistently after a fixed time regardless of task: Strongly suggests heat — clean the system and check cooler mounting.
  • MemTest86 reports errors: Replace or reseat the faulty RAM module identified.

When to Call a Professional

If you've cleaned the system, confirmed temperatures are within normal limits, tested with a different PSU or PSU tester, and run a full RAM test with no errors found, but shutdowns persist, the issue may lie with the motherboard itself or a less common component fault that needs specialist diagnostic equipment to isolate. A professional can also help if you're not confident opening the case or handling components like RAM and cabling yourself.

Was this guide helpful? Explore more Power & Boot Issues guides, or browse all Desktop PC Repair articles.