Due to the sheer number of components involved, troubleshooting a PC that keeps crashing can be intimidating. But it’s very doable if you know where to look.
- For instance, driver problems often cause programs to crash and generate an error message without shutting down the system.
- It’s the same with Blue Screens of Death (BSODs). The Stop Code can help you figure out what causes this exact bug and how to fix it.
- If some component is silently failing and you’re not sure where to begin, you can isolate and test them one by one to single out the faulty part.
In all cases, the key detail is that you should identify the source of the crashes first. Then, you can apply the appropriate steps and get rid of the crashes.
Identify the Cause of the Crashes
Ideally, you’ll have some kind of lead to start with, like an error message or a stop code.
- Look up the error code online to figure out what the problem is.
- With BSODs, you can check the Stop Code or analyze the dump files to find the cause.
- The Windows Event Viewer often logs useful info about crashes.
If you don’t have any leads, note whether the system only crashes to the desktop or shuts down and restarts entirely.
Programs crashing to the desktop are often due to driver problems or faulty components like the GPU or SSD.
If the system fully restarts after the crash, consider the following possibilities instead:
- Your components may be overclocked to the point of instability. Remove the overclock to check for this.
- Your components could be overheating. Pay attention to the fan speed or directly monitor the temps to verify this.
- Components like the PSU, CPU, RAM, motherboard, or cables could also be failing. Perform diagnostic tests to check for this.
- The last option is to temporarily swap components that you suspect.
Fixing RAM Problems
- Monitor and lower the RAM usage if required.
- Disable the XMP Profile if it is currently enabled. Remove any manual overclocks as well.
- Reseat the RAM modules and try different slots.
- Use the Memory Diagnostic Tool or any other method to test your RAM for errors. If your RAM modules turn out to be faulty, you’ll likely need to replace them.
Fixing GPU Problems
- Disable any overclocks on your GPU and ensure it’s not overheating.
- Ensure your GPU drivers are up to date.
- If the latest drivers don’t work well, uninstall them and try an older stable version instead.
- Try the GPU on a different PCIe slot to account for faulty slots.
- On that note, if you’ve vertically mounted your GPU, a faulty PCIe riser could also be the problem.
Fixing PSU Problems
If your PSU restarts, that means your entire PC loses power and restarts as well. Here’s what you should do in this case:
- First, ensure your PSU isn’t restarting due to overload. It’s best if your system’s power requirement doesn’t exceed 75% of the total PSU wattage.
- Ensure your PSU isn’t throttling due to overheating.
- Damaged power cables often lead to crashes. Swap your power cable to check for this.
- Also, ensure none of the connections are loose. This is especially important with modular PSUs as loose connections can easily lead to restarts.
- Finally, you can test your PSU with a Multimeter or PSU Tester. If the readings turn out to be unusual, swap the PSU for a while to check if the old one was faulty.
Solving Overheating Problems
If your fans sound like they’re about to take off before the PC crashes, you’ve probably got a heating problem. After passing a certain temperature, components like the CPU, GPU, and PSU either throttle performance or simply restart to prioritize component health.
On the other hand, if some of your fans are not working, that could also cause the components to overheat and lead to the same result.
- Clean dust from your PC. Focus especially on parts like the case fans, CPU fan, GPU, etc.
- Replace the thermal paste on components like the CPU and GPU.
- Disable any overclocks and try out undervolting instead.
- Ensure your case isn’t placed in a constricted spot that would restrict airflow.
Fixing Driver Issues & BSODs
On the software end, there are two main reasons for PC crashes.
A faulty graphics driver often causes certain programs to crash. Refer to the fixes from the GPU section in this case. Specifically,
- Disable any GPU overclock.
- If a GPU driver update is available, install the new version.
- If you’re having problems despite having the latest version, remove it instead. Then, manually install an older version that’s known to be stable.
Sometimes, the GPU driver or any other may lead to BSODs instead. In that case, use the following steps for troubleshooting:
- First, note the Stop Code. Look it up on our site or on Microsoft’s Bug Check Code Reference and try the suggested fixes.
- Use tools like WinDbg Preview to analyze the dump file. This can help you identify which driver generated the bug check.
- If required, use Driver Verifier to stress test your drivers. This will also help single out problematic drivers.
- Once you figure out which driver is causing the crashes, troubleshoot it. Update it, or roll back to find a version that works for you.
Fixing Storage-related Crashes
Problems with your storage drive (e.g., bad sectors or corrupt system files) can also lead to PC crashes. You should first use your device manufacturer’s tool (e.g., Samsung Magician) to perform a health check.
On Windows, dism, sfc, and chkdsk can help with storage-related crashes.
- DISM can repair a corrupt OS image.
- SFC can repair other corrupt system files.
- CHKDSK can repair logical bad sectors.
Troubleshooting All Components
If the crashes have persisted till this point, you have three more options.
Second, you could swap out hardware components that you suspect and replace them with known good ones. You’ll have to do this one by one for each part that may be faulty, from the CPU and RAM to the power cable and motherboard.
If both of these options seem like a hassle, you can also take your system to a repair professional. They should be able to diagnose the problem for you and suggest the best way to proceed.