My PC, which I bought around 2019 or 2020, worked reliably until the CPU cooler failed in March. I replaced it with a cooler from the original system builder, and since then the computer has randomly frozen and restarted. It can happen while I'm watching a movie, while the system is idle, or during normal work. Sometimes it runs for several days without an issue, but other times it freezes two to four times within 24 hours. I work from home, so reliability is important. I've already tried replacing the SSD, updating the BIOS to the 2025 F19 version, installing the latest chipset and other drivers, and running stress tests. The system builder checked the event logs and found multiple errors. They suggested trying a newer BIOS update, followed by sending the PC in for repair if that doesn't help. Before doing that, I'd like to know what else I should check—especially whether the new cooler was installed correctly, whether the protective film might still be attached, or whether memory settings such as XMP or EXPO could be causing instability.
3 Answers
Double-check that the protective plastic film was removed from the cooler’s contact plate before installation. Leaving it on can cause poor heat transfer and lead to shutdowns or instability. Also verify that the cooler is mounted evenly, the fan or pump is connected properly, and the CPU temperatures are normal under load. Since the problem started immediately after the cooler replacement, the installation and temperatures are worth checking first.
Run MemTest86 from a bootable USB drive to test the RAM outside of Windows. Let it complete several passes, ideally overnight. Memory errors can cause random freezes and restarts even when CPU stress tests look fine. If errors appear, test one memory stick at a time and try the default BIOS memory settings.
I haven’t used MemTest86 before, so I’ll look into creating the bootable test drive and running it for several passes.
Temporarily disable XMP or EXPO in the BIOS and run the memory at its standard default speed. These profiles overclock the RAM and can become unstable after hardware or BIOS changes. If the crashes stop with the profile disabled, update the BIOS, check that the memory voltage and speed are supported, and then either leave the profile off or use a slightly less aggressive memory setting.
Typical clues include freezes, application crashes, blue screens, or restarts that happen unpredictably, especially during memory-heavy tasks. The simplest test is to disable XMP or EXPO, save the BIOS settings, and use the computer normally for a while to see whether the behavior changes.
I’m fairly sure I removed the film, but I can’t remember with absolute certainty, so I’ll inspect the cooler and mounting. The inconsistent timing makes it harder to tell, since it sometimes works for days before freezing again.