Frequently Asked PC Hardware & Windows Diagnostic Questions
Verified solutions authored by pcfix Technical Diagnostic Laboratory.
Why is my motherboard showing a solid red or yellow DRAM debug LED with no display?
A solid DRAM debug LED indicates the motherboard UEFI failed memory initialization during Power-On Self-Test (POST). This is usually caused by unseated RAM sticks, oxidation on RAM gold pins, memory installed in incorrect DIMM channels, or unstable XMP/EXPO memory profiles.
Solution Checklist:
- Power off the power supply unit (PSU) and disconnect the AC wall power cord completely.
- Remove all RAM modules. Inspect the golden contacts for dust, debris, or oxidation, and clean them with 99% isopropyl alcohol or a clean eraser.
- For a standard two-stick configuration, seat the sticks into DIMM slots A2 and B2 (slots 2 and 4 away from the CPU socket). Press firmly until both end latches click securely.
- Clear the motherboard CMOS by shorting the CLRTC / JBAT1 pins with a screwdriver for 10 seconds or removing the CR2032 coin cell battery for 5 minutes.
- If the LED remains lit, test each RAM stick individually in slot A2 to isolate a defective stick or memory channel.
Citation: JEDEC Solid State Technology Association Standard JESD79-4 / JESD79-5 DDR Memory Specifications
Why does my PC turn on for a split second, spin the fans, and then immediately shut down?
A split-second power cutoff is triggered by the power supply or motherboard short-circuit protection (OVP/OCP/SCP). When a voltage rail detects a dead short or a missing 4/8-pin CPU 12V EPS connector, the PSU cuts power instantly to prevent silicon damage.
Solution Checklist:
- Inspect the motherboard standoffs inside the case to ensure no extra metal brass standoff is contacting the bare solder points on the backside of the motherboard.
- Verify that the dedicated 8-pin or 4+4 pin CPU EPS power cable is firmly clicked into the top-left motherboard header, not a PCIe graphics cable.
- Disconnect all non-essential peripherals: front panel USB 3.0 headers, RGB hubs, SATA drives, and case fans.
- Perform a breadboard test: place the motherboard on a non-conductive cardboard box outside the chassis with only the CPU, CPU cooler, 1 RAM stick, and PSU connected.
- Test the PSU independently with a paperclip bridge between Pin 16 (PS_ON green wire) and Pin 17 (Ground black wire) to confirm the fan continues spinning.
Citation: Intel ATX 3.0 / ATX 3.1 Power Supply Design Guide Specification
How do I fix the IRQL_NOT_LESS_OR_EQUAL (Bug Check 0x0A) Blue Screen of Death in Windows?
Bug Check 0x0000000A occurs when a kernel-mode process or hardware driver attempts to access pageable memory at an elevated Interrupt Request Level (IRQL). This is typically caused by a corrupted device driver, faulty system memory (RAM), or an unstable CPU voltage/overclock.
Solution Checklist:
- Boot Windows into Safe Mode with Networking to prevent third-party kernel drivers from initializing.
- Run the Windows System File Checker and DISM repair in an administrative terminal: DISM /Online /Cleanup-Image /RestoreHealth followed by sfc /scannow.
- Test system memory stability using Windows Memory Diagnostic (mdsched.exe) or bootable MemTest86 for at least 4 full passes.
- Roll back or cleanly reinstall the most recently updated driver (especially GPU, Wi-Fi, or chipset drivers) using Device Manager.
- Open Windows Minidump files located in C:\Windows\Minidump using WinDbg to inspect the exact .sys driver module listed under PROCESS_NAME and FAULTING_MODULE.
Citation: Microsoft Learn Windows Hardware Developer Documentation: Bug Check 0x0A IRQL_NOT_LESS_OR_EQUAL
What causes WHEA_UNCORRECTABLE_ERROR (Bug Check 0x124) and how do I resolve it?
WHEA_UNCORRECTABLE_ERROR is a hardware exception reported by the Windows Hardware Error Architecture. It indicates a fatal hardware bus timeout, CPU L1/L2/L3 cache parity error, defective NVMe SSD controller, or unstable CPU core voltage (Vcore).
Solution Checklist:
- Enter the UEFI/BIOS and reset all performance settings to Optimized Defaults to disable CPU curve optimizer, PBO, or manual overclocks.
- Check CPU core temperatures under load with HWInfo64; thermal throttling above 95°C can destabilize the internal CPU voltage regulator.
- Update the motherboard UEFI BIOS to the newest AGESA (AMD) or microcode (Intel) release, which patches known voltage transient drops.
- Inspect the Windows Event Viewer under System logs for Event ID 18 (WHEA-Logger) to identify whether the error stems from Processor Core, Cache Hierarchy, or PCIe Bus.
- Reseat the primary NVMe M.2 SSD in its slot and check drive SMART health parameters for unsafe shutdowns and media errors.
Citation: Microsoft Learn Windows Driver Kit Architecture: Windows Hardware Error Architecture (WHEA)
Why is my laptop or desktop CPU clock stuck at 0.79 GHz or 0.54 GHz?
A CPU locked at 0.79 GHz (799 MHz) is triggered by an asserted BD PROCHOT (Bi-Directional Processor Hot) flag. This sensor signal tells the CPU that a motherboard component (such as VRM MOSFETs or battery charger) is overheating, forcing the processor to its lowest power multiplier.
Solution Checklist:
- Check power delivery: on laptops, an aftermarket or failing AC adapter with incorrect wattage identification pin will trigger BD PROCHOT as a safety limiter.
- Clean laptop vents and heat pipe fins of compacted dust mats, and replace dried thermal paste on CPU and GPU dies.
- Disconnect the battery and hold the power button for 30 seconds to drain motherboard embedded controller (EC) capacitive charge, then power on with the AC adapter alone.
- Test with ThrottleStop on supported Intel platforms by unchecking the BD PROCHOT flag to verify if CPU frequencies immediately restore to full turbo clocks.
- Update motherboard BIOS/firmware to resolve errant sensor threshold tables that falsely report thermal over-limit states.
Citation: Intel 64 and IA-32 Architectures Software Developer's Manual Volume 3B: System Programming Guide
What should I do if the Wi-Fi icon completely disappeared from Windows 10 or 11?
A missing Wi-Fi icon means the operating system does not detect the network interface card in the device tree, or the WLAN AutoConfig service has crashed. This frequently happens after sleep states on Intel Wi-Fi 6 (AX200/AX201) and MediaTek wireless adapters.
Solution Checklist:
- Perform an electrical power cycle: shut down the PC, unplug the AC power cord, hold the chassis power button for 30 seconds to discharge motherboard capacitors, and reconnect power.
- Open Device Manager (devmgmt.msc), expand Network adapters, and check if the wireless card shows a yellow warning triangle with Code 10 or Code 43.
- Right-click the Wi-Fi adapter, select Uninstall device, leave the driver deletion checkbox unselected, and click Action -> Scan for hardware changes.
- Open an administrative Command Prompt and restart the WLAN service: net stop WlanSvc followed by net start WlanSvc.
- Execute network socket resets: netsh winsock reset && netsh int ip reset, followed by a system reboot.
Citation: Microsoft Windows Networking Architecture & WLAN AutoConfig Service Documentation
Why is my CPU temperature reaching 85°C to 100°C while idle or under light gaming?
Abnormally high idle temperatures are caused by a detached cooler mounting bracket, forgotten clear plastic peel film on the cooler base, dried or pumped-out thermal paste, or a dead AIO liquid cooler pump.
Solution Checklist:
- Touch both liquid cooling tubes if using an AIO: if one tube is hot and the other is cold while the pump reads 0 RPM in BIOS, the liquid pump or impeller is dead or clogged.
- Remove the CPU cooler base and inspect the copper cold plate to ensure the factory transparent protective plastic sticker was removed prior to installation.
- Check the thermal paste spread: a clean cross or pea-sized dot of quality non-conductive thermal compound should evenly cover the integrated heat spreader (IHS).
- Tighten cooler mounting screws in a diagonal cross pattern (top-left, bottom-right, top-right, bottom-left) with even hand torque until firmly bottomed out.
- Verify that the CPU cooler fan or pump is plugged into the dedicated CPU_FAN or AIO_PUMP header on the motherboard, not a generic chassis fan header.
Citation: Intel & AMD Thermal Design Integration and Mechanical Guidance Standards
Why does my monitor display 'No Signal' even though the PC lights and fans are working?
A 'No Signal' screen despite active fans occurs when the display cable is plugged into the motherboard I/O rather than the dedicated graphics card, the monitor is set to the wrong input source, or PCIe graphics initialization fails.
Solution Checklist:
- Confirm that the HDMI or DisplayPort cable is plugged directly into the horizontal ports on the dedicated graphics card (GPU), not the vertical ports on the motherboard backplate.
- Use the monitor physical button controls to manually select the correct active video input channel (DisplayPort 1, HDMI 1).
- Test with a different high-speed certified display cable or test a different display output port on the graphics card.
- Reseat the graphics card: release the PCIe slot retention latch, remove the GPU, clean the golden PCIe edge contacts, and seat firmly until the latch clicks shut.
- Ensure all individual 8-pin PCIe power connectors from the PSU are securely clicked into the graphics card, using separate cables rather than daisy-chained pigtails.
Citation: VESA DisplayPort Standard 2.1 & HDMI 2.1 Specification Guidelines
How do I cleanly uninstall and reinstall GPU drivers to fix game crashes and display stuttering?
Corrupted graphics driver registry branches and residual shader cache files can cause recurring DirectX and Vulkan crashes. Using Display Driver Uninstaller (DDU) in Windows Safe Mode completely purges display driver remnants before a clean driver installation.
Solution Checklist:
- Download the latest official graphics driver installer from NVIDIA, AMD, or Intel, and download the portable ZIP of Display Driver Uninstaller (DDU).
- Disconnect your PC from the internet (unplug Ethernet cable or disconnect Wi-Fi) to prevent Windows Update from automatically installing a generic display driver.
- Hold the Shift key while clicking Restart in the Windows Start menu. Go to Troubleshoot -> Advanced options -> Startup Settings -> Restart, then press 4 for Safe Mode.
- In Safe Mode, run Display Driver Uninstaller (DDU), select Device type: GPU, choose your GPU vendor, and click 'Clean and restart'.
- After rebooting into normal Windows, run the downloaded official GPU driver installer, select 'Clean Installation', and reconnect your internet connection once complete.
Citation: Wanyna DDU Official Guidelines & Microsoft Windows Display Driver Model (WDDM)
How do I properly clear the CMOS / reset BIOS settings on any desktop motherboard?
Clearing the Complementary Metal-Oxide Semiconductor (CMOS) resets the motherboard UEFI non-volatile RAM (NVRAM) back to factory defaults. This clears bad overclocking profiles, corrupted memory training data, and improper hardware boot flags.
Solution Checklist:
- Power down the computer completely, switch off the PSU toggle switch (set to 'O'), and pull out the AC wall power plug.
- Locate the two-pin header labeled CLRTC, CLEAR_CMOS, or JBAT1 on the motherboard surface (typically near the bottom edge or near the coin cell battery).
- Touch the metal tip of a flathead screwdriver across both pins simultaneously for 10 to 15 seconds to discharge the RTC memory circuit.
- Alternatively, use a non-conductive plastic pry tool to release the metal retention clip holding the CR2032 3V coin cell battery, remove the battery for 5 minutes, then reinsert it with the '+' positive side facing up.
- Reconnect AC power, power on the system, and allow up to 2 minutes for initial DDR memory retraining before entering UEFI with the Delete or F2 key.
Citation: Unified Extensible Firmware Interface (UEFI) Specification Version 2.10
How do I stop 100% Disk Usage caused by TiWorker.exe, SysMain, or Windows Search?
100% disk usage occurs when Windows background indexing, automated update servicing (TiWorker.exe / TrustedInstaller), or the SysMain prefetch service overwhelms the I/O queue depth, particularly on mechanical HDDs or failing SATA SSDs.
Solution Checklist:
- Open an administrative PowerShell terminal and execute: Stop-Service -Name SysMain -Force followed by Set-Service -Name SysMain -StartupType Disabled.
- Disable Windows Search temporary indexing overload: Stop-Service -Name WSearch -Force.
- Run a disk integrity check: chkdsk C: /f /r in an administrative command prompt to repair file system allocation table corruption.
- Check drive read/write health using CrystalDiskInfo: look for 'Reallocated Sectors Count', 'Current Pending Sector Count', and 'Uncorrectable Sector Count'.
- If the operating system is running on a mechanical spinning hard drive (HDD), clone the Windows installation to a modern NVMe PCIe solid state drive.
Citation: Microsoft Learn Windows Internals: I/O System Architecture and Storage Performance Analysis
How do I determine whether a computer crash is caused by bad RAM or a bad motherboard RAM slot?
To distinguish between a bad RAM stick and a defective motherboard DIMM slot, you must run systematic single-stick isolation testing across known functional channels while logging parity and memory address errors.
Solution Checklist:
- Label your memory modules as Stick A and Stick B using a small piece of painter's tape.
- Insert only Stick A into primary slot DIMM A2 (slot 2). Run MemTest86 for 4 passes. If it passes with 0 errors, Stick A is verified functional.
- Now insert Stick A into slot DIMM B2 (slot 4) and run the test. If it passes, slot B2 is also functional.
- Next, remove Stick A and insert Stick B into slot DIMM A2. Run the test. If errors immediately appear, Stick B is internally defective.
- If both sticks fail in slot B2 but pass in slot A2, the motherboard slot traces, bent CPU socket pins, or memory channel controller on the CPU IMC is damaged.
Citation: JEDEC Memory Reliability Test Standards and Intel Core / AMD Ryzen Memory Architecture Guides