Interactive Troubleshooter

Published: Sep 9, 2026

HWiNFO: Diagnose Thermal Throttling

Time: 15 mins Level: Beginner

By StepDiag Engineering Team

Reviewed by Ibraheem Taofeeq Opeyemi

An exhaustive root-cause diagnostic tree using HWiNFO64 to isolate thermal throttling across system hardware components. This guide details how to decode IA Thermal Throttling flags, distinguish between VRM and package limits, diagnose GPU memory hotspot issues, and remediate NVMe thermal drops.

Full Diagnostic Tree & Step-by-Step Overview

Which component or sensor section in HWiNFO displays active thermal throttling flags?

CPU DTS / Core section flags 'IA Thermal Throttling' or 'Core Thermal Throttling' as Yes
GPU section displays 'GPU Thermal Throttling' or extreme Delta between Core and Memory Hotspot
Motherboard / VRM section shows high 'VR MOS' temperatures or 'VRM Thermal Throttling'
NVMe / Storage section reports high 'Drive Temperature 2' (Controller) and read/write drops

What is the primary power/thermal characteristic of the CPU during the throttling event?

CPU temperatures instantly spike to TjMax (e.g., 95°C–100°C) within seconds of load application
CPU temperature rises slowly over 5–10 minutes before triggering thermal limit flags
CPU clock frequencies drop sharply while package temperature remains well below TjMax (Power Limit Throttling)

Fix Immediate CPU Thermal Spikes & Heat Transfer Breakdown

Solution:

Root Cause: Thermal Interface Material (TIM) Breakdown or Cooler Mounting Fault

An instantaneous spike to TjMax (100°C on modern Intel architecture or 95°C on AMD Ryzen) upon applying load indicates a catastrophic breakdown in thermal conductivity between the CPU integrated heat spreader (IHS) and the cooler cold plate. The CPU core logic generates heat faster than it can be conducted away, triggering the internal Digital Thermal Sensors (DTS) to trip IA Thermal Throttling or PROCHOT flags to prevent physical silicon destruction.

Diagnostic Verification:

1Open HWiNFO64 in Sensors-only mode.

2Locate the CPU section and observe CPU Package Temperature alongside IA Thermal Throttling.

3Execute a short stress test (e.g., Prime95 or Cinebench).

4If temperatures jump from 40°C idle to 100°C within <2 seconds while fan speeds are at maximum RPM, heat transfer is severely compromised.

Step-by-Step Fix:

1Inspect & Dismantle Cooler Mounting Assembly:

Power off the system, unseat the AIO/air cooler cold plate, and check for unremoved plastic protective films on the cold plate surface.

Clean off old, dried thermal compound from the IHS and cold plate using 99% Isopropyl Alcohol.

2Apply High-Performance Thermal Interface Material:

Apply a fresh dot/cross pattern of high-viscosity thermal paste or Phase Change Material (PCM) pads (e.g., Honeywell PTM7950).

3Evenly Torque Mounting Hardware:

Reattach the cooler using a diagonal cross-pattern to distribute mounting pressure evenly across all socket standoffs.

4Verify AIO Pump Operation (If using liquid cooling):

In HWiNFO, verify that the AIO Pump RPM sensor reports active rotation (>2000 RPM). A dead pump impellor leaves liquid stagnant, causing instant thermal throttling.

Prevention & Long-Term Monitoring:

Re-paste high-TDP CPUs every 18–24 months to prevent pump-out effects caused by thermal cycling.

Fix Gradual CPU Thermal Saturation & Exhaust Airflow Restrictions

Solution:

Root Cause: Heat Dissipation Capacity Exhaustion or Case Thermal Saturation

When a CPU takes several minutes of sustained load before hitting its thermal threshold, the immediate heat transfer from the IHS to the cooler is functional, but the cooling ecosystem lacks sufficient heat dissipation capacity. This occurs when radiator fins accumulate dust, ambient case air becomes heat-soaked due to poor exhaust CFM, or the cooler TDP rating is under-spec for the processor's active power draw.

Diagnostic Verification:

1Open HWiNFO64 and launch CSV logging by clicking the Logging Start button at the bottom right.

2Run a 15-minute Cinebench loop.

3Analyze the CSV log or graph view for System Temperature, CPU Package, and Fan RPM.

4Observe if internal ambient sensor values rise continuously alongside the CPU package, indicating hot air trapping.

Step-by-Step Fix:

1Clear Heat Exchangers & Dust Filters:

Clean heatsink fins, radiator meshes, and case dust filters using compressed air.

2Optimize Fan Curves in UEFI/BIOS:

Boot into BIOS/UEFI, navigate to Hardware Monitor / Fan Control, and adjust fan curves to hit 100% duty cycle when the CPU reaches 80°C.

3Rebalance System Airflow Pressure:

Ensure case intake CFM exceeds or matches exhaust CFM to maintain positive static pressure. Ensure front/bottom fans act as intake and rear/top fans act as exhaust.

Prevention & Long-Term Monitoring:

Set up HWiNFO automatic alert actions under Options -> Alerts to sound an alarm or trigger a script when CPU Package exceeds 90°C for more than 60 seconds.

Fix CPU Power Limit Throttling (PL1 / PL2 / EDC / TDC Restrictions)

Solution:

Root Cause: Firmware Power Limit Enforcement (PL1/PL2 or PPT/EDC/TDC Limits)

If CPU clock speeds drop while package temperatures remain well below thermal limits (e.g., 70°C–80°C), the processor is experiencing Power Limit Throttling rather than Thermal Throttling. Modern motherboard BIOS profiles enforce strict Short Power Duration (PL2 / PPT) and Long Power Duration (PL1) limits. Once the tau timer expires, the CPU scales back clock frequencies to stay within the long-term TDP envelope.

Diagnostic Verification:

1Expand the CPU status section in HWiNFO64.

2Check flags for Power Limit Exceeded, PL1 Limit, PL2 Limit, or Package Power Limit Active.

3Compare active package power (Watts) against rated processor Tau specifications.

Step-by-Step Fix:

1Adjust Power Limits in UEFI/BIOS:

Restart into BIOS, locate CPU Power Management (e.g., ASUS Performance Enhancement, MSI PL1 Limits, or Gigabyte Multi-Core Enhancement).

Adjust PL1 and PL2 power limits to match your cooler's rated dissipation capacity (e.g., set to 253W for Intel i9 or unlock PPT for AMD Ryzen).

2Apply Undervolting via Offset / Curve Optimizer:

Reduce operating voltage to lower power draw while maintaining clock speeds.

For AMD platforms, use PBO Curve Optimizer to set a negative per-core offset (e.g., -15 to -20).

For Intel platforms, apply a negative Vcore voltage offset (e.g., -0.050V) using official tools like Intel Extreme Tuning Utility (XTU).

Prevention & Long-Term Monitoring:

Ensure the motherboard VRM phases can supply the required continuous current before removing BIOS power limits.

What specific GPU thermal symptom or delta is reported in HWiNFO?

High Delta (>20°C–25°C) between GPU Core Temperature and GPU Memory / Hotspot Temperature
GPU Thermal Throttling flag triggers 'Yes' while GPU Fan RPM is at 100% (Core Overheating)

Fix GPU Hotspot Delta & VRAM Thermal Pad Degradation

Solution:

Root Cause: Uneven Cold-Plate Contact Pressure or VRAM Pad Breakdown

A large delta (exceeding 20°C–25°C) between the standard GPU Core Temperature and the GPU Hotspot or Memory Junction Temperature indicates uneven mounting pressure or degraded thermal interface pads on VRAM/VRM modules. When hotspot temperatures exceed 105°C (NVIDIA) or 110°C (AMD), the GPU driver automatically forces core clocks down to prevent VRAM degradation, even if the primary core sensor reads a cool 65°C.

Diagnostic Verification:

1Run a 3D graphics benchmark (e.g., 3DMark or Unigine Heaven) while logging with HWiNFO64.

2Scroll to the GPU sensor block and monitor GPU Temperature, GPU Hot Spot Temperature, and GPU Memory Junction Temperature.

3Subtract GPU Core Temp from Hotspot Temp. If Hotspot - Core > 22°C, thermal contact across the silicon die is non-uniform.

Step-by-Step Fix:

1Disassemble GPU Cooler Assembly:

Remove backplate screws and carefully disconnect fan/ARGB headers.

2Replace VRAM Thermal Pads with Exact Thicknesses:

Measure original thermal pad thicknesses using digital calipers (e.g., 1.0mm, 1.5mm, or 2.0mm). Incorrect pad thickness prevents the main cold plate from contacting the GPU core die.

3Apply Viscous Phase-Change Material to GPU Die:

Clean the GPU die with Isopropyl Alcohol and apply a high-viscosity compound or PTM7950 pad to prevent paste pump-out.

4Re-torque GPU Retention Bracket:

Tighten the four spring-loaded screws surrounding the GPU core in an alternating X-pattern to ensure equal mounting pressure.

Prevention & Long-Term Monitoring:

Support heavy graphics cards with a GPU anti-sag bracket to prevent PCB warping and mounting plate separation over time.

Fix GPU Core Thermal Throttling via Undervolting & Custom Fan Curves

Solution:

Root Cause: Excessive Default Voltage Frequency Curve & Dust Saturation

Modern GPUs ship with aggressive factory boost algorithms that push core voltages high to guarantee stability across all silicon quality bins. Over time, as heatsinks gather dust or ambient case temperatures rise, the GPU core reaches its thermal limit (83°C default on NVIDIA, 85°C on AMD), causing the driver to lower core clock states and trigger the GPU Thermal Throttling flag.

Diagnostic Verification:

1Open HWiNFO64 and scroll to the GPU section.

2Monitor GPU Performance Limit - Thermal and GPU Thermal Throttling flags.

3If these flags switch to 'Yes' while fans run at max speed, core cooling is insufficient for the current power delivery.

Step-by-Step Fix:

1Clean GPU Heatsink & Fans:

Use compressed air to blow out dust trapped within the shroud and heatsink fins.

2Apply GPU Core Undervolt via MSI Afterburner:

Open MSI Afterburner and press Ctrl + F to open the Voltage/Frequency Curve Editor.

Flatten the curve above a stable target frequency (e.g., set 1900 MHz at 875mV instead of the factory 1050mV).

Click Apply and test stability in a looping graphics benchmark.

3Set Custom Aggressive Fan Curve:

In Afterburner settings, enable user-defined software fan control and set fans to reach 80% speed at 75°C.

Prevention & Long-Term Monitoring:

Periodically monitor GPU fan RPMs in HWiNFO to ensure secondary slave fans have not suffered bearing failure.

Which VRM / Power Delivery symptom or sensor reading is active in HWiNFO?

Motherboard 'VR MOS' / 'VRM Temperature' sensor exceeds 105°C during sustained CPU loads
CPU severely drops frequency to minimum base clock (800 MHz) without high CPU core temps

Fix Motherboard VRM Overheating & MOSFET Airflow Deficits

Solution:

Root Cause: Inadequate Voltage Regulator Module (VRM) Cooling

The motherboard VRM steps down 12V power from the PSU to the low-voltage (~1.2V) high-current power required by the CPU. Under heavy multi-threaded workloads, uncooled or budget VRM MOSFETs generate massive heat. When VRM temperatures exceed 105°C–120°C, the power delivery system triggers internal safety cutoffs, forcing the CPU to throttle power draw to prevent MOSFET destruction.

Diagnostic Verification:

1In HWiNFO64, navigate to the Motherboard sensor section.

2Monitor sensors labeled VR MOS, CPU VRM, or VRM Temperature.

3Observe if VRM temperatures climb past 100°C while the CPU power draw drops sharply.

Step-by-Step Fix:

1Provide Direct Active Airflow across VRM Heatsinks:

Liquid cooling AIOs remove CPU fan airflow from the socket area, leaving VRM heatsinks stagnant. Re-orient top/rear chassis fans to direct airflow over the VRM heatsinks surrounding the CPU socket.

2Install Aluminum VRM Heatsinks & Thermal Pads:

On budget motherboards lacking factory heatsinks, apply aftermarket copper/aluminum heatsinks using thermally conductive adhesive tape to all exposed power phases.

3Lower CPU Load Line Calibration (LLC) & Vcore Voltage:

In BIOS, avoid aggressive LLC modes (e.g., Level 1/2) that cause VRM switching losses. Reduce CPU Vcore to decrease current demand on the power phases.

Prevention & Long-Term Monitoring:

Pair high-TDP multi-core CPUs only with motherboards featuring robust multi-phase VRM designs and factory heatsinks.

Fix BDPROCHOT Signals & Fake Thermal Throttling Locks (800 MHz Multiplier Lock)

Solution:

Root Cause: External BDPROCHOT (Bi-Directional Processor Hot) Signal Tripping

BDPROCHOT is a system signal line that allows non-CPU components (such as motherboard power phases, power supplies, or external sensors) to inform the CPU that the system is overheating. When a broken motherboard sensor or faulty power supply trips BDPROCHOT, the CPU immediately locks itself to its lowest energy state (typically 800 MHz or a 8x multiplier), even if CPU core temperatures are completely cool (e.g., 30°C).

Diagnostic Verification:

1Open HWiNFO64 and observe CPU Core Clocks locked at 800 MHz under full load.

2Check BDPROCHOT under CPU flags—if it displays 'Yes' while core temperatures are low, an external motherboard sensor is sending a false throttling command.

Step-by-Step Fix:

1Reset Motherboard CMOS & Power Drain:

Shutdown, unplug the PSU, and remove the CR2032 CMOS battery for 5 minutes to clear misconfigured sensor states.

2Inspect Physical Sensors & Power Supply Cable Connections:

Ensure all 8-pin CPU power cables are securely seated and not delivering unstable voltage rails.

3Bypass BDPROCHOT via ThrottleStop (Software Workaround):

If caused by a failed motherboard sensor on a legacy system, download and launch ThrottleStop.

Uncheck BDPROCHOT on the main interface and click Save.

Verify in HWiNFO that CPU clock speeds immediately restore to normal operating frequencies.

Prevention & Long-Term Monitoring:

Replace failing power supplies or damaged motherboards if BDPROCHOT trips due to genuine electrical fault conditions.

Which NVMe storage thermal metric or behavior is reported in HWiNFO?

Drive Temperature 2 (ASIC Controller) exceeds 80°C–85°C while NAND Flash stays cool
NVMe read/write transfer speeds collapse under sustained file transfers with high overall drive temps

Fix NVMe Controller Thermal Throttling (Drive Temperature 2)

Solution:

Root Cause: NVMe ASIC Controller Overheating

Modern PCIe 4.0 and PCIe 5.0 M.2 SSDs report multiple temperature metrics in HWiNFO: Drive Temperature (NAND Flash cells) and Drive Temperature 2 (ASIC Master Controller). While NAND flash memory operates best at warm temperatures (40°C–60°C), the memory controller chip generates intense heat. When the controller exceeds 80°C–85°C, it enters self-preservation thermal throttling, reducing I/O queue depth and transfer speeds.

Diagnostic Verification:

1Open HWiNFO64, expand your NVMe drive section, and monitor Drive Temperature 2.

2Run a CrystalDiskMark benchmark.

3If Drive Temperature 2 hits 85°C and the Thermal Throttling attribute switches to 'Yes', the controller chip requires immediate heat sinking.

Step-by-Step Fix:

1Install M.2 Heatsink with Thermal Pads:

Remove the M.2 SSD and apply a thermal pad directly over the ASIC controller chip.

Attach a dedicated aluminum/copper M.2 heatsink or connect it to the motherboard's integrated M.2 shield.

2Ensure High Thermal Conductivity to the Heatsink:

Check that thermal pads make firm physical contact with both the controller chip and the heatsink surface.

3Avoid Direct GPU Exhaust:

If the primary M.2 slot sits directly beneath an open-air GPU shroud, relocate secondary NVMe drives to lower motherboard M.2 slots equipped with passive shields.

Prevention & Long-Term Monitoring:

Always use dedicated heatsinks on high-speed PCIe 4.0 and 5.0 NVMe storage drives.

Fix PCIe Bus Lane Saturation & Overall NVMe Heat Soaking

Solution:

Root Cause: Enclosure Heat Trapping & Thermal Pad Misalignment

Sustained large-scale read/write operations (e.g., video editing or database transfers) cause both the NAND flash modules and the SSD controller to emit heat simultaneously. In poorly ventilated PC cases or fanless external USB-C enclosures, the thermal mass of small M.2 heatsinks saturates, locking the drive into persistent thermal throttling.

Diagnostic Verification:

1Open HWiNFO64 and scroll to the NVMe SMART telemetry section.

2Check Warning Composite Temperature Time and Critical Composite Temperature Time.

3If these timers show active increments, the drive has spent extended duration operating in dangerous thermal zones.

Step-by-Step Fix:

1Upgrade M.2 Thermal Solutions:

Replace basic flat metal heatspreaders with taller, finned passive heatsinks or active fan-assisted M.2 coolers.

2Increase Enclosure Airflow (For External NVMe Enclosures):

Use aluminum external enclosures that utilize thermal transfer pads to turn the entire outer shell into a passive heat sink.

3Force Active Drive Idle States in Windows Power Options:

Press Win + R, type powercfg.cpl, navigate to Change plan settings -> Change advanced power settings -> PCI Express -> Link State Power Management, and ensure power saving states allow the drive to drop voltage when idle.

Prevention & Long-Term Monitoring:

Monitor NVMe composite health status in HWiNFO regularly to ensure high operational thermals have not reduced remaining drive endurance life.

Related Diagnostic Guides

Windows Update Error 0x80070002 Missing Files Diagnostic Guide

System Restore Failed 0x80070005 Antivirus Conflict Tree

Windows Photo Viewer Not Showing Thumbnails: Codec Diagnostic

Safe Mode Password Not Working With Microsoft Account Fix

Clipboard History Not Working Windows 11: Sync Service Tree

Subscribe to StepDiag

Embed Guide on Your Website