Calibration and Curves
Fan Control Sensor Name Changed or Missing
Stop rapid fan-speed changes around a temperature threshold.
By GetFanControl Editorial Team · Published 2026-09-23 · Last tested 2026-09-23
Purpose of this guide
This guide addresses one task: stop rapid fan-speed changes around a temperature threshold. Fan Control is free to use, but the main application is closed-source; its public releases, open libraries, and plugin projects should not be confused with source code for the complete application.
Before making changes
- Record the current Windows build, Fan Control version, motherboard model, BIOS header mode, connected fan, and temperature source.
- Save the working configuration to a separate folder and keep firmware fan protection enabled.
- Close competing motherboard, GPU, or cooling utilities so two programs do not write to the same controller.
- Confirm that every CPU, GPU, pump, and case fan is physically spinning before calibration.
- Use the current required Microsoft .NET Desktop Runtime shown by the official release. Current modern releases use .NET 10; requirements can change again.
Recommended procedure
- Identify one control. Do not calibrate or modify all headers together. Match one physical fan with its RPM sensor and label it clearly.
- Establish a safe baseline. Note idle and sustained-load temperatures while the existing BIOS or vendor profile is still active.
- Set conservative limits. Use the component manufacturer’s thermal specifications. Example percentages are starting points, not universal safe values.
- Apply one change. Keep a visible temperature monitor open and be ready to return control to firmware if RPM falls unexpectedly.
- Test transitions. Check idle, a moderate load, and a sustained load. Verify that output rises when the selected sensor gets hotter.
- Verify recovery states. Restart Windows and test sleep/resume. Confirm that the saved profile loads and that fans do not remain at an unsafe minimum.
How to validate the result
A curve displayed in the interface is not proof that the physical fan follows it. Compare commanded percentage, reported RPM, component temperature, and audible response. The four signals should change consistently. If RPM becomes zero or unknown while temperature rises, stop the workload and return control to the BIOS or vendor utility.
Run the same repeatable workload before and after the change. Record ambient conditions and avoid comparing a cold first run with a heat-soaked later run. For a CPU or GPU, check temperature, clock behavior, throttling flags, and fan RPM—not temperature alone.
Failure and rollback plan
If a fan stops responding, end the workload first. Disable the affected control card or close Fan Control so firmware can recover control where supported. Reopen the previous saved configuration, restart the computer, and inspect BIOS header settings. Do not repeatedly run calibration on a pump or a critical CPU fan that is not responding.
After a software update, a sensor identifier or controller mapping can change. Keep the prior installer or archive and configuration backup until the new version has passed restart and sleep tests. Never disable Windows Security simply to force a file or plugin to load.
Test environment and sources
Editorial checks use Windows 11 24H2 in a controlled test environment, with before-and-after configuration records and one changed variable per test. Hardware compatibility varies, so results are qualified rather than presented as universal.
Primary references: official release repository, official documentation, and LibreHardwareMonitor. Check current release notes before following version-specific instructions.
Related guides
Quick Answer
Recover curves when hardware sensors are renamed, reordered, or no longer match saved controls. Start with conservative settings, verify the physical fan and sensor mapping, and test temperatures under a representative workload before relying on the profile.
Recommended workflow
- Record the current BIOS/UEFI fan behavior and keep thermal failsafes enabled.
- Confirm the correct temperature sensor and fan header before applying a curve.
- Change one setting at a time and verify RPM response.
- Test idle, sustained load, restart, and sleep/wake behavior.
- Restore firmware control immediately if temperatures or fan response become unsafe.
Frequently asked questions
Why can sensor names change?
Driver, firmware, hardware-monitoring, or device changes can alter exposed sensor identifiers.
What should I do with broken curves?
Re-select the verified physical sensor rather than assuming a similarly named entry is identical.
How do I confirm the mapping?
Apply a controlled workload and verify the selected temperature changes as expected.