Fan Control Module Using GPU Power and Temperature Signals

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing fan control algorithms in computer systems face challenges in determining an optimal temperature range for fan operation, leading to frequent cycling, which causes thermal stress and user annoyance, as the difference between fan-on and fan-off temperatures can be too small or too large, affecting thermal management effectively.

Innovation Solution

A computer-implemented method that controls fan speed based on both temperature and power levels, turning the fan on in response to increasing power before temperature increases and turning it off only when both temperature and power are below respective thresholds, using a fan control module to manage fan operation and reduce cycling.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Speed

If the fan-on temperature and fan-off temperature are set close together, then the fan responds quickly to temperature changes, but fan cycling becomes frequent causing user annoyance

Engineering Contradiction:
Improvefan response speedVSAvoiduser experience
Core Design Contradiction:
SpeedVSReliability

Solution Approach 1:

The system performs preliminary cooling action by turning on the fan when power increases, before the temperature actually rises to the fan-on threshold. This anticipatory approach prevents temperature spikes that would trigger frequent cycling, thereby maintaining user comfort while ensuring rapid thermal response when needed.

Inventive Principle:
Principle #10Preliminary action

2Reliability

If the fan-on temperature and fan-off temperature are set far apart, then fan cycling is reduced, but thermal stress on components increases

Engineering Contradiction:
Improvefan cycling reductionVSAvoidthermal stress
Core Design Contradiction:
ReliabilityVSStress or pressure

Solution Approach 1:

The system takes preliminary cooling action by activating the fan based on power level increases before temperature reaches critical thresholds. This prevents large temperature excursions and reduces thermal cycling stress on components while still maintaining adequate cooling coverage.

Inventive Principle:
Principle #10Preliminary action

3Device complexity

If the fan-on temperature and fan-off temperature are set far apart, then thermal management is simpler, but the fan may not turn off when it otherwise could

Engineering Contradiction:
Improvetemperature threshold managementVSAvoidfan energy consumption
Core Design Contradiction:
Device complexityVSUse of energy by moving object

Solution Approach 1:

The system uses power level as a preliminary indicator to trigger fan operation before temperature reaches the fan-on threshold. This allows for more precise control where the fan can turn off at appropriately set temperatures while ensuring it activates in time to prevent excessive heating, optimizing both energy efficiency and thermal management.

Inventive Principle:
Principle #10Preliminary action

Applied Scientific Principles

This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.

Function Achieved in This Case

This approach reduces fan cycling, minimizes thermal stress on components, and provides a more graceful temperature rise, enhancing system reliability and user experience by anticipating temperature increases based on power usage and maintaining fan operation only when necessary.

Implementation Method 1

Computer systems typically include one or more mechanical fans to provide convection cooling of dissipating components.

Methodology Applied
Scientific EffectConvection cooling: Convection

Data Source

PatentUS10571978B2Techniques for reducing fan cycling
Publication Date: 2020.02.25 NVIDIA CORP
  • US10571978B2 patent drawing
  • US10571978B2 patent drawing
  • US10571978B2 patent drawing

AI summary

A fan control module configured to control the speed of a fan receives a signal that indicates the power used by a graphics processing unit (GPU) and a signal that indicates the GPU temperature. Whenever the GPU power exceeds a power threshold level, but the GPU temperature is below a temperature threshold level, the control module turns the fan on and causes the fan to operate at a minimum speed. Whenever the GPU temperature is above the temperature threshold, the control module causes the fan speed to increase with increasing temperature, regardless of power. The control module turns the fan off only when both the GPU temperature is below the temperature threshold and the GPU power is below the power threshold. Although the algorithm is discussed in conjunction with a GPU, the algorithm can be implemented with any type of processor or subsystem that needs to be fan-cooled.