Add-In Card Backup Cooling for Host Power Loss Shutdown

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Standard rack mount servers often lose power, causing add-in devices to stop operating and overheat due to the lack of cooling, as the server fans cease functioning, and existing solutions do not allow for extended operation after power loss.

Innovation Solution

Add-in cards are equipped with a backup power system, such as a battery unit, and an air circulator, like a fan or blower, to continue operation and provide forced convection cooling, maintaining airflow even when the host server loses power.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Duration of action of moving object

If add-in devices continue operating after host server loses power, then extended operation time is achieved, but heat generation increases causing potential overheating

Engineering Contradiction:
Improveoperation durationVSAvoidcomponent temperature
Core Design Contradiction:
Duration of action of moving objectVSTemperature

Solution Approach 1:

The add-in device detects power loss from the host server and proactively activates its backup power system and air circulator before critical overheating occurs. This preliminary action allows the device to transition to self-powered operation and establish cooling airflow immediately, preventing the temperature rise that would otherwise occur during extended operation.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The air circulator acts as an intermediary cooling mechanism that introduces external airflow into the add-in device when the host server's cooling system becomes unavailable. This intermediary cooling approach enables the device to dissipate heat generated during extended operation on backup power, resolving the contradiction between continued operation and temperature control.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Temperature

If host server fans provide cooling for add-in devices, then cooling efficiency is improved, but system dependency increases making add-in devices vulnerable to power loss

Engineering Contradiction:
Improvecooling efficiencyVSAvoidsystem reliability
Core Design Contradiction:
TemperatureVSReliability

Solution Approach 1:

The add-in device is equipped with its own backup power system and air circulator, enabling it to provide self-service cooling when the host server fails. This self-service capability allows the device to maintain its own cooling airflow independently of the host server's cooling system, thereby improving reliability without sacrificing cooling efficiency.

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The cooling system is segmented into host server-level cooling (fans) and add-in device-level cooling (air circulator). This segmentation allows the add-in device to have its own dedicated cooling capability that operates independently when needed, reducing vulnerability to host server failures while maintaining efficient cooling when the host is operational.

Inventive Principle:
Principle #1Segmentation

3Reliability

If backup power system and air circulator are added to add-in cards, then fault tolerance is improved, but device complexity increases

Engineering Contradiction:
Improvefault toleranceVSAvoiddevice complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The backup power system and air circulator are designed as integrated components that serve multiple functions: providing extended operation time, enabling independent cooling, and ensuring safe shutdown capability. This multi-functionality approach improves fault tolerance while minimizing the increase in device complexity by combining multiple protective functions into unified subsystems.

Inventive Principle:
Principle #6Universality (Multi-functionality)

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

The add-in cards can operate for several minutes post-power loss, ensuring safe shutdown and preventing overheating, thus preserving data and critical components.

Implementation Method 1

the air circulator in the add-in card may be turned off during normal operation and operated only if cooling is needed when the add-in card continues operation while server power or cooling is off

Methodology Applied
Scientific EffectForced convection: Forced Convection

Data Source

PatentUS12504796B2Fault-tolerant cooling for add-in card
Publication Date: 2025.12.23 NVIDIA CORP
  • US12504796B2 patent drawing
  • US12504796B2 patent drawing
  • US12504796B2 patent drawing

AI summary

An add-in device (100) for use in a host server (300) is capable of operating for an extended time after the host server (300) stops providing power to the add-in device (100). The add-in device (100) may include a circuit board assembly (110) and a cover (120) that define a duct through the add-in device (100). An inlet (170) of the duct receives cooling air flow that the host (300) normally drives through the duct. A blower (160) inside the duct may operate from a backup power source (140) to pull air flow through the duct to maintain forced convection cooling when the host (300) fails to provide power or cooling air flow to the add-in device (100). The backup power can also continue operation of the device (100) to execute a shutdown procedure after the host fails.