Cooling Heat Path

What this experiment teaches

• Keep captured and residual heat visible
• Expose entered support power
• Avoid assigning one universal efficiency to a cooling family

This standalone teaching model isolates one mechanism so its inputs and consequence remain inspectable. The calculated label describes the output kind; it does not turn the result into measured facility data, product selection, a commissioned design or professional advice.

Predict before changing the model

Estimate how much room-air heat remains after changing liquid capture, before reading the calculated split.

Record the expected direction in your own words. Then change one input at a time and compare the result with that prediction. If the output surprises you, open the model card and inspect the boundary and assumptions before creating a story around the number or state.

Stress and debrief

Raise IT heat while holding support assumptions fixed, then explain which unmodeled plant limits would need real engineering data.

A useful debrief names four things separately: the input that changed, the modeled consequence, the important effects that were not evaluated and the site data or engineering work needed before a real decision. The lab deliberately avoids a universal score because energy, resilience, capacity, maintainability and risk are different dimensions.

CRAC and CRAH are not interchangeable labels

A computer room air conditioner commonly uses a direct-expansion refrigeration circuit in or near the room. A computer room air handler uses a chilled-water coil supplied by a central plant. Both move room air, but the heat travels through different equipment and failure domains.

Names in the field are sometimes used loosely. Read the actual mechanical sequence, refrigerant or water connections and controls rather than inferring the whole design from one acronym.

The plant rejects heat outside

In a chilled-water design, the CRAH transfers room heat into chilled water. The chiller moves that heat to a condenser loop or another rejection path; cooling towers may then reject it to ambient air through evaporation. Pumps, fans and controls are part of the energy and reliability picture.

Economizers can use favorable outdoor conditions to reduce compressor work. Water availability, water quality, climate and local rules shape the best design.

Liquid moves closer to the chip

Rear-door heat exchangers capture rack exhaust. Direct-to-chip cold plates carry liquid to processors. Immersion places equipment in dielectric fluid. These approaches can support high heat flux with less room-air movement, but they introduce manifolds, leak detection, water quality, service procedures and new skills.

There is no universal winner. Density, hardware compatibility, heat-reuse goals, redundancy, maintenance and the existing building determine the practical path.

Continue the evidence trail

Open the connected Academy lesson. The lesson provides the formula or mechanism, common misconception, knowledge check and source context that surround this compact experiment.

• Hybrid liquid-cooled zone — Liquid capture can remove part of the rack heat close to the source while a residual air system still serves the rest.

Starting references

• U.S. Department of Energy — Cooling Water Efficiency Opportunities
• ASHRAE — AI Data Center Energy and Thermal Efficiency

Current standards, adopted requirements, verified site information, manufacturer data and qualified professional review remain necessary for real work.