Power Path

What this experiment teaches

• Trace complete source-to-rack paths
• Expose a shared upstream dependency
• Separate component count from path resilience

This standalone teaching model isolates one mechanism so its inputs and consequence remain inspectable. The illustrative 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

Name the component whose loss will remove both labelled feeds before you open anything.

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

Compare one failure in a shared A/B topology with the same failure in independent lanes; do not infer certification from the result.

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.

Receive, transform and switch

Utility service enters through protection and switchgear. Transformers change voltage to the levels used by the facility. Automatic or static transfer systems move eligible loads between sources. Every component has a rating, protection scheme and maintenance state.

A generator is a longer-duration source, but it does not carry the first instant of an outage. Detection, start and transfer take time. Fuel quality, cooling, batteries and regular loaded testing matter as much as the generator nameplate.

Bridge and condition with UPS

The uninterruptible power supply bridges the gap when normal power disappears and conditions the output seen by critical loads. Its stored energy is finite. The design question is not “do we have a UPS?” but which load it protects, for how long, through which path, and how the system can be maintained.

UPS conversion creates loss and heat. Efficiency changes with technology and loading, so an oversized lightly loaded unit can perform differently from its peak specification.

Distribute to the rack

Power distribution units and remote panels divide the protected source into branch circuits. Rack PDUs deliver those circuits to equipment. Dual-corded servers can use separate A and B paths, but the redundancy is real only when both cords remain independent upstream.

Operators track kW, amperage, voltage, phase balance and breaker limits. A rack may have empty units but no usable electrical capacity; space and power are separate constraints.

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.

• Utility-to-rack power path — Map every dependency between incoming service and the IT load before discussing redundancy.
• A/B rack feeds — Two rack feeds help only when the upstream paths and the load itself preserve the intended independence.
• Maintenance bypass — A bypass creates an alternate operating state so protected equipment can be isolated—but it also creates new hazards.
• Overhead busway distribution — A modular busway can simplify rack-feed changes, but fault duties, tap-off practices and overhead coordination still govern.

Starting references

• U.S. Department of Energy — Best Practices Guide for Energy-Efficient Data Center Design
• Telecommunications Industry Association — ANSI/TIA-942

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