Testing backup power, one power cycle at a time

USB power was removed from the base's power-selection board, leaving battery power available. USB power was then restored. During this cycle, the phone received both the power-lost and power-restored text messages.

Afterward, the recorded running times of the base and its Wi-Fi companion extended back before the test. That supports the conclusion that neither restarted during this particular power cycle. This was a test of the existing bench electronics, not hardware fitted inside the new concept enclosures.

Remaining powered and telling the family about a power change are different things to check. In this test we had evidence for both: continued running time and actual receipt of the two text messages. A connected modem or a message accepted for sending would not, by itself, establish phone receipt.

We did not measure phone-delivery latency or capture the brief electrical changes during transfer. This test does not establish performance under every load, a guaranteed backup duration or recovery from every kind of outage.

Earlier battery work included brownout resets, and the cause of a previous thermal incident remains unresolved. Those limits are not erased by a successful power cycle. Battery life, calibration, thermal behavior and more demanding power-transfer scenarios still need qualification.

Further controlled testing needs to check transfer under heavier loads, electrical margins, temperature and usable backup duration. We will report those results with their conditions rather than turn a single observation into a runtime promise.

Alert delivery also depends on available cellular coverage or internet access and outside messaging services. Backup power helps address one dependency; it cannot guarantee that a message reaches someone or that assistance arrives.