Visible light is a pointing check, not flare evidence

The VEML7700 is the visible-light sensor in the Rev A schematic, which made it tempting to treat its response as flare evidence while the actual X-ray detector path was still missing.

CubeSat Solar X-Ray Flare Payload — Rev ASensorsRequirements

Problem

A flare is a change in soft-X-ray output. Visible irradiance barely moves during most events, so a visible-light sensor cannot identify or classify a flare — and worse, the things that do move it sharply are exactly the artefacts the instrument has to reject: eclipse entry and exit, loss of pointing, a covered aperture.

Decision

I fixed the sensor's role as a quality check and wrote the constraint into the requirements rather than leaving it as an understanding: visible light is treated as a quality check, not as flare evidence, and a visible-light change alone never produces a flare alert. Its outputs are Sun presence, eclipse entry and exit, coarse pointing, and contamination or aperture problems.

Test / evidence

The acceptance criteria say the same thing from the other side — the sensor must distinguish sunlight, eclipse and covered-sensor conditions, and off-Sun observations must be correctly flagged. It also needs optical attenuation if direct sunlight would saturate it, which is Rev B item 10.

Result

The sensor that is easiest to read is not the sensor that answers the question. Writing down what a measurement may not be used to conclude turned out to be as useful as writing down what it measures.