CubeSat Exoplanet Transit Photometer — Rev A

The intended system would point a small telescope at a selected bright star, focus its light onto the BPW34, and measure brightness every few seconds while holding pointing and detector temperature stable. It would record a baseline before and after the predicted transit, build a light curve of brightness against time, correct that curve for temperature, pointing and electronic drift, and compare the measured transit depth and timing against published predictions. The measurements would be stored onboard and returned through the spacecraft.

That is intended mission behaviour. Rev A is a schematic and a review, and demonstrates none of it.

A transit is a dip of a fraction of one per cent. So is almost everything else that can go wrong.

If the measured brightness falls, the candidates are: a real transit; the star drifting across a detector whose response is not perfectly uniform; a change in focus; a change in detector temperature; amplifier drift; ADC drift; a change in background or stray light; a change in the calibration state; or samples that went missing or arrived corrupted. Every one produces a decrease in a number. Only the first is astronomy.

So the hard part is not detecting a decrease — a 16-bit converter will report decreases all day. It is being able to say afterwards that nothing else could have caused it, which means each of those effects has to be measured well enough to be subtracted or excluded.

The source sets the standard: a transit claim has to be statistically significant and repeatable. Validation is stricter — either two of the payload's own observations agree, or one agrees with an independently confirmed transit, and temperature, pointing and background changes have to be shown not to explain the dip. No precision has been achieved here; the targets exist because the drift sources do.

The acquisition electronics are drawn. The review was about the distance between that and an instrument whose light curve would mean something.

The optical system is the largest gap and has its own section below: no telescope or lens and mirror system, no baffle, no optical filter, no optical mounting. Nothing on the sheet collects light.

Then the missing measurements. There is no temperature sensor near the BPW34, the MCP6002 or the ADS1115, and all three drift with temperature. Fine pointing depends on host-spacecraft systems not present on the payload schematic. Timing has to come from the spacecraft or GNSS and is not a complete Rev A subsystem. The flight power and data interface needs formal definition rather than a development connector.

There is also a labelling error worth stating plainly. The schematic annotates the photodiode as "BPW34 = LED", and the design source is explicit that it is a photodiode and should not be identified as an LED. It is a drawing-label defect rather than a circuit error, and the artefact above is shown as drawn rather than quietly corrected.

Finally the unverified analog: transimpedance gain, noise, stability and bandwidth; ADC gain, rate and linearity; and the calibration LED driver and its optical path.

The Rev A schematic contains the detector electronics, but it does not define how starlight is collected, filtered, focused, and delivered to the photodiode.

The intended optical payload is a chain in its own right: telescope, lens or mirror system, a baffle with internal vanes, an optical filter, then the BPW34 at the focus. None of it is shown in the EasyEDA schematic, and the source says plainly that it is essential.

This is not packaging around the electronics. The telescope sets how many photons arrive, which sets the photometric precision achievable at all. The baffle decides whether sunlight, Earthshine or Moonlight reaches the detector alongside the star. The filter defines the wavelength band, which has to enter the photometric model. The mounting decides whether the stellar image stays on the active area, repeatably.

No aperture or focal length is proposed here. The source requires the telescope to be sized by a photon-budget calculation for a target that has not been selected.

A photometer measures one number, and two things outside the electronics can change it without the star changing at all.

Pointing first. A photodiode's response is not perfectly uniform across its active area, so a star that drifts across the detector changes the measured brightness by itself. Hence the requirements: acquire and verify the target before science collection begins, hold it inside a defined region of the photodiode with motion below 5 per cent of the useful detector width, attach pointing telemetry to at least 99 per cent of science samples, and correct afterwards so pointing-related brightness error falls below 0.05 per cent. Fine pointing is the host spacecraft's job — star tracker, reaction wheels, gyroscope, attitude sensors — none of which is on this schematic.

Temperature second. Photodiode response, amplifier behaviour and converter response all drift with it. The source requires a temperature sensor near the BPW34, the MCP6002 and the ADS1115, sampled at least once per second, accurate to 0.1 degrees Celsius, with temperature attached to at least 99 per cent of photometry samples and corrected drift below 0.05 per cent per hour. The present schematic does not contain one.

Neither correction has been validated, because neither exists yet.

H1 is a two-pin header for an external calibration LED, intended to sit inside the telescope assembly with a stable current driver and a fixed optical path.

Its purpose is real: checking photodiode response, measuring amplifier drift, testing the ADC, and verifying the instrument before and after an observation. The requirement is output repeatable within 0.1 per cent, with LED current and temperature recorded, and an interlock so calibration cannot accidentally run during a stellar observation.

What it is not, in the source's own words, is a substitute for an actual star. It does not arrive through the telescope's pointing, it does not carry the star's spectrum, and it does not reproduce external stray light. It is an internal reference source with a controlled optical path, not a substitute for observing a star through the complete flight observation geometry.

So it answers "did the instrument drift?" It cannot answer "did the star dim?"

I started with an electronics question: can a photodiode measure a star precisely enough to see a transit? That is answerable with a gain calculation and a noise budget, and it is the part I enjoyed.

The review turned it into a systems question, which is harder: can I prove a small dip came from the star rather than from the spacecraft or the instrument? Pointing, temperature, stray light, calibration and baseline stability stopped being support functions and became part of the science measurement, because each can manufacture the exact signal I am looking for.

The calibration LED is the clearest example. It is the only way to tell whether the instrument changed between two moments — but it arrives without the telescope, without the star's spectrum and without the sky behind it. It tells me my ruler moved. It cannot tell me the thing I measured did.