Master-Thesis-2024

Exp-H: Long-run OPL + pixel feedback benchmark

Aim

Exp-H tested the limit of temperature-only feedback over a multi-hour interval with optical path length (OPL), environmental and detector-plane measurements evaluated together. The purpose was to determine whether thermal correction could keep the image motion bounded under a strong refractive-index disturbance envelope, and to identify the residual that a second correction layer would need to address.

Selected comparison

Quantity	Warm-up	Feedback	Reading
RMS dY	0.79 px	0.83 px	Detector-plane motion remained comparable to the warm-up scale over the selected long-run comparison.
RMS dX	-	0.16 px	Cross-axis motion stayed smaller than the dominant dY response.
RMS OPL residual	0.77 µm	6.47 µm	OPL residual increased under the environmental disturbance, despite bounded pixel motion.
Pressure span	-	~5.14 hPa	Strong pressure forcing for a temperature-only control experiment.
Approximate duration	-	~8 h	Long-run stress-test window.
Feedback updates	-	19	Event-based thermal corrections during the selected run.

Interpretation

The main result is not that temperature control removed all optical-path variation. It did not: the OPL residual remained sensitive to pressure- and humidity-driven refractivity changes. Rather, Exp-H showed that the temperature loop could maintain first-order image-plane stability under a disturbance envelope far larger than that of the earlier short, quiet tests.

This makes Exp-H a useful temperature-only benchmark. It also defines the reason for the later hierarchy: TEC control handles the slow, coarse component of drift, while a separate fine actuator is needed for residual image motion that cannot be efficiently absorbed by thermal correction alone.

Reporting note

The values above describe the explicitly selected warm-up and feedback windows used in the benchmark analysis. They should not be read as a release of raw telemetry or as a claim that every section of the full run had identical stability.