What we did
A known effect was planted in designed test data with a known reference value. RefCal was asked to recover that value. The same run checked that already-shipped results stayed unchanged while the new capability was present but inactive.
What we saw
RefCal recovered the planted effect within the tolerance set before the run. Its existing test suite and a check of every already-shipped result showed no change to results already in production. A further run on a small set of real measurements completed without error, but that data set was too small to produce a usable fitted result and none was used.
What it means
RefCal can recover a known effect from designed data without disturbing anything already live. This establishes the mechanism on data where the answer is known. The next step is a real-data run large enough to support a checkable result.