One reference set, three annual physicist workflows.

Warm-up the dose calibrator and let the source set come to room temperature alongside it. Thermal drift is the #1 single-day variance noise.

Read background; subtract; verify the calibrator is on the correct isotope setting before each source.

Insert source in the standard orientation. Read three times, take the mean.

Decay-correct the certificate activity to the time of measurement. Δ = (A_measured − A_expected) / A_expected × 100 %.

PASS criterion: Δ within ± 10 % across all four isotopes (AERB / AAPM TG-181 annual accuracy threshold).

Document the result in the annual QA dossier alongside the constancy trend log from the past year.

Use the long-lived Cs-137 reference (T½ 30.1 y) as the activity-decade-spanning probe. Decay won't close the working range during the test.

Read the source through shielded attenuators producing known transmission factors (typically 5 attenuator pots producing 1× / 0.5× / 0.2× / 0.1× / 0.05× geometry).

Plot measured activity vs expected activity on a log-log axis. Fit line; expected slope = 1.0, intercept = 0.

PASS criterion: deviation from linearity ≤ 5 % across the clinical activity range (typically 37 MBq to 37 GBq).

Document the linearity plot in the annual dossier. Re-run after dose-calibrator service or chamber replacement.

Energy calibration: use Ba-133 / Co-57 / Cs-137 / Co-60 sequentially to map the energy-channel response of the well counter's NaI(Tl) crystal.

Plot channel number vs photon energy; expected linear fit. Slope and intercept feed the spectrum-analysis software.

Counting efficiency: at each isotope, measure CPM at known activity; efficiency = CPM / (Bq × 60). Plot efficiency vs energy.

Re-run after PMT replacement, photomultiplier high-voltage adjustment, or annual scheduled QA.

Cross-check against the dose calibrator linearity result — well counter and dose calibrator should agree on the same Cs-137 activity to within their combined uncertainties (typically ± 10 %).