Premium
TH‐C‐19A‐02: Characterization of An Actively Controlled Graphite Probe Calorimeter
Author(s) -
Renaud J,
Sarfehnia A,
Seuntjens J
Publication year - 2014
Publication title -
medical physics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.473
H-Index - 180
eISSN - 2473-4209
pISSN - 0094-2405
DOI - 10.1118/1.4889587
Subject(s) - thermistor , calorimeter (particle physics) , resistive touchscreen , reproducibility , materials science , temperature control , linearity , detector , standard deviation , calorimetry , calibration , analytical chemistry (journal) , optics , electronic engineering , physics , mathematics , electrical engineering , chemistry , statistics , thermodynamics , engineering , chromatography
Purpose: To construct and experimentally evaluate the performance of a miniature probe‐format graphite calorimeter (GPC) with built‐in active control. The GPC is the first ever calorimeter designed specifically for routine clinical use. Methods: The original GPC design, developed as part of our previous work, was modified to accommodate a micro‐thin, resistive heating element and sensing thermistors embedded in the outermost graphite layer. A software‐based process control loop was written to maintain a predefined set point temperature throughout the detector by precisely modulating the current flowing through the resistive element. Continuous feedback is provided to the controller by the additional thermistors. A functioning prototype was constructed in‐house and operated in adiabatic mode in a 6 MV photon beam. Ten sets of measurements of varying duration (10 to 60 seconds) and repetition rate (200 to 600 MU/min) were analyzed. Reproducibility, linearity and dose rate dependence were assessed. Results: A total of 47 individual measurements were performed using the active GPC. The reproducibility, defined here as the sample standard deviation weighted across all 10 measurement sets, was found to be 0.37%. Similarly, the standard error (type A uncertainty) was 0.17%. Linearity was quantified by plotting signal as a function of monitor units delivered at 600 MU/min. The adjusted R‐square of the resulting linear fit was 0.9996. Dose rate dependence, evaluated as the change in normalized response (signal/MU) as a function of repetition rate, was found to be statistically insignificant. Conclusions: The incorporation of active control has resulted in a marked improvement in the GPC's reproducibility and it allows the user to now perform measurements within minutes of setup. With this refinement, it is estimated that the overall standard uncertainty in the absolute determination of dose to water is 1%, making the GPC comparable to a calibrated ionization chamber. This research has received financial support from Sun Nuclear Corporation.