[en] The Human Monitoring Laboratory's portable whole body counter has been calibrated using Monte Carlo simulations. The validity of the calibrations was checked by measuring a 95-percentile man phantom distributed by the International Atomic Energy Agency as part of an international In Vivo intercomparison exercise. The agreement between the predicted activity of 137Cs and 60Co and the amounts that were measured in the Human Monitoring Laboratory's fixed whole body counter was 4%. A function was also found that related counting efficiency to photon energy and size of the phantom (or person) being measured. The accuracy of this function was in the range of 8% to -9%. Finally, some retrospective design was performed using a combination of Monte Carlo simulations and measured background data to show that the optimum configuration for the portable whole body counter would have been two detectors that were 9.2 cm in diameter and 12.7 cm thick instead of the current configuration: upper detector, 7.6 cm in diameter and 7.6 cm thick; lower detector, 9.2 cm in diameter and 12.7 cm thick. The modified portable whole body counter would have a minimum detectable activity that was lower than the original design by 13% - 23%, in the range 126 keV to 2,754 keV, for the Reference Male phantom. (Author) 11 refs