Measured Biases between Calorimeter Electrical Calibration System and 238Pu Heat Standards

Year
2002
Author(s)
M. Newell - Los Alamos National Laboratory
David S. Bracken - Los Alamos National Laboratory
J.E. Schauer - Los Alamos National Laboratory
Abstract
A calorimeter electrical calibration system (CECS) has been developed for use in existing radiometric calorimeters without the need for any mechanical design modifications. The CECS is composed of a resistive heater probe and a control unit. Once inserted into the measurement chamber, the power output of the probe is set by the operator at the control unit. The CECS was tested and compared with 238Pu heat standards to determine its bias by means of a sensitive twinbridge calorimeter operated in passive mode. The desire for an electrical calibration system to replace or supplement 238Pu heat standards is apparent throughout the Department of Energy (DOE) complex for a number of reasons. The most pressing recent issue with the use of plutonium heat standards is the lack of any certified shipping containers, making it difficult to ship heat standards for recertification. The calibration certification period is usually five years. Periodic recalibration is necessary because of the DOE Albuquerque Operations Office calibration program order AL 57XA. Other issues associated with using plutonium heat standards will be presented in this paper. Because of measured biases between the CECS and 238Pu heat standards over a power range of 0.006–1.8 watts, this CECS can be used only as a calibration device when bias corrections are made. However, this must be done with a calorimeter that has been previously calibrated by using 238Pu heat standards. Even with bias corrections, the uncertainty in calibrating with the CECS is only comparable to calibrating with 238Pu heat standards above settings of 1 watt. Here the increase in uncertainty was about 0.15%. Below 1 watt, the increase in uncertainty in calibrating with the CECS grows exponentially with decreasing power. At low powers, the total measurement uncertainty that is due to using the CECS and a bias correction can be two to four times larger.