Sensor Response Time Monitoring Using Noise Analysis
H. M. Hashemian, J. A. Thie, B. R. Upadhyaya, K. E. Holbert
Random noise techniques in nuclear power plants have been developed for system surveillance and for analysis of reactor core dynamics. The noise signals also contain information about sensor dynamics, and this can be extracted using frequency, amplitude and time domain analyses. Even though noise analysis has been used for sensor response time testing in some nuclear power plants, an adequate validation of this method has never been carried out. This paper presents the results of limited work recently performed to examine the validity of the noise analysis for sensor response time testing in nuclear power plants. The conclusion is that noise analysis has the potential for detecting gross changes in sensor response but it cannot be used for reliable measurement of response time until more laboratory and field experience is accumulated. The method is more advantageous for testing pressure sensors than it is for temperature sensors. This is because: 1) for temperature sensors, a method called Loop Current Step Response test is available which is quantitatively more exact than noise analysis, 2) no method currently exists for on-line testing of pressure transmitters other than the Power-Interrupt test which is applicable only to force balance pressure transmitters, and 3) pressure sensor response time is affected by sensing line degradation which is inherently taken into account by testing with noise analysis.
Progress in Nuclear Energy, Vol. 21, pp. 583-592, 1988, Proceedings of the Fifth Specialists Meeting on Reactor Noise (SMORN) V, Munich, Germany, October 1987.
Degradation of Nuclear Plant Temperature Sensors
H. M. Hashemian, K. M. Petersen, T. W. Kerlin, R. L. Anderson, K. E. Holbert
A program was established and initial tests were performed to evaluate long term performance of resistance temperature detectors (RTDs) of the type used in U.S. nuclear power plants. The effort addressed the effect of aging on RTD calibration accuracy and response time. This Phase I effort included exposure of thirteen nuclear safety system grade RTD elements to simulated LWR temperatures. Full calibrations were performed initially and monthly, sensors were monitored and cross checked continuously during exposure, and response time tests were performed before and after exposure. Short term calibration drifts of as much as 1.8°F (1°C) were observed. Response times were essentially unaffected by this testing.
NUREG/CR-4928, Prepared for U.S. Nuclear Regulatory Commission, June 1987.
A Remote Method For The In-Situ Response Time Testing of Force Balance Pressure Sensors
Keith E. Holbert
To comply with the NRC requirements of Regulatory Guide 1.118, utilities operating nuclear power plants have begun response time testing of safety system sensors, such as pressure sensors. The pressure sensors are currently tested using a hydraulic ramp generator. This method is both time consuming and requires entry into containment, and therefore, can only be performed at shutdown. A new method has recently been developed for the remote in-situ response time testing of Foxboro force balance pressure sensors. This method involves "tricking" the sensor, through the removal and reinitiation of power to the sensor, into the simulation of a pressure perturbation. The response transient of the sensor is measured, and the ramp response calculated. This transient is identical to one generated by a step change in pressure. This test can be completed in ten minutes from the control room with the plant operating at full power.
Proceeding of the 21st Annual ANS Midwestern Student Conference, Purdue University, March 24, 1984.