Abstract
The Berkeley TRIGA Mark III Reactor has three distinct reflector arrangements, depending on the position of the core in the pool. The control rods must be calibrated in each position, making 12 rod calibrations required, in all. To eliminate the human element and the drudgery involved in this repetitious task, a computer-assisted semi-automatic method has been devised to perform the necessary period methods, and to produce the resultant rod-calibration curves. The method is based on the use of a signal from the linear-power-channel recorder to feed a voltage comparator which generates a pulse at a preselected voltage 'B' and also at '1.50V'. The 2 pulses are used to start and stop pulses for an electronic timer, which easily measures the time difference to 0.01 second. The comparator actually consists of two such pulse-pair generating circuits, so that 2 measurements of t{sub 50} can be obtained on each range of the linear-power channel. Before the comparator is used for a series of rod calibrations, the voltage discrimination levels are checked with a precision voltage source to verify that they are set at 3.50, 5.25, 6.00, and 9.00 volts. Corrections in the discrimination levels can be made by means of front-panel potentiometer
More>>
Citation Formats
Ruby, L, and Wang, H -K.
Eliminating the human element and the drudgery from control-rod calibrations.
United States: N. p.,
1974.
Web.
Ruby, L, & Wang, H -K.
Eliminating the human element and the drudgery from control-rod calibrations.
United States.
Ruby, L, and Wang, H -K.
1974.
"Eliminating the human element and the drudgery from control-rod calibrations."
United States.
@misc{etde_21217742,
title = {Eliminating the human element and the drudgery from control-rod calibrations}
author = {Ruby, L, and Wang, H -K}
abstractNote = {The Berkeley TRIGA Mark III Reactor has three distinct reflector arrangements, depending on the position of the core in the pool. The control rods must be calibrated in each position, making 12 rod calibrations required, in all. To eliminate the human element and the drudgery involved in this repetitious task, a computer-assisted semi-automatic method has been devised to perform the necessary period methods, and to produce the resultant rod-calibration curves. The method is based on the use of a signal from the linear-power-channel recorder to feed a voltage comparator which generates a pulse at a preselected voltage 'B' and also at '1.50V'. The 2 pulses are used to start and stop pulses for an electronic timer, which easily measures the time difference to 0.01 second. The comparator actually consists of two such pulse-pair generating circuits, so that 2 measurements of t{sub 50} can be obtained on each range of the linear-power channel. Before the comparator is used for a series of rod calibrations, the voltage discrimination levels are checked with a precision voltage source to verify that they are set at 3.50, 5.25, 6.00, and 9.00 volts. Corrections in the discrimination levels can be made by means of front-panel potentiometer adjustments. As voltage is gradually increased past each of the pre-set discrimination levels, a panel light comes on, indicating that a pulse has been formed. The comparator circuit also accepts a reset command from a push button held in the hand of the reactor operator, which command is then converted into an electrical reset signal for the electronic timer. The system provides non-prejudiced measurements for t{sub 50} as short as 5 seconds, with no concern about pen lag. The only manipulation of the data is to determine the best value of t{sub 50}, which is done by averaging those values which agree to within 0.1 second. The program ''RODCALN'' is used to calculate the rod worth remaining (in dollar units) versus control rod position, and a RODCALN provides both a table and a plot of the results. From an input of t{sub 50}'s obtained over the whole rod length, the program determines the reactivity for each increment of rod distance by a continuous function which approximates the table of Inhour solutions prepared by GA. The plots are of sufficient detail to be used in routine reactor operation for reactivity determinations. The results obtained with this program have been compared with Randall's contention that a consine-squared function is an excellent fit to the data for the differential worth of a control rod. The close agreement confirms Randall's hypothesis. (author)}
place = {United States}
year = {1974}
month = {Jul}
}
title = {Eliminating the human element and the drudgery from control-rod calibrations}
author = {Ruby, L, and Wang, H -K}
abstractNote = {The Berkeley TRIGA Mark III Reactor has three distinct reflector arrangements, depending on the position of the core in the pool. The control rods must be calibrated in each position, making 12 rod calibrations required, in all. To eliminate the human element and the drudgery involved in this repetitious task, a computer-assisted semi-automatic method has been devised to perform the necessary period methods, and to produce the resultant rod-calibration curves. The method is based on the use of a signal from the linear-power-channel recorder to feed a voltage comparator which generates a pulse at a preselected voltage 'B' and also at '1.50V'. The 2 pulses are used to start and stop pulses for an electronic timer, which easily measures the time difference to 0.01 second. The comparator actually consists of two such pulse-pair generating circuits, so that 2 measurements of t{sub 50} can be obtained on each range of the linear-power channel. Before the comparator is used for a series of rod calibrations, the voltage discrimination levels are checked with a precision voltage source to verify that they are set at 3.50, 5.25, 6.00, and 9.00 volts. Corrections in the discrimination levels can be made by means of front-panel potentiometer adjustments. As voltage is gradually increased past each of the pre-set discrimination levels, a panel light comes on, indicating that a pulse has been formed. The comparator circuit also accepts a reset command from a push button held in the hand of the reactor operator, which command is then converted into an electrical reset signal for the electronic timer. The system provides non-prejudiced measurements for t{sub 50} as short as 5 seconds, with no concern about pen lag. The only manipulation of the data is to determine the best value of t{sub 50}, which is done by averaging those values which agree to within 0.1 second. The program ''RODCALN'' is used to calculate the rod worth remaining (in dollar units) versus control rod position, and a RODCALN provides both a table and a plot of the results. From an input of t{sub 50}'s obtained over the whole rod length, the program determines the reactivity for each increment of rod distance by a continuous function which approximates the table of Inhour solutions prepared by GA. The plots are of sufficient detail to be used in routine reactor operation for reactivity determinations. The results obtained with this program have been compared with Randall's contention that a consine-squared function is an excellent fit to the data for the differential worth of a control rod. The close agreement confirms Randall's hypothesis. (author)}
place = {United States}
year = {1974}
month = {Jul}
}