Booster Magnet Circuit Measurements2/18/99
The period at the end of January and beginning of February 1999 was a time of struggle for Booster beam operation with constant tuning necessary and sometimes even unsuccessful to prevent large beam loss early in the cycle. In addition to the tuning, many hours were spent trying to identify the cause of the problem.
Ultimately, it became apparent that the observation of beam motion of varying amplitude at different locations around the ring, as large as ± 10 mm, was the clue. The motion was present only in the horizontal plane. The frequency of the beam motion was 200Hz at some times, 250Hz at others. Finally, signals of the same frequency were observed on the GMPS power supply voltage to ground monitors.
As the problem was diagnosed, it was found that the slew rate of the power supply voltage program signals in the two galleries were not equal; one gallery ramped up in 5msec while the other took 6 msec. The beam motion and voltage to ground signal at the offending frequencies were greatly attenuated by adjusting the slew rate of the power supply voltage program signals in the two galleries to be equal. During the days of investigation, common mode filter capacitors had been several times connected and disconnected from the magnet circuit. It was finally determined that the offending frequency shifted from 250Hz to 200Hz when the 1uF common mode filter capacitors on each output terminal of each of the four GMPS supplies were connected.
Our lack of understanding of this mechanism and the response of the circuit led to measurements of the ring gradient magnet circuit properties.
The total capacitance of the ring magnet circuit to ground was measured and found to be 6.6 microfarads with all four power supplies switched in and 6.46 microfarads with all four supplies switched out. Note that even with the supplies switched into the circuit the power supply transformer capacitance is not included since the contactor on the transformer secondary was open. The measurement was made by raising the ring to one or two hundred volts with the highpot power supply, disconnecting the supply, and observing the RC time constant of the voltage decay. The value of R was determined by hipotting the ring at different voltages and recording the leakage currents. This showed the normal ring leakage to be a good, constant resistance independent of voltage. R had to be measured separately for the two conditions of power supplies in (135K) and power supplies out (210K).
The troublesome beam motion and the corresponding power supply voltage to ground signals are shown in the first two following slides. Results of impedance and transmission measurements made of the Booster gradient magnet circuit are shown in the remaining slides.