Background
In a conventional vacuum capacitor, such as the CSV4-1400, the variable electrode is attached to one end of the variable shaft. The shaft itself is constrained to the body of the capacitor by a tight fit to the alignment bearing. When the capacitor is adjusted away from minimum capacitance, the shaft slides through the guide bearing, extending the variable electrode into the fixed electrode. This movement sets up the variable shaft as a cantilever type support of the electrode (Figure 2).

Shock impulses that have a component orthogonal to the axis of the shaft will impart a bending moment on the variable shaft assembly. Smaller impulses will cause a temporary change in capacitance due to the elastic deformation of parts during the acceleration event. After the event the elastic deformation relaxes and no permanent change in electrode alignment is realized. At some magnitude of impulse, however, the stress from the applied moment will cause a plastic deformation of the weakest part of the rigid body.

In a conventional vacuum capacitor this plastic deformation normally occurs in the copper seal that sits between the bearing and the ceramic body, and causes a permanent change in the alignment of the electrodes. This change in alignment can be either toward or away from the ideal axis of the high voltage capacitor. Alignment changes toward the ideal axis will reduce the value of capacitance while increasing the voltage withstand gap between the electrodes. Conversely, alignment changes away from the ideal axis will increase the value of capacitance and reduce the voltage withstand gap between the electrodes.

Test
Five capacitors were tested, at 20, 30, 40, 50, and 60g respectively. Each was subjected to three hits at the specified level. The variable end of the capacitor was bolted to a thick phenolic disc. This disc was clamped into a rigid test fixture, which was bolted securely to the working table of the shock test rig (Figure 1). The fixed end of the capacitor was left free to simulate a typical mounting scenario in a tuner. An accelerometer was mounted to the shock table away from the unit, to simulate a shock of the specified value to the case of the tuner. Shock levels were monitored on a Tektronix TDS 320 oscilloscope.

All units were set at C(max), with the shock plane orthogonal to the variable electrode axis (worst case scenario). The capacitance was measured just prior to, and just after each drop to measure the change in capacitance from the individual shock event.

Results

All capacitors showed variable electrode displacement from shock impulses of 30g and greater. Results were consistent within each shock magnitude group, and increased linearly with increasing shock magnitude (Figure 3). The graph of Figure 2 shows the average change in maximum capacitance for an individual impulse, with repetitive impulses causing an accumulation of change (i.e. three hits at 60g = ?C of 4.8pF). After surviving the three hits, each capacitor was voltage tested. All units passed 60Hz high potential testing at 3.5kVp (specification is 3kVp).

Conclusion
Based upon these results, a CSV4-1400 capacitor is able to withstand #3 60g shocks without voltage degradation.The CSV4 model of capacitor was chosen for this test because it represents the worst case test sample of all Jennings capacitors. The other Jennings capacitor families and models will demonstrate better results when tested for shock durability.