Downloaded from https://www.everyspec.com MIL-STD-1311D 6.2 Capacitive divider circuit. The capacitive divider circuit is particularly useful at very short pulse durations. Its use for longer pulse durations is restricted by the amount of pulse droop which can be tolerated. a. The divider is made up of a capacitor C 1 developing a high voltage in series with a capacitor C 2 developing a low voltage; the divider ratio being approximately inversely proportional to the ratio of these values. The divider network is connected to the CRO through a series matching resistor R, a coaxial cable and if necessary a blocking capacitor C 3 .
The input resistance of the CRO should be two or three orders larger than R. The capacitor C 1 has a value usually in the range of 1 to 10 pF and should be shielded to avoid stray pickup. The dielectric may be ceramic, resin, oil, or vacuum as required by the working voltage. The capacitor C 2 should have very low inductance and a safety spark gap should be connected across this capacitor. b. The divider circuit may be calibrated using other methods described (for example, resistive divider and peak voltmeter within its range) using a suitable pulse duration and where the errors due to the known characteristics of the methods are minimized. c.
For pulse durations down to about 0.05 microsecond, the length of the coaxial cable should not exceed that which has a two-way transit time of about 10 percent of the pulse duration. For pulse durations below about 0.05 microsecond the cable length becomes impractically short and the divider may be mounted directly on the CRO. 6.3 Balanced divider circuit. The balanced divider circuit is particularly useful when the pulse waveform is to be faithfully reproduced in detail. a. The divider consists of a high-voltage section R 1 and C 1 , and a low-voltage section R 2 and C 2 , and the CRO matching components. The time constants of each section, including the effects of strays, shall be equal.
Usually, the value of R 2 is made equal to the characteristic impedance of the coupling line between the divider network and the CRO. In this case, the resistance R 3 in figure 4306-6 is zero, and R 4 = R 2 . This gives a division ratio 1 2 C C R R = . The time constants 2 1 2 of the two circuits are R 1 C 1 and 2 2 C R , respectively. When the required division ratio is greater than 2 1 2 R R , a smaller value of R 2 may be used, and the match condition for the cable restored by inserting a nonreactive series resistor R 3 , 2 where R 3 = Z o - R 2 . In this case, R 4 shall be made equal to Z o , and the division ratio is:
4 2 R R R R R R R R R + + + + ) ( ) ( 4 3 2 4 3 2 1 The time constants are then given by R 1 C 1 and R 2 C 2 where R 2 = ) ( ) ( 4 3 2 R R R R R R + + + 4 3 2 It is difficult to use this circuit when R 2 exceeds Z o of the cable. b. When assembled, the divider shall be checked for division ratio and pulse distortion due to possibly overlooked stray effects. This may be done using a calibrated square-wave generator whose rise and fall time characteristics are similar to the pulse to be measured. Alternatively, calibrated signal sources in the frequency range dc up to twice the reciprocal of the pulse duration may be used. NOTE:
In practice, it may be useful to isolate the divider network from the cable and the CRO by using a value of R 3 several times greater than Z o , but the cable is then not matched at the sending end. METHOD 4306A 4306-4 25 March 1970