standard practice to test such capacitors at that frequency rather than at the more readily available 60 cycles. To meet this requirement, a modification of the 60-cycle Type 1611-A Capaci
Figure 1. View of the Capacitance Test Bridge (left) with 120-cycle oscillator (center) and Unit Variable Power Supply (right) for furnishing the d-c polarizing voltage.
tance Test Bridge has been available for some time.1 The modification as shown in the schematic diagram (Figure 2), consists primarily of the addition of terminals for connection of an external generator and of providing 120-cycle tuning for the internal detector. Switching is provided so that either the standard circuit or the special configuration can be used.
Because the input impedance to the bridge varies between 1 ohm and 1000 ohms, most available audio oscillators are not capable of delivering enough voltage directly to the bridge over its entire range for adequate sensitivity of balance. A multiwinding impedance-matching transformer is frequently required to deliver adequate energy to the bridge, particularly for the 1000 multiplier setting, and, in any event, a transformer is required to isolate the a-c source from the d-c polarizing voltage.
A recent modification of the Type 1214-A Unit Oscillator, the Type 1214-AS2 120-cycle Oscillator, has a power output approaching the maximum that the bridge arms will withstand. This inexpensive oscillator provides the 120-cycle source, the impedance matching, and the isolation. An output transformer is provided, tapped to provide optimum match to the bridge for each of the four applicable multipliers. A panel switch selects the proper tap, and the engraving corresponds to the multiplier markings of the bridge, XI, X10, X100, and X1000.
For the measurement of electrolvtic
Figure 2. Elementary schematic of the bridge.
capacitors over a range of low audio frequencies, it is convenient to make use of the impedance-matching transformer and switching provided in the new oscillator. Accordingly, a jack is provided on the panel of the oscillator by means of which ail external source can be connected.
The bridge circuit is designed for optimum sensitivity at 60 cycles, and substantially the same performance is obtained at 120 cycles. As the measuring frequency is raised, however, the bridge sensitivity factor decreases and above about 400 cycles is inversely pro-tional to frequency. Figure 3 shows the variation of the bridge sensitivity factor2 S for the different settings of the CAPACITANCE dial. The sensitivity factor is independent of multiplier setting, but the voltage applied to the bridge varies with ratio arm setting, as a Defined as -S =
"Electrolytic Capacitor Testing at 120 Cycles", General Radio Experimenter, Vol. 28, No. 6, November, 1953, p. 8.
Za Zb d, where d is the precision of setting of the reactive balance shown in the following table.
Approx. Voltage (rms) Applied to Bridge
The upper frequency limit for satisfactory operation is thus a function not only of available detector sensitivity and applied voltage but also of the magnitude of the capacitance being measured arid of the accuracy of measurement desired. With the internal detector, measurements can be made at 1 kilocycle on even the highest multiplier with a resolution of about 1%. If higher resolution is desired, a more sensitive detector can be connected externally to the external-filter jack of the bridge. For best results the external detector should have a 20-db discrimination to harmonics and noise.
The dissipation factor range of the bridge is directly proportional to frequency. At 60 cycles the range is 0.6 (60%), at 120 cycles 1.2 (120%) and at 400 cycles 4.0 (400%). This variation is compatible with the normal tendency of high-value electrolytic capacitors to show dissipation factors rising with frequency as a result of fixed series resistance.
The effective accuracy of dissipation factor measurement decreases at the high frequencies. In addition to the problem of residual phase-angle errors within the bridge, the problem of making a satisfactorily low resistance connection to the capacitor under measurement is a serious one. For example, at
Figure 3. Variation of bridge sensitivity with frequency for different settings of the CAPACITANCE dial.
10,000 juf and 1000 cycles a series resistance of 0.001 ohm in lead or connection produces a dissipation factor of 0.6 (60%). For such an extreme case the limitation on accuracy is external to the bridge; for less extreme combinations of frequency and capacitance the accuracy of dissipation factor is ±2% of dial reading ±.0005 (//60).
The a-c voltage applied to the capacitor under test is always somewhat lower than the voltage applied to the bridge, shown in the table above, since the ratio arm is in series with the unknown. The d-c polarizing voltage should normally be greater than the peak value of a-c test voltage. The voltages applied by the Type 1214-AS2 Oscillators are safely below ordinary voltage ratings for any given range. If, however, reduced test voltage is desired for capacitors of very low d-c rating or for any other reason, an adjustable resistor may be connected (in series or shunt) between the oscillator and the bridge to set to an arbitrarily specified level.
FREQUENCY IN CPS
FREQUENCY IN CPS
Was this article helpful?