Purpose of Investigation
CISPR 22 (1997) requires the measurement of common mode voltages on Information Technology Equipment (ITE) cabling to determine conducted emission levels over the 150 kHz to 30 MHz range. Two techniques that do not require breaking into the insulation of the cable under test to make galvanic contact with wires under test are a Capacitive Voltage Probe and a Capacitve Clamp. This application note compares making common mode voltage measurements with these two instruments in a simple and controllable test setup.

CISPR 22 Annex C.1.3 provides the guidance for making this voltage measurement, and indicates that the common mode voltage measurement is to be made no more than 80 cm from the Equipment Under Test (EUT). The length of the cable under test from the voltage measurement location to the Auxiliary/Associated (AE) can be any length. The effect on the
accuracy of the measured common mode voltage due to uncontrolled impedance/cable length resulting in a voltage standing wave is addressed in the measurements shown in this application note.

CISPR 22 requires the common mode capacitive loading to the ITE cable under test by the voltage measurement technique to be less than 5 pF. Measurements were made to determine the common mode capacitive loading of the Capacitive Voltage Probe and the Capacitive Clamp to the cable under test.

Voltage Measurement Techniques
The Capacitive Voltage Probe (CVP) used for these measurements was a Fischer Custom Communications, Inc. Model F-CVP-1. The cable under test is placed through the cylindrical aperture of the CVP.

The Capacitive Clamp (CC) used for these measurements was an EM Test Model HFK. This is the standard Capacitive Clamp commonly used for injection of CW and fast transients into cable bundles. To use the CC as a voltage probe requires the use of a traditional high impedance, low capacitance voltage probe. The active FET Voltage Probe (FET VP) used in conjunction with all measurements made with the Capacitive Clamp was a Cal Test Model CT2865 which is rated at 750 MHz bandwidth with 2 pF of loading. The test setup used allowed voltage measurements to be made with the FET VP without the CC.

Test Setup
Test setup is shown in Figure 1. A Hewlett Packard 4396B configured as a Network Analyzer was used to provide a CW stimulus from a 50 W source impedance. The test load was a 147 W resistor located at the end of a 2.7 meter length of #14 wire (insulated) mounted over a 30 cm wide copper ground plate. The 147 W was selected to be consistent with the 150 W load cited by CISPR 22. The 2.7 meter length was chosen for two reasons. First, this length allowed standing waves to develop. Second, the wire length is a conceivable length in a laboratory environment. Voltage measurement positions denoted A, A’, B, C and C’ as shown in Figure 1 were used for the measurements in this document.

CVP Measurements
The voltage on the #14 wire was measured at positions A, B, and C as shown in Figure 1. The results of the measured voltages at A, B, and C for this set of measurements are shown in Figure 3.

These measurements show the effect of the standing wave above a few MHz generated on the #14 wire due to the mismatch of the 50 W source impedance and the 147 W load. The difference in measured signal at 30 MHz is about 4 dB from position A to position C.

FET VP Measurements
The measured voltage at positions A’, B, and C’ as measured with the FET VP is shown in Figure 3. The FET VP and the CVP measurements (Figure 3) are about 2 dB apart.

Capacitive Clamp Measurements
The data in Figure 3 have been corrected only for the FET VP calibration curve shown in Figure 2. Based on the theory of the CC and FET VP operation cited above, only the FET VP should affect the voltage measurement.

Capacitive Loading
To determine how the capacitance of the #14 wire over the ground plane was loaded by the CVP and the CC, the Hewlett Packard 4275A LCR meter was used to measure the capacitance for these various physical configurations. The HP 4275A was placed at one end of the #14 wire (position A in Figure 1). At the other end of the #14 wire, the 147 W load was
removed, and the #14 wire open circuited.

The CVP (including foam insert) was centered in the middle of the 2 meters of #14 wire. The total capacitance was measured with the HP 4275A at 100 kHz, 1 MHz, and 10 MHz. The CVP was then removed, keeping the #14 wire in the same physical position. The capacitance was then re-measured.

These measurements were repeated for the CC, keeping the #14 wire in the same relative position dictated by the CC size. In addition, the capacitance measurements were made with and without the FET VP connected to the CC. The output of the FET VP was loaded with a 50 W termination, and the low side (shield) of the 50 W termination was connected to the ground plane.

These capacitance measurements are shown in the Table below. It can be seen that the CVP generally adds 3 pF of capacitive loading to the #14 wire. The Capacitive Clamp without the FET VP adds 25 to 31 pF to the #14 wire. With the FET VP attached, the CC adds 28 to 32 pF to the #14 wire. The CISPR 22 requirement is for the VP to be used in conjunction with the Capacitive Clamp to have less than 5 pF loading from the wire/bundle under test to ground.