Effect of Stray Capacitance on Testing

We occasionally receive emails from customers who feel their balun is defective because they are unable to replicate our test results. This is due in large part to the effect of stray capacitance picked up by the testing method they use. For Common-Mode chokes used in RF applications, stray test-fixture capacitance (which might be only a few pF) can resonate with the choke's inductance and greatly change impedance measured. As you can see below, this is especially noticeable at the higher frequencies.

Below you will see a pictorial review of how our testing is performed and the methods used to eliminate the effect of stray capacitance. We also show a scan using a non inductive 50 ohm resistor with wire leads and the magnitude of error this typical method introduces in the results.

Test Setup using High Precision 50 ohm Load

  Figure 1

The scan in Figure 1 shows the results using an AIM 4170UHF network analyzer utilizing a high precision encapsulated/shielded 50 Ohm load.

Test setup utilizing a 50 ohm non inductive resistor w/ wire leads

Figure 2

This is the same scan with the only difference being the 50 ohm load now has wire leads. Note the dramatic impact of stray capacitance picked up by the leads of the resistor.  If the model being tested has studs on the sides or top, the longer wire leads required to connect the resistor will increase the capacitance pick up area further exacerbating the resulting impedance mismatch.

For those interested, here is the full layout of our test setup using the AIM 4170 and the VNA 2180.  The VNA is used to measure insertion loss and suppression in db.  We should also note the impact of this capacitance is the most noticeable on 1:1 baluns. The complex winding of other ratios can reduce or eliminate the effect of this capacitance on test results.