Since 2019, I’ve been part of a fantastic standards working group helping draft the standard IEEE P2855 “Recommended Practices for the Electromagnetic Characterization of Cable/Connector Assembly Shielding Effectiveness in Frequency Range of Direct Current (DC) to 40 GHz.” This group has consistently been meeting monthly, and we’re just getting ready to submit a draft for balloting—which gives you an idea of the speed of standards development, even when everything goes very well and you have a very conscientious working group chair, secretary, and membership.
Reviewing the draft got me thinking about transfer impedance (Zt) and shielding effectiveness (SE). When I was just starting out, someone told me they’re reciprocals of each other, that SE was just 1/Zt. That’s directionally true—when you have a very good shield, Zt will tend to be very small and SE will tend to be very large. However, it isn’t technically true, since Zt and SE are measured in quite different ways.
While there are many ways to characterize the quality of a shielding material or configuration (many of which will be captured in the future IEEE 2855 standard!), Zt is often measured with some variation of a triaxial fixture, which you can find described in IEC 62153-4-3. This test setup runs the cable under test (CUT) down the center of a long conductive tube. Current is induced to flow on the shield of the cable, using the conductive test fixture to return. Voltage is then measured between the inner conductor of the CUT and the CUT shield. Transfer impedance then is defined as:
Zt = V/I
Where I is the induced current (Amps) and V is the measured voltage (Volts). You can see that when you have a really good shield, even if you have a lot of current flowing on the shield, you’ll only have a small voltage induced on the interior. Thus, with a good shield Zt tends to be very small, often in the mΩ range.
In contrast, shielding effectiveness is determined by exposing the CUT to a particular stimulus, both with and without the shield in place, and then expressing the difference between the two cases in dB. There are many ways to do these measurements: in a semianechoic chamber (ALSE), on a bench, in a test jig, etc. Possibly, my favorite is to take the measurement in a reverb chamber (RC), as seen in the test method IEC 62153-4-21.
Imagine you have a CUT and you expose it to a certain stimulus, whether by radiating it in an ALSE or RC or by directly injecting current onto it. You can measure the power picked up by the inner conductor in both the shielded and unshielded cases, at which point shielding effectiveness is:
SEdB – Pu – Ps
Where Pu is the power picked up in the unshielded case (dBm) and Ps is the power picked up in the shielded case (dBm). If you have excellent shielding in place, you should pick up dramatically more power in the unshielded case than the shielded case, leading to SEdB being a large, positive number.
When doing one-off measurements, you’ll often find SE to be easier to measure. Shield manufacturers have fixed test setups for characterizing all their samples, so they tend to advertise their shield material quality in terms of its Zt. It’s important to keep in mind what number you’re looking at and how it was measured. While a large SE or a small Zt can both indicate a good shield, they’re not directly related.
