Insertion Loss, Mismatch Loss, Transmission Loss
A correspondent asks about the effect of RCA connectors at HF on his proposed noise bridge. The question is very similar to that considered at Exploiting your antenna analyser #13 for UHF series connectors.
I have made a simple measurement of a BNC 50Ω termination (to check calibration) then inserted a BNC-RCA and RCA-BNC adapter.
Measurements of input impedance only for such an electrical short transmission line will not give useful data for determining TransmissionLoss which is the result of conversion of RF energy to heat. The measurements do give ReturnLoss and given that InsertionLoss=MismatchLoss+TransmissionLoss, they set a lower bound for InsertionLoss.
To jump to the chase, it also has a Smith chart plot up to 200MHz that suggests it might be well modelled by a TL segment of 30-35Ω.
Above is a plot of VSWR when Zref is adjusted for the flattest response from DC, and it can be seen that with Zref=33, response is quite flat to 200MHz.
If the AIM software was any good, I could use the refer to antenna tool to backout the length of 33Ω line and adjust that length until the result matched the actual load (50+jΩ). unfortunately, one of the many defects of the AIM software is that it changes Zref to match Z used in the refer to antenna facility. Locking these two parameters together prevents full use of the facility.
I can find the length of 33Ω line to transform the impedance as shown on the plot at 200MHz, and it is 150mm electrical length. Physical length will be reduced by the effect of the solid dielectric.