nanoVNA-H – measure 40m low pass filter for WSPRlite flex describes measurement of the response of a filter.
The filter is of a type that depends on its source and termination impedance for as designed performance.
The article mentioned the use of a 10dB attenuator on the nanovna-h Port 2 for the purpose of improving the accuracy of the load impedance for the filter.
Like most low end vnas, the nanoVNA Port 2 VSWR or Return Loss is not wonderful, not as good as needed for some types of measurement. Return Loss can be improved by placing an attenuator ahead of the port. The nanoVNA-H v3.3 already includes an attenuator on the PCB, and the nanovha-H v3.4 increased that attenuation by about 5dB to improve Return Loss by about 10dB.
In my own case, I am using a nanoVNA-H and upon measurement of |s11| (-ReturnLoss) I determined that it needed to be improved by 20dB for my use so I purchased and installed a 10dB attenuator semi permanently on the Port 2 connector.
Above, the 10dB attenuator is semi permanently attached to Port 2 and also serves the purpose of a connector saver. There is a connector saver semi permanently attached to Port 1.
So, to measurement of |s11|, first step is to carefully calibrate the vna paying particular attention to torquing connectors properly (it can make a difference). The outboard 10dB is attached for the calibration process, so the calibrated inputs in this case are the male side of Port 1 connector saver and the male side of the 10dB attenuator on Port 2. The |s11| measurement of Port 2 is done with the same connections as the THRU part of the OSLT calibration process, we are reading |s11| of Port 2 as an input.
Above is a plot of |s11| from 0.1-300MHz. I don't really regard this as a very useful instrument above 150MHz (for many reasons), and the ‘improved' |s11| plot is one of the reasons. The effect of the 10dB attenuator is to decrease |s11| by 20dB, or to increase ReturnLoss by 20dB (ReturnLossdB=-|s11|dB)… so you can see that at 300MHz |s11| is -27dB and would be more like -7dB without the 10dB attenuator. For most purposes |s11|=-7dB, ReturnLoss=7dB, VSWR=2.6 would not be regarded as a good termination.
Above for convenience is the Return Loss plot.
Adding an external attenuator is not without cost, it reduces the dynamic range of Port 2, so choice of the amount of attenuation is a compromise between good measurement accuracy that flows from high Port 2 Return Loss and noise degrading measurements. The nanoVNA is a very diverse thing, to asses the need and size of an external attenuator, make the Port 2 |s11| measurement of the bare port and based on that and your own applications, you might choose an appropriate attenuator.