RF Glossary
Special capability that allows user to directly connect to receivers. Useful for load pull and high power measurements.
The sweep speed, or the time required to present a complete S-Parameter measurement from the start to the stop frequency on the screen depends on three things. First, the IF Bandwidth determines how long it takes to make a measurement at each point. This is due to the delay through the Finite Impulse Response (FIR) filter which applies the IF Bandwidth to the signal. This delay is approximately 1.5/(IF Bandwidth) per point.
Secondly, the number of measurement points obviously affects the sweep speed and thirdly the type of calibration in place. A full two port calibration must operate on all four S Parameters in order to calculate any single corrected S Parameter. If a full two port calibration has been performed then the VNA must sweep twice, once in each direction in order to obtain all four required parameters. This doubles the time required to present a sweep on the screen. A full four port calibration would take four times as long.
Additionally, each sweep requires a small setup time, usually a few milliseconds.
So Sweep Time is approximately Setup + CalFactor * NP * 1.5/IFBW
Where NP is the number of points and CalFactor is 1 for no calibration or 1 port, 2 for 2 port and so on. and Setup is perhaps 1 to 5 mS.
The trace noise of a VNA measurement is related to the signal to noise ratio of the measurement or the distance between the reading and the noise floor at a particular IF Bandwidth. If the Dynamic range of the VNA is given as 140 dB at maximum output level and 10 Hz IF Bandwidth and the user makes a measurement at those settings at -130 dB then the signal to noise ratio is 10 dB.
If the measurement is made at a more convenient 10 kHz IF Bandwidth for faster sweep speed then the dynamic range is reduced to 110 dB ( 30*Log(10K/10) ) . Now if a measurement is made at -100 dB then the signal to noise ratio is again 10 dB.
That noise is 10^(-10/20) or a ratio of 0.316 RMS in linear terms. So a normalized unit vector sees a 0.316 vector added to it with random phase. Worst case that’s 0.684 to 1.316 so the trace noise is -3.3 to +2.4 dB RMS for this example.
This is why the Dynamic Range specification of a Vector Network Analyzer is so important.
Scattering parameters. How much RF signal has passed through the DUT and at what phase.
The device being tested by a vector network analyzer.