AnsweredAssumed Answered

Two Port VNA Measurement Setup

Question asked by chard on Sep 9, 2011
Latest reply on Sep 10, 2011 by Dr_joel
I'm measuring a passive two port network using an E5071B and performing a SOLT calibration with an 85033E cal kit.  I'm using some good Gore cables between the VNA and DUT.

The DUT is a low loss network where loss is defined as Loss=|S21^2|/(1- |S11|^2).  Measurements are made over a wide band.  Over this band, the DUT has several resonances where it is alternately matched well (wrt 50 ohm) and anti-resonances where it's highly reflective.  My problem is that I'm seeing some small ripples in the calculation for Loss that I speculate should not be occurring.

My question is whether the ripples might be a measurement error (or increased uncertainty) when the DUT is swept through the highly reflective state.  Would it help if I added a couple attenuators at the end of the Gore cables just prior to the DUT on the input and output?  If this is a viable technique, can you point me to an app note that explains why.  Are there any other measurement techniques I should consider when measuring a low loss, highly reflective network?