Has the HMC849A RF switch changed in performance recently?
The reason for the question: its S-parameter model has changed dramatically since 2017. Below is a plot for the HMC849A S-parameter model downloaded in 2017 versus a plot for supposedly the same model downloaded this month in 2018. As a result of the new 2018 S-parameter model, the simulated VSWR at the RF input for a board that we're currently designing is much greater than it was before using the older 2017 model.
If the HMC849A part's performance hasn't changed, what is the reason for the S-parameter model change? Was the original model in error, or is the current model possibly now in error?
Plot for HMC849A S-parameter model for Vctl=5V, Ven=0 downloaded in 2017:
Plot for HMC849A S-parameter model for Vctl=5V, Ven=0 downloaded in 2018:
I'm also attaching the S-parameter models for the HMC849A that we downloaded in 2017 and 2018 to this question.