When I use the Port Extensions in PNA, I want to know how to get the DC loss term in the port extension setup ? In Dr.Joel's book ,Chapter 9,equatin 9.23 and 9.24 don't have this DC loss term, if the term is added into the equations, how to use the the least squares fitting over the current span of the measurement in chapter 9.4?
Thanks for your reply,Dr.Joel. I still have a question.
“”“In APE, the loss term is extracted as the Y-axis intercept.”
Does the Y-axis intercept mean log(A) in equatin 1.81 of your book?
No, this DC offset is applied before Equ 1.81. The DC offset is applied to the data, and then the data is fitted to the eqn to get the loss vs. freq terms. It is essentially a combination of skin effect loss plus dielectric tangent loss.
There can be a constant loss that goes from DC to high frequency in addition to the loss above. Or more particularly to the case where a fixed attenuation is also in the path. The port-ext with loss applies the offset first, then the eqn.
Dr.Joel. I have another question.
In the above, you replied that "DC offset term is needed for the case where there is some fixed attenuation in the path, in addition to cable loss."
I use the APE, I find that DC offset term is always zero when the loss at the start frequency is relatively big(for example -1dB).Why?
Hope you give some help!
thanks
DC offset term is needed for the case where there is some fixed attenuation in the path, in addition to cable loss. For example, if there were a 3 dB pad and a cable. The PNA applies the loss vs. freq equation and then adds the DC offset as a fixed offset.
In APE, the loss term is extracted as the Y-axis intercept.