AnsweredAssumed Answered

de-embedding

Question asked by e4411a on Dec 4, 2007
Latest reply on Dec 10, 2007 by daras
Hi,

I want to test an amplifier under different temparatures in a thermal chamber. After calibrating the PNA at the network analyzer end ,I connect two cables to the amplifer i/p and o/p ports in the thermal chamber and connect them to the network analyzer ports. These are sucoflex phase stable cables and the loss is 0.5dB each and I also have the s2p files that I use for de-embedding the cables.

I have the following questions.

1. When the amplifier is in the linear region, and when I check the results with de-embedding on in the PNA and without de-embedding, I get a difference of 1dB in the S21, that clearly shows that the gain of the amplifier gets reduced by 1dB because of the cable at the i/p and o/p.

2. However when I test the amplifier in the non linear region at 2dB and 3dB compression, I do not see this behaviour. The difference between the results in the S21 measurements with de-embedding and without de-embedding is not 1dB. Is this how de-embedding would work, since the amplifier is in the nonlinear region?

Pls clarify.  

Outcomes