AnsweredAssumed Answered

GCA Application

Question asked by nickmmw on Oct 9, 2011
Latest reply on Oct 12, 2011 by Dr_joel

We would like to measure AM/AM and AM/PM using the PNA-X's GCA application -

(if anyone happens to know of a document that describes, in-depth, the GCA application, it would be greatly appreciated (other than the PNA-X online help file.))

We would appreciate any information on the following:

1. If I understand correctly, the GCA application walks you through three required calibrations:
1st cal - Source-power calibration: to enable the PNA-X to measure absolute-power accurately at the port-1 reference-plane
2nd cal - RCVR calibration: the PNA-X uses the source-pwr calibration to calibrate the b2-RCVR --- this procedure calculates the loss-between the port-1 reference-plane and where the b2-RCVR measures absolute-power. This is done at 1-pwr level only, since the PNA-X's rcvr are accurate to within 0.05 dB
3rd cal - Relative calibration: either full-2-port vector corrected cal or response cal.
Question: I do not understand why or how this final calibration step is used. I can see why it would be needed if relative-phase/AM-PM is needed - it would allow an accurate measurement of phase(s21) as a functin of Pin. However, say we are only interested in AM/AM - then is the final relative-calibration step necessary?

2. If these measurements are obtained on-wafer, and say we have accurately obtained the S2P-data of each of the two-probes, can we load these s2p files as 'embed' blocks into PNA-X ? Will the source-power-cal and RCVR-cal consider the losses of these 'embed-blocks' ?

3. Are there any detailed application notes/documentation describing the math of the GCA application's calibration steps - especially, the relative-calibration? (Any other application notes than: '1408-8 - Linear and Gain Compression Measurement'.)

4. In order to sweep the source-input power, is it correct to assume that the source-power attenuators are engaged if we have to step past a 10 dB step ? For example, if our source-power sweep is on greatter than a 10 dB range - say a 30dB range - then would we have to have 3-different calibration channels for maximum accuracy - each channel corresponding to an individual source-att setting ?

5. Can the PNA-X output a P2D and/or S2D file using the calibration performed with the GCA application ?

6. Are the PNA-X's receivers more accurate in measuring absolute power than the previous genration's PNA receivers ?  On the PNA, I recall having to do a source-power calibration by fixing the frequency and then sweeping power during the source-cal. However, the GCA application in the PNA-X only calibrates at a single-power level.

7. If we are not intersted in the harmonics, then is it poissible to obtain a 'fundamental-tone X-Parameter' using the calibrations performed with the GCA application ? Or - does this require a different PNA-X application or set of steps ? (in this case, we will not be needing the harmonic-phase reference calibration for the X-Parameters).

Thank you for your time,