Network Analyzer shows positive S11 on passive PCB

A freshly calibrated HP8753ES Network Analyzer measured a positive S11 on a passive PCB. The blank PCB will be used for an amplifier, and is composed of input and output matching networks. The only mounted components on the board are 0402 capacitors and SMA connectors. What am I missing?

Can you show the polar display of S11 (Smith diagram)?

Your fixture and adapters must be included in the calibration.

Then insert DUT for accurate readings.

Thank you SunnySkyguy. I calibrated the VNA to the end of the test cables, and I have SMA connectors soldered directly to the PCB. Do I need to include the SMA connectors in the calibration? If so, how?

Looks like a wrong open/short calibration.

Are you seeing positive |S11| using the factory calibration, or user calibration (with passive networks attached)? If the former, then something is either wrong with the instrument or the factory calibration.

I used the Agilent 85033C Cal Kit. When I finished the Cal procedure, I re-checked the short, open and load terminations on the VNA's Smith Chart, and they were correct.

So I am puzzled by the positive S11 response.

It's matter of calibration. Have a look to this:

http://anlage.umd.edu/Microwave%20Me...980-2784EN.pdf

Sure it matches because you used them for calibration, but I think it's not a broken standard, otherwise s22 would be strange too. Try to change cable and connectors of port 1.

Thank you all for your help. I tried a different Cal Set and results were substantially closer to predicted (simulated) results.