Hospital communications issue

[Mike_G_D]

I put the BNC T on the VNA and had N to BNC adapters on the T, no cables.

Hmm, in that case, I either suspect the calibration or the quality of the connectors and/or dummy load.

1) Was a full calibration done at the test frequency using lab standards, open, short, and load?

2) What is the rated frequency range of the dummy loads used on the T.

I can see some reactive component from the T and BNC adapters but the test frequency seems a little low for that to contribute much. Anyway, the real part confirms the overall point: two 50 ohm loads in parallel = expected 25 ohms (or close enough).

-Mike

 

[speedway_navigator]

I didn't look at the T closely, the first one I pulled out of the bag had the center conductor for the long part of the T sticking out on one side, so quality is probably an issue.

I calibrated the device at IWCE with an Anritsu cal kit another instructor had, I don't own one but plan on getting one. I have a hard time spending $300+ for a calibrated open and short. The VNA has not been factory calibrated for 3 years, hard to believe I've had it that long.

>>Anyway, the real part confirms the overall point: two 50 ohm loads in parallel = expected 25 ohms (or close enough).

It is nice to see theory and measurements agree.

 

[Mike_G_D]

I didn't look at the T closely, the first one I pulled out of the bag had the center conductor for the long part of the T sticking out on one side, so quality is probably an issue.

I calibrated the device at IWCE with an Anritsu cal kit another instructor had, I don't own one but plan on getting one. I have a hard time spending $300+ for a calibrated open and short. The VNA has not been factory calibrated for 3 years, hard to believe I've had it that long.

>>Anyway, the real part confirms the overall point: two 50 ohm loads in parallel = expected 25 ohms (or close enough).

It is nice to see theory and measurements agree.

At this frequency and for this purpose, just using an SMA female pcb mount with the through mounting legs soldered to the center pin for a short and another with nothing (preferably with the mounting legs cut off) for the open should be good enough. And if you lack a load, use another with a small (say, 0402 size) 50 ohm surface mount resistor (the tighter the tolerance the better) soldered between the center pin and the outer shield, again, with legs cut off preferably. I did this a lot when lab standards were not easily available and it gets you pretty close.