AoA RF switch - terminated/unterminated

[Stetina]

Hello,

does anyone tried terminated vs unterminated RF switch in AoA / AoD application? According to forum, some people use terminated switch, some don't.

Also, there should not be any DC level, but someone form nordic recommended to use DC blocking cap:

https://devzone.nordicsemi.com/f/nordic-q-a/57001/we-make-a-board-for-nrf52811-drection-finding-can-it-work

My problem is that i'm getting wide spread of phase shift from the same place, so im looking for the origin of that issue.

Thank you,

Hynek Stetina

[Dmitry]

Hi Hynek,

do you get different phase shift in static environment (no people walking near antennas)? Maybe you could show some IQ waves you got?

Stetina said:

Also, there should not be any DC level, but someone form nordic recommended to use DC blocking cap:

There will always be some DC level, if the switch IC requires blocking caps, you have to use them.

[Stetina]

Yes, i see phase shift in static enviroment. Even in the middle of the garden, where there should be not much reflections.

Here is how the IQ data look like:

Here is the gif from data processing. Im able to compensate all data and don't see there any reflected signal.

https://gfycat.com/jealousremarkabledaddylonglegs

Legend from bottom:

-I,Q, I*Q from reference period

-phase from measuring slots

- after freq compensation

- median between antena switching (ant1 vs ant2 vs ant1....)

- arrow with phase sift from the center of measuring slot (3 samples: center sample+-1)

Our current RF switch in unterminated BGS13SN8E6327 without blocking caps.