RF circulator / front end question

Projects
1

I’ve been doing some work on the RF front end for the LoRa modules in my wind monitoring system.

I’ve sourced some really nice waterproof 4Mhz cavity filters.
https://www.aliexpress.com/item/1005004141233710.html

These are not directional and have an insertion loss of between 1 and 2 dB

cavity filter passthrough
cavity filter passthrough1920×1080 173 KB

I was doing some research and came across this setup

two circulators
two circulators913×527 109 KB

I thought that this would suit my needs without the Power Amplifier and with the cavity filter in place on the Low Noise Amplifier. The idea is that this should give a lower insertion loss on transmissions as they do not go through the filter but with properly filtered reception. I also had a couple of circulators in stock.

So I made this:

IMG_20241206_113923_1
IMG_20241206_113923_11200×1600 181 KB

But the results are not what I expected…
The transmission path is great with a flat insertion loss of 0.5dB but the return path is not what I expected!

return path
return path1920×1020 45.9 KB

It looks like the circulators are leaking around the filter. I tested them with a 50 ohm load, and they work as expected.

It was an interesting test, but sadly not successful!
What have I misunderstood?
Thanks

2

Reverse transmission path has no impedance match + using narrow bandwidth circulators with poor isolation.

It will be more obvious on the phase diagram or smith chart.

1 Like
3

Thanks Ni

4

I had those circulators in stock.

I’ll see if I can find a couple of reasonably priced wider bands ones.