Why is there a slope of Easy Radio Astronomy (ezRA) ezCol data collection from my radio telescopes?

The screenshot below shows data collection on ezCol from both LRO-H2 Monopole 1 and LRO-H2 Monopole 2 (2 orthogonal polarisations on same cantenna waveguide on same 150cm parabolic reflector used for hydrogen line (1420.405 MHz) mapping of Milky Way. The system is set up to collect every other sample from 1417.405 MHz which is used to calibrate the 1420.405 MHz signal.

The following explanation for the slopes seen in the graphs above was posted by Joergen on SARA mailing list 28/9/2025, in response to my own post that included the above screenshots:

Good effort Andrew!

I think your problem is the Nooelec LNA and its performance when its NOT seeing 50 ohm clean. Have done a lot of experimentation with this LNA and it is sensitive to impedance, ohm value not phase angle. If it is under 50 ohm, your ezCol will show a slope and for higher than 50 ohm still a slope but in the other direction. Your ezCol is set to RefDiv and this is enhancing the slope.

Have 3 Noelec LNAs and they are all behaving similary, its not a faulty LNA.

In your case Andrew, the precise reason is the the LNA gain ripple caused by the antenna inpedance is not 50 ohm at one or both frequencies in play here. 1420 and 1417 will have different slopes and ezCol presents the difference as shown.

In my own case i use a relay and a good 50 ohm load and gets…. the same result.

From my experimentation, VSWR has to be better than 1:1,2 or better than 20 dB return loss to avoid the slope and of course on both frequencies in play. As you can see on my graph i havent reached 50 ohm myself.

A bare cantenna can easilly be 50 ohm and on the reflector, be far from 50 ohm wich is the case for me.

Any experience on how we achieve 50 ohm when we are using the reflector?

Joergen