ezRA - How do you do reference mode calibration?

[Tom Nathe]

After reading through the documentation and watching the videos, I'm more confused than before as to how ezConRefMode is set up and utilized.

I can use a loading resistor, so that would be setting #10, which is also the default, but I'm seeing how that option is used in conjunction with the samples being pulled off of the antenna. 

At the moment, I am using the frequency offset as the reference (1423.4050 MHz) and using RefDiv as the result. That graph, along with RefSub, does display an odd graph that isn't like anything that I've seen in the documentation.

I've included screenshots of what I'm seeing. I would like to know just what I am doing incorrectly here.

Thanks in advance!

[tedcl...@gmail.com]

Hi Tom,

I see 2 subjects (RefDiv and RefSub, then ezConRefMode) in your questions above.

1.

ezCol and the spectrum display (RefDiv and RefSub).

I added the top left spectrum display to the ezCol collecting program to provide live reassurance that the data being recorded to file was worthy.

But the standard spectrum had 3 big humps and maybe a tiny bump somewhere near the middle that was the Galactic hydrogen.  Hard to see.

So I added the   ezColYLimL   parameter to change the spectrum plot's vertical limits, to allow zooming in on the top part of the plot to see the interesting Galactic hydrogen bump.  It helped a little.  You can change the   ezColYLimL   parameter live on the graphics screen (with perhaps a slow response to your changes).

I wanted ezCol to focus on collecting data as fast as it could, but what could I do to enhance the spectrum plot without  distracting a lot of CPU cycles ?  So I added the RefDiv display choice to divide the most recent Ant sample by the most recent Reference sample.  And I added the RefSub display choice to have the most recent Ant sample subtract the most recent Reference sample.  This helped.  This does not change the data written to file.  I like the RefDiv plot more than the RefSub plot.  These enhance the Galactic hydrogen signal and provide reassurance that the data being recorded to file is worthy.

2.

Later, the ezCon analysis program processes those recorded data files (ezConRefMode).

It is often hard to see the Galactic hydrogen in the raw data.

The AntB signal tries to enhance the Galactic hydrogen signal, without using Reference samples.

The AntRB signal tries to enhance the Galactic hydrogen signal, using Reference samples.

ezCon compares the Ant samples to some Reference that does not have a Galactic hydrogen signal.

You can change the ezConRefMode parameter, which defines the Reference samples.

Looking at the Ant samples available, maybe you see that Ant sample zero has no Galactic hydrogen signal.

ezConRefMode of 0 uses Ant sample zero for every Reference sample.

ezConRefMode of -1 uses Ant sample 1 for every Reference sample.

ezConRefMode of -1234 uses Ant sample 1234 for every Reference sample.

ezConRefMode of 2 uses the RawByFreqBinAvg spectrum average spectrum for every Reference sample.

The ezConRefMode default is 10 which, for each Ant sample, uses the most recent Reference sample.

If there are no recorded Reference samples, ezCon uses Ant sample zero as every Reference sample.

I am adding new ezConRefMode options.

ezConRefMode of 5 uses the one ant spectrum at the minimum of AntAvg.

ezConRefMode of 7 uses the one ant spectrum at the minimum of AntBAvg.

ezConRefMode31 uses the one AntB spectrum closest to a requested Right Ascension (quiet like at 11.5 hours).

ezConRefMode30L selects the Ant spectrum beside the Reference spectrum from the same wide bandwidth Raw spectrum (Airspy, SDRplay, Ettus USRP B210).

Helpful ?

---
Ted Cline   N0RQV
TedClineGit at gmail.com

On Saturday, March 21, 2026 at 11:46:00 PM UTC-6 Tom Nathe wrote:

...

[tedcl...@gmail.com]

Hi Tom,

You attached 2 plots.

The RefDiv.png plot looks great.

Yup, there is a hint of some Galactic hydrogen signal in the middle.

I think the RefDiv works better than the RefSub choice.

ezCon does better analysis than either.