JA_IFAvg2csv_IFavg2.8_01Jun2501 Right Ascension Problem

[cgd]

If I choose to index by RA, the software does not appear to reset to zero when 24 hr is reached. Has anyone else encountered this? Am I using the latest version? Thanks for replies.

[Alex P]

YES, that is a fact and related to the Graphic software.

IF enough  people start using Rinearn, I may be able to convince the author to add this.

Post some plots !

Regards,

Alex

[cgd]

Alex,

See RA axis

[b alex pettit jr]

=========

! Excellent !

=========

==============================================================================

2D Files :  Post the Raw, Drift, and mx+b Corrected .   

The 2D plots are most important as they characterize the performance of the system,  identify RFI, 

and verify the selection of the low & high CSN points..

=============================================================================

3D File : Try adjusting the Z axis range to improve the color display range and decreasing the freq span to stretch that axis.also

( too much Green ! ) 

===============================================================================

https://github.com/AP-HLine-3D/HLine3D/blob/main/21cm_Step3B_HL3D_Rinearn_Graphics_Rev001.pdf

On Saturday, November 29, 2025 at 11:27:06 PM EST, 'cgd' via Society of Amateur Radio Astronomers <sara...@googlegroups.com> wrote:

Alex,

See RA axis

.

[b alex pettit jr]

BTW, Dec + 40 Deg is a good place to start ... I have lots of test data from that slice of the sky

[b alex pettit jr]

FYI,

this is data from  2 yrs ago at Dec +50 Dg with my 1.2 m WiFi dish system

[cgd]

Thanks for advice and your data Alex.

My current equipment is limited to a Discovery Dish LNA feed adapted to a Nooelec mesh dish which is supposedly "good" for 21cm data collection.

I worked on the 1st image in this post.

 

3D version

2D Charts

[b alex pettit jr]

1) The Drift Looks reasonable ..

2) There are RFI spikes in the data that throw off the Z axis plot autoscaling

3) The high freq spike might be within the range of the default CSN Hi setting.. try changing it to 80%

4) The 0.9 dB re ColdSky Signal Amplitude looks as expected for that  system. 

5) TRY A 24 HOUR DRIFT SCAN THROUGH Dec + 40 Dg.

    (  for future reference : build a Loop Feed s to replace the DiscoDish Dipole Unit ?? )

                                 

             

Only took me ~ 1 yr to get plots like this (smile) the first processing was done in Matlab & then converted to Java

******   EXEC SUMMARY .. WELL DONE ...  except for the RFI, All Looks Good ! 

   >>>>>   (  See final comment at end ) <<<<<

====================================================================================

What I typically do is leave the HL3D window open, process the set at the default 5% & 90%, look at the 2D plot and adjust  the CSN parameters..

Label the files being saved in HL3D by those CSN values to make them easy to sort out later

Alex

==================================================================

On Sunday, November 30, 2025 at 02:18:12 PM EST, 'cgd' via Society of Amateur Radio Astronomers <sara...@googlegroups.com> wrote:

Thanks for advice and your data Alex.

My current equipment is limited to a Discovery Dish LNA feed adapted to a Nooelec mesh dish which is supposedly "good" for 21cm data collection.

I worked on the 1st image in this post.

 

You know .. THIS looks more like a slice through Dec + 60 Degrees ..

( I flipped mine to match your Y scale )  plotted in Matlab

Alex

[b alex pettit jr]

BTW :

[b alex pettit jr]

AND, if you Saved the BC file, it is a *.txt file and can be Edited = you can Manually Remove that Anomaly !

[cgd]

Alex, please elaborate on what the BC file is and how I would go about removing the RFI spikes.

[b alex pettit jr]

Sorry for being so cryptic ..

IF you Save the Background Correction ( BC ) file, it looks pretty much like a data file ..

Plot it in Excel   remove the ( now positive ) Spike ( make the level = to the other nearby values ) , and ReSave for future acquisition 

===========================================================================

[cgd]

Alex, wow your +60 Dec image does look similar to my "supposedly" +50 Dec image. I have very crude pointing capability. Using an Iphone compass for azimuth and an electronic angle finder cube for altitude. I could easily be that much off.  I don't have my dish on a goto mount yet. I have a Celestron AVX equatorial mount, but it is a pain to setup and align. Not weatherproof either. Would have to take it all down if any chance of rain in forecast.

[b alex pettit jr]

You have made some Major Accomplishments in a very few days  

[cgd]

Alex, I collected background data 2 Mhz above 21cm frequency to avoid a dip in my data due to some H which always seems to be  present regardless of where I collect "cold sky" background data. Accordingly, the spikes near 21cm were not removed by my backgound calibration because they were not in the background data.

I may go back to collecting background data by disconnecting the dish antenna and terminating my SDR with a 50ohm terminator.

[b alex pettit jr]

Unfortunately, that will Not Work : the M curve is from the SAW filter of the LNA & so it must be part of the system being corrected

SEND me one of your Background Correction files

=======================================================================

[cgd]

Also, thanks Alex for your patience and help.  Glad I joined SARA and this forum.

[b alex pettit jr]

THIS COMMENT MAY BE INVALID ..

Unfortunately, that will Not Work : the M curve is from the SAW filter of the LNA & so it must be part of the system being corrected

TRY IT

Attach one of your Bkgnd Corr files

Glad to Help 

===================================================================

[cgd]

I believe this to be the background file I used.

[b alex pettit jr]

I don't see the spike in the BC file ??  

[cgd]

No I collected background data 2 Mhz above the 21cm frequency to prevent H present in cold sky from negatively affecting my data collection at 21cm. The spikes near 21cm were not removed because I didn't collect background at 21cm. At least that is how it appears to me. The spikes were troubling me when I was collecting data at 21cm and I thought I might be able to notch them out in sdr# but sdr# doesn't have the notch feature that other sdr software has.

[b alex pettit jr]

I understand that.... since there is a Notch in the spectrum, that indicates there Should be a high + spike in the correction file...

( and I just 'upgraded' to Excel 2024 . Tasks which used to be trivial have become show-stopper today )

=======================================================

[cgd]

I just compared the date on the background file that I sent you and the date on the first data file in the set for the data we are looking at here. The background file I sent you was dated ~ 20 hrs earlier. I distinctly remember creating a new background file for the data we are looking at here so I guess I didn't save it and what I sent you was not used for this data set. Sorry for the confusion.

[b alex pettit jr]

 There's Hope the process will work in the Future.

Can you USE the file you posted ? It may provide a clean background correction for new acqs

[cgd]

Yes I can use it, but if there are spikes near 21cm they are not going to be removed are they?

[b alex pettit jr]

Oh, that was an non  Band_Shifted correction .

Yes, you are correct .. 

[b alex pettit jr]

Try getting a background corr file pointing the antenna towards the ground ?

Also  try the load on the SDR input . will be interesting to see the results.

[b alex pettit jr]

ALSO, an In_Band Bkgnd Corr obtained around RA 12 should work fine also

[cgd]

Thanks for the tip on RA 12 Alex. I will also try terminated backgrounds again if necessary but I, like you I believe, feel background data is best collected with all the equipment in the data collection train in place.

[b alex pettit jr]

The LNA Must Be Included in the Background Correction ..

This is the SDR alone ..

Alex

[b alex pettit jr]

SDR comparisons 

This is the SDR alone ..

AirSpy mini SDR

nooelec SDR

[b alex pettit jr]

Installing the DiscoDish Feed on a 1m WiFi Grid Dish

Disco Dish parameters 

Model of a 70cm dia f/D 0.35 Dish puts the Focal Point @ 24.5cm

The Focal Point of the Disco Feed is     24.5 cm from the base of the Dish ( as installed on the Disco Dish )

The Focal Point for the 1m WiFi Dish is 35.0 cm from the base of the Dish 

Therefore : The  Disco Feed support needs to be Extended by 10.5 cm for proper positioning on the Grid Dish

Alex

[cgd]

Thanks Alex. I'll check my DD feed and make sure the reflector is 35cm from the base of the Nooelec dish. 

[b alex pettit jr]

The problem Is: to what point on the feed is that measured ? I do not know .. 

But, as if is designed for a 70cm f/D 0.35, its position needs to be extended by 100/70 from its stock length when 

used on a 100cm f/D 0.35 dish.  aka  + 10.5cm

============================================================================

[b alex pettit jr]

Just found a system similar to the Discovery Dish on Ebay  ..  

The Deep-Sky Dish Kit by R3 Radio Research provides the DIY kit components to enhance your Hydrogen Line or 

other Discovery Dish observations by up to 6.6 dB with an improved 1.5m diameter parabolic reflector

Guidance on Feed Distance from Reflector:

Discovery Dish fails to disclose the most crucial operational parameters for Discovery Feeds. Deep-Sky Dish provides a reference starting point. Deep-Sky Dish should have Discovery Feeds installed with the 140mm measuring mark aligned to the feed-side edge of the boom diameter adapter when using the included 25 cm 1" PVC extension tube aligned with the base chamfer stop points as shown in the instruction manual photos.

Real-World Testing of Improved Performance:

The Designers can attest that performance improvement of ~6 dB was observed with Deep-Sky Dish when receiving GOES-16 at ~45° elevation angle across 5 unique dish units when compared to multiple Discovery Dishes. Average SNR in HRIT demodulation using SatDump improved from ~5 dB to ~11 dB for consistent decoding of GOES broadcasts with high margin, a strong testament to the similar f/D of Discovery Dish and the converted solar cookers lending itself to equalized feed performance between the 2 apertures as 6 dB improvement closely aligns with 20*LOG10(1.5/0.7) = 6.6 dB theoretically possible improvement from the increase in dish aperture size.

On Monday, December 1, 2025 at 11:47:02 AM EST, 'cgd' via Society of Amateur Radio Astronomers <sara...@googlegroups.com> wrote:





Thanks Alex. I'll check my DD feed and make sure the reflector is 35cm from the base of the Nooelec dish. 

On Monday, December 1, 2025 at 7:28:04 AM UTC-6 b alex pettit jr wrote:
> Installing the DiscoDish Feed on a 1m WiFi Grid Dish
>
> Disco Dish parameters 
>
>
>

> Model of a 70cm dia f/D 0.35 Dish puts the Focal Point @ 24.5cm
>
>
>

> The Focal Point of the Disco Feed is     24.5 cm from the base of the Dish ( as installed on the Disco Dish )
> The Focal Point for the 1m WiFi Dish is 35.0 cm from the base of the Dish 
>
> Therefore : The  Disco Feed support needs to be Extended by 10.5 cm for proper positioning on the Grid Dish
>
>

> Alex