15-Day Drift Scan at Dec +40 Degrees from my 2.4 m Dish

[Ayushman Tripathi]

Hi,

I managed to resolve the artifacts generated by the baseline correction in my script.

Here are the plots with this week’s data included.

I’ll also adjust some observation settings to evaluate the impact on the results.

Also, for my smaller dish, I’ve added a steel mesh around it and will share its data.

Thanks.

[Alex P]

Excellent results :)

What techniques did you implement ?

Thanks,

Alex

[Ayushman Tripathi]

Hi Alex,

Here's the pipeline method I used:

1. Per-observation calibration. Each 5-minute scan is recorded raw with Airspy Mini (3 MHz span, 1024 FFT bins at 1420.406 MHz). VIRGO processes it against a separate cold-sky calibration scan: it divides the observation by the cold-sky spectrum, then normalizes it by the noise standard deviation in the line-free channels to put everything in signal-to-noise (SNR) units and finally applies a median filter in frequency. The output is one "filtered" spectrum per scan.

2. Wing-linear residual. I measure the remaining slope and offset using only the true line-free wings and then subtract that linear baseline from the whole spectrum.

3. VLSR correction + RA stacking. Each observation is Doppler-shifted onto a common LSR velocity grid, then observations are binned by RA (0.10 h bins) and median-combined across all sidereal loops.

I'm constantly experimenting with these techniques to improve the pipeline.

Thanks.

[b alex pettit jr]

Hello Ayushman,

Sounds very close to my sequence, except I subtract vs divide as the shifts are not gain but environment amplitude shift related.

And I use simple mX+b slope correction ..

I probably will not incorporate VLSR correction in the HL3D code.

Thanks,

Alex

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

[Ayushman Tripathi]

Hi Alex,

Yes, I’m still trying to improve it. I can also try HL3D’s plotting on the same collected data later this week and will share the results.

Thanks.

Sent from my iPad

[b alex pettit jr]

Attached is an example of the IFavg format for *.txt Input  and the *.csv output

*.CSV format as required for Rinearn Graphics 2D / 3D .

A1 Blank  

B1 ....  (RA) Time Hrs

ColA = Freq

Ampl Field dB re cold sky

Thanks,

Alex

[Ayushman Tripathi]

Hi Alex,

I wrote a rough converter from my pipeline output to the Rinearn CSV format you shared.

Here are some plots I generated using Rinearn Graphics.

The converter might still have a few bugs, I'll try to improve this along with my processing pipeline.

Thanks.

[b alex pettit jr]

On Friday, April 24, 2026 at 05:51:19 PM EDT, Ayushman Tripathi <ayushmantr...@gmail.com> wrote:

Hello Ayushman,

>>>  Well I am Impressed   !!   

Very small adjustments to the "Range" Z  min settings can make major changes in the appearance of the plots

example : auto scale set Z min @ -0.2543    changing to -0.15 or -0.35 might improve the plot ( OR More, of course )

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

Attached are a few plot formats to try ..

Copy them to  the Rinearn3D directory tree folder :  Rinearn3DQuickSetting

(  use Wordpad, etc to view & edit the  *.ini   text files )

This is a scan @ Dec +60 deg with the HC_Blue   w/ my 1.2m system 

Regards,

Alex P

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

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

[b alex pettit jr]

On Friday, April 24, 2026 at 07:31:30 PM EDT, 'b alex pettit jr' via Society of Amateur Radio Astronomers <sara...@googlegroups.com> wrote:

@ Dec +40

HC_blue

[Ayushman Tripathi]

Hi Alex,

Here are some plots using the formats you shared: HC_blue, True-Zaxis, etc.

Your scan at Dec +60 DEG looks very nice. My dish is currently pointed at Dec +40, next time I visit, I'll try Dec +60. I had to put the dish away from the city, it's near a forest with very little RFI, and it sends data back over cellular internet. I put the Raspberry Pi, cellular modem, UPS, etc. in a metal box, the only thing outside the box is the antenna for the cellular router. No Wi-Fi is used, everything runs over an Ethernet cable.

Thanks.

[b alex pettit jr]

Hello Ayushman,

Thanks. Not sure Rinearn is any improvement over your software, but its nice we can Both Plot using the same format !

Regarding the Gradient_Color_List, the colors are in Quads RGBI :  Red_Green_Blue_Intensity ;  ......

You can change the list to just a few or extend the list of color quads and the software will blend the transitions.

standard

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

A suggestion : Verify the "dB" scaling of your data .

Use a 50 ohm terminated LNA as the noise source and impose several steps of attenuation into the SDR,

process by your Software and verify there is consistency .. 

Adding a fixed Attenuator on the input to the SDR seems to improve the test results - 

probably because it creates a constant 50 ohm resistive Z on the SDR input vs what would normally be done :

 comparing attenuated signals against a directly connected LNA<>SDR for '0'dB reference.

Your values seem a bit high ..  

I've found 3dB over cold sky  is a typical max value regardless of dish size ... even to 12 meters in diameter !

Alex

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

[Ayushman Tripathi]

Hi Alex,

Yes this is great, now we can use same plot format and compare them easily.

I’ll experiment with color gradients.

Regarding dB scaling, I’ll check and confirm/correct that soon.

Thanks

Sent from my iPad

On 25 Apr 2026, at 1:03 PM, 'b alex pettit jr' via Society of Amateur Radio Astronomers <sara...@googlegroups.com> wrote:



Hello Ayushman,

Thanks. Not sure Rinearn is any improvement over your software, but its nice we can Both Plot using the same format !

Regarding the Gradient_Color_List, the colors are in Quads RGBI :  Red_Green_Blue_Intensity ;  ......

You can change the list to just a few or extend the list of color quads and the software will blend the transitions.

standard

<1777135940493blob.jpg>

<1777136201053blob.jpg>

<1777136250717blob.jpg>

<1777136319132blob.jpg>

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

A suggestion : Verify the "dB" scaling of your data .

Use a 50 ohm terminated LNA as the noise source and impose several steps of attenuation into the SDR,

process by your Software and verify there is consistency .. 

Adding a fixed Attenuator on the input to the SDR seems to improve the test results - 

probably because it creates a constant 50 ohm resistive Z on the SDR input vs what would normally be done :

 comparing attenuated signals against a directly connected LNA<>SDR for '0'dB reference.

<1777135438053blob.jpg>

Your values seem a bit high ..  

I've found 3dB over cold sky  is a typical max value regardless of dish size ... even to 12 meters in diameter !

Alex

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

On Saturday, April 25, 2026 at 10:10:51 AM EDT, Ayushman Tripathi <ayushmantr...@gmail.com> wrote:

Hi Alex,

Here are some plots using the formats you shared: HC_blue, True-Zaxis, etc.

Your scan at Dec +60 DEG looks very nice. My dish is currently pointed at Dec +40, next time I visit, I'll try Dec +60. I had to put the dish away from the city, it's near a forest with very little RFI, and it sends data back over cellular internet. I put the Raspberry Pi, cellular modem, UPS, etc. in a metal box, the only thing outside the box is the antenna for the cellular router. No Wi-Fi is used, everything runs over an Ethernet cable.

Thanks.

--
--
You received this message because you are subscribed to the Google
Groups "Society of Amateur Radio Astronomers" group.
To post to this group, send email to sara...@googlegroups.com
To unsubscribe from this group, send email to
sara-list-...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/sara-list?hl=en
---
You received this message because you are subscribed to a topic in the Google Groups "Society of Amateur Radio Astronomers" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/sara-list/ALXILeChUUI/unsubscribe.
To unsubscribe from this group and all its topics, send an email to sara-list+...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/sara-list/1926761313.2024462.1777136613463%40mail.yahoo.com.

<1777135438053blob.jpg>

<1777135940493blob.jpg>

<1777136201053blob.jpg>

<1777136250717blob.jpg>

<1777136319132blob.jpg>

[b alex pettit jr]

Hi Ayushman,

Process data from several days individually and plot this region. 

Does each show this same Light/Dark pattern ?

Alex

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

[b alex pettit jr]

VERTICAL_SPLIT   (Hidden Option)

If this text is added to the upper left ( currently empty ) cell of the CVS file, 

 the LINES function will draw each spectrum  as an individual line on the plot.

Regards,

Alex

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

[Ayushman Tripathi]

Hi Alex,

Thanks.

Yes, I found an issue with one of the 15 days of data I was using for the stacked plot, it had incomplete data. After removing that day and regenerating the stack, the pattern is no longer that noticeable.

Attached is the new 14-day stacked plot, along with a few individual day plots that go into it. I'll experiment more with this later today.

I'll also try VERTICAL_SPLIT, right now the CSV I'm loading into Rinearn is already multi-day stacked by my own pipeline before I send it in.

Thanks.

[Ayushman Tripathi]

Hi Alex, I tried VERTICAL_SPLIT, and it's really cool. I'll experiment more with it. Thanks

[Ayushman Tripathi]

Attached is the last 7-day stacked data with VERTICAL_SPLIT, and a VLSR-corrected 3D plot.

[Ayushman Tripathi]

Hi,

Here's a plot I made with Starplot using the same data.

Also for my smaller 1m WiFi grid dish, I'm getting a telescope alt-az mount with tripod and adapting it for the dish, and I'm building a new cantenna for the 2.4m dish with a choke ring as per Alex's advice. I'll share photos soon.

I also ordered a NanoVNA V2 clone from eBay, the smaller one, for $64. Not sure about the quality, but the price looked good, so I'll give it a try.

Thanks.


NanoVNA:

[b alex pettit jr]

Hello Ayushman,

The narrow beam width of the 2.4m dish is quite apparent in the plots.

Clean RFI free data !

The " StarPlot" is interesting.

Regards,

Alex

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

On Tuesday, May 12, 2026 at 07:04:58 PM EDT, Ayushman Tripathi <ayushmantr...@gmail.com> wrote:

Hi,

Here's a plot I made with Starplot using the same data.

Also for my smaller 1m WiFi grid dish, I'm getting a telescope alt-az mount with tripod and adapting it for the dish, and I'm building a new cantenna for the 2.4m dish with a choke ring as per Alex's advice. I'll share photos soon.

I also ordered a NanoVNA V2 clone from eBay, the smaller one, for $64. Not sure about the quality, but the price looked good, so I'll give it a try.

Thanks.


NanoVNA:

On Monday, April 27, 2026 at 6:41:38 PM UTC-4 Ayushman Tripathi wrote:

Attached is the last 7-day stacked data with VERTICAL_SPLIT, and a VLSR-corrected 3D plot.