Parabolic dish question from a newbie

[JERRY TAYLOR]

While waiting for a replacement Nooelec H1 LNA I decided to see if I could use my cantenna as a feedhorn for a Direct TV Slimline dish antenna.  The cantenna works well and I have been able to detect hydrogen easily.  It is an 8" to 6" duct reducer with a 6" end cap.

The dish is 32"x22" (solid, offset feed, dish doesn't appear bent or damaged) and still has the old LNB's so I have a good idea of where the focal point is.  However, I have been unable to see any additional gain from the dish (maybe 1 db or so).  I have tried using the Total Power app and background noise with a generic LNA to see if I can get any gain from the dish but that hasn't worked.  I also tried detecting hydrogen and adjusting the feedhorn location multiple times but was unsuccessful.

I realize that the larger feedhorn creates a shadow on the dish but I assumed I would still get some gain regardless.

Am I trying to do something not possible?

Does anyone have any suggestions?

Thanks!

Jerry Taylor

[Marcus D. Leech]

What is it you're expecting?

The absolute most you'll see in H1, regardless of aperture is about 2.2dB above background.  That's true whether you have
  a small dish or a large one.

A better test would be to do a Sun transit, record the total-power, and see if the resulting beam-width is consistent with
  the size of your dish.

--
--
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 the Google Groups "Society of Amateur Radio Astronomers" group.
To unsubscribe from this group and stop receiving emails from it, send an email to sara-list+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/5d5d6603-2846-4290-8f44-406bb34f115dn%40googlegroups.com.

[JERRY TAYLOR]

Thanks Marcus.

To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/162e07b9-373e-4437-a073-025f3703718d%40gmail.com.

[Eduard Mol]

Hi Jerry, great work detecting the hydrogen line with a simple cantenna!

It might seem counterintuitive at first but the hydrogen line signal does not really scale with aperture or antenna gain. The reason is that the hydrogen line signal is very evenly spread out over the sky- yes, it is stronger towards the galactic plane but there is atleast some detectable hydrogen in practically any direction. When you use a small cantenna instead of the larger dish, the lower gain of the cantenna is partially compensated by its much larger beamwidth (or antenna "field of view"; there is simply much more hydrogen gas in the wide beam of the cantenna. For sources that are much smaller than the beamwidth (e.g. sun, radio sources, masers etc...) this situation does not apply and the received power scales with antenna gain just as one would expect. That being said, gven that your dish is only about four wavelengths across it probably does not offer all that much extra gain anyway.

The main advantage of using a dish over a small cantenna is that the dish gives you more spatial resolution. With the cantenna you are seeing a huge swath of sky (perhaps 100 degrees across) so signals from different parts of the milky way get blended together. With the dish you should have a much smaller beamwidth (in the order of 15- 20 degrees) so you should be able to resolve individual arms of the Milky Way as distinct humps in your hydrogen line spectra. Try pointing your dish at different sections of the Milky Way to see if you can detect this.

Good luck with your observations, 

Eduard

[JERRY TAYLOR]

Thanks so much Eduard!  That explains some of the results I’ve been getting.

I was expecting a significant signal increase but now I know why that didn’t happen.

The dish I’m using was meant for a much higher frequency. So although it may have high gain at the design frequency, since I’m using it for a considerably lower frequency, the gain would also be considerably lower. Is that correct?

Jerry

To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/CAFcm1Q6QNV-tO5NDjpzzYb%3D-E7Jcx9vPHD0C6a8qq%3De7%3DnUVTg%40mail.gmail.com.

[Marcus D. Leech]

On 12/06/2024 10:26, JERRY TAYLOR wrote:

Thanks so much Eduard!  That explains some of the results I’ve been getting.

I was expecting a significant signal increase but now I know why that didn’t happen.

The dish I’m using was meant for a much higher frequency. So although it may have high gain at the design frequency, since I’m using it for a considerably lower frequency, the gain would also be considerably lower. Is that correct?

Jerry

Indeed, the gain of a dish antenna scales with both its size and its operating frequency:


G = k*((pi * D) / lambda)^2)

The efficiency factor, "k" is usually between 0.5 and 0.6 or thereabouts.

Notice that as the wavelength gets shorter, the gain increases for a fixed-size antenna  (assuming surface quality
  adequacy, etc).

To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/CAMMXHa4oNLxrRYaZ8XWDxZAg-PVDjv%3D5wFCW0QS0uzS_Qz9Hrg%40mail.gmail.com.

[JERRY TAYLOR]

Using my buddy Chat GPT this is the calculation for the theoretical gain of the dish I am using:

G=(λπD)2×η

Where:

• D=0.6858 meters
• λ≈0.211 meters
• η≈0.6

G=(0.211π×0.6858)2×0.6

G=(0.2112.154)2×0.6 G=(10.21)2×0.6 G=104.24×0.6 G=62.54

The gain G is approximately 62.54. To convert this to decibels (dBi):

GdBi=10log10(62.54)

GdBi=10log10(62.54)≈10×1.796 GdBi=17.96

Therefore, the gain of the Direct TV Slimline dish at 1.42 GHz is approximately 17.96 dBi.

The theoretical HPBW is:

HPBW≈D70λ

Where:

• λ is the wavelength of the signal.
• D is the diameter of the dish.

Given:

• λ≈0.211 meters (calculated earlier for 1.42 GHz).
• D≈0.6858 meters (as calculated from the average dimensions of the dish).

Let's compute the HPBW:

HPBW≈0.685870×0.211

Now calculate:

HPBW≈0.685814.77 HPBW≈21.54 degrees

Therefore, the Half-Power Beamwidth (HPBW) of the Direct TV Slimline dish at 1.42 GHz is approximately 21.54 degrees

If I can impose on you one more time, is Chat GPT correct with these calculations?

To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/6bc66d63-e40b-4c9e-9278-1c858f9bd545%40gmail.com.

[James Abshier]

No.

To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/CAMMXHa4BPtQbLk16Mtrh9p4S_3%2BRDsQec-ot1kKQ2As%3DGkrAmw%40mail.gmail.com.

[Eduard Mol]

Nope, the HPBW it spits out somehow seems to be in the right ballpark but the “calculations” it provides are rubbish.

I think the equation for HPBW it gives is scraped from the Wikipedia article on parabolic dishes, which gives the approximation HPBW ~ 70 X lambda / D

If you look closely you will recognize this equation in GPT-4s output but the model dropped the division… then it proceeds to just multiply lambda and D followed by some more gibberish.

I could probably go on and point out many more errors, but I think you get the point. GPT-4 is a language model, so no surprise it may perform poorly on physics problems, or any other task that does not relate to text generation for that matter.

Some say LLMs are “word calculators”. If that is correct then you’re probably better off just using a regular “numbers calculator” for these sort of simple physics questions 😉

Eduard

Op wo 12 jun 2024 om 23:25 schreef James Abshier <abs...@ameritech.net>

To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/f4c8b436-84f4-4050-ba69-eca3d28e74ae%40ameritech.net.

[JERRY TAYLOR]

Thanks Edward.  I suspected that GPT might not be that accurate. I’ve experienced other errors previously. I guess I better brush up on my mathematics.

Jerry

To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/CAFcm1Q58x58Q4pAWFsurnMhCia7htVxK93%2B1rfZgfNywvriA-Q%40mail.gmail.com.