My take on the SETI-Feed

[Jimmy Fransson]

Hello,

Let me tell you about me and my recent project with building a SETI-waveguide

I always meant to go down this path and learn a bit more about antennas and general RF as well as being able to scan the sky for myself but life didn’t let me, until now.

 

Then Atlas arrived and I thought to myself that this is a perfect excuse to finally learn the heavy stuff and so I started constructing a semi-printed version of a waveguide. As I had just started to use Chat-GPT for philosophical discussions I thought it would home me straight…

Chat is a really helpful tool and as dumb Rainman with confirmation bias as a standard… It also hallucinates and you really must have a strict BS-filter in order to use it successfully. After navigating around the BS I decided to build a “mix-feed” in order to be able to record 

1420MHz as well as 1665MHz. This was to be a disaster as a can with 78mm in diameter was by Chat marked as “Sweet spot”... I lost two weeks here.

The first iteration with a printed can of 135mm in diameter and 300mm in height was bad and I got a S11 value of maybe -5db at best… I now decided to demote Chat, he was not to be trusted in decision making.

I then did what I should have done in the first place, I followed the design of Apostolos Spanakis-Misirlis and built a “standard” SETI-waveguide and lo and behold, this sucker worked!

I have chosen an approach with PetG plastic for both inside and outside, I know there must be some decibels lost on S11 with me having a plastic liner around the walls but this makes it really sturdy and keeps the shape neat.

And here is the result:

It seems to work but other than being really good at getting my test-signal I really haven’t seen or even processed anything for real.

If anyone is interested in this building technique and would like to know more, just send me a message or a reply here. That goes for telling me I am wrong too. 

Kind regards // Jim

[b alex pettit jr]

If you widen the analysis band width, you can see more of the response curve. ?

[Captain Anne Flint]

Hi //Jim: 

What is PETG made of - it is copper-colored? or is that your lining? Sure is a handsome setup. Have you tried Wikipedia at all for a reference? IMHO, Chat GPT is developing but not that helpful yet (sounds as if you proved that empirically), but to troubleshoot both at the same time … I would try to get ahold of Pablo. Actually I would just do RA by itself, as I’ve had some annoying experiences with AI. Last time I WON, but I’m sure it’ll catch up in time. 

Wende 

On Feb 11, 2026, at 10:23 AM, 'b alex pettit jr' via Society of Amateur Radio Astronomers <sara...@googlegroups.com> wrote:

If you widen the analysis band width, you can see more of the response curve. ?

--
--
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 visit https://groups.google.com/d/msgid/sara-list/1701424308.947440.1770827035509%40mail.yahoo.com.

[Jimmy Fransson]

Yes, it's like taking advice from a really smart person, who has smoked an ounce of bad weed, easy to get fooled it's the expert when it suddenly has facts from Narnia.

I think I have found how, where to use it and just accepted that it can't really think as of yet, but it can for sure help to "Brainstorm".

The PetG is transparent in order to make it a bit more Tesla-like and also... That's what I had on the shelf and it works really well.

The copper-sheet is just 0.2mm and the combination of inner and outer plastic walls makes it pretty solid structure but at the same time makes it really light.

This plastic is also great in water and sunlight (And snow).

Everything is connected/attached with copper-tape and without the cocoon it would be a good laugh being as thin and flimsy as it is, but in this way it's almost perfect and by using this method I can now print pretty much any dimension and structure I want to try. 

I still could tweak it a lot but as it's working I decided to put in into use first.

Not claiming it to be the best, not at all but I think it's a pretty neat future for DIY-folks that means with this method you can skip the heavy metalwork, just printing, taping and cutting foil with a pair of scissors. 

To the right is Chat's "Sweet spot". :)

When the Swedish winter is taking a break I will most likely buy a proper dish and do some general tweaking.

I am pretty pleased with it but then again, no alien signal... ;)


//Jim

[fasleitung3]

Clever idea to put the copper in between two layers of printed PETG. I was wondering how you connected the copper tube to the ground of the connector. Also how you electrically connected the back reflector to the tube.

Both cannot be seen on the picture.

Best regards,

Wolfgang

Am Mittwoch, den 11.02.2026, 03:55 -0800 schrieb Jimmy Fransson:

Hello,

Let me tell you about me and my recent project with building a SETI-waveguide

I always meant to go down this path and learn a bit more about antennas and general RF as well as being able to scan the sky for myself but life didn’t let me, until now.

 

Then Atlas arrived and I thought to myself that this is a perfect excuse to finally learn the heavy stuff and so I started constructing a semi-printed version of a waveguide. As I had just started to use Chat-GPT for philosophical discussions I thought it would home me straight…

Chat is a really helpful tool and as dumb Rainman with confirmation bias as a standard… It also hallucinates and you really must have a strict BS-filter in order to use it successfully. After navigating around the BS I decided to build a “mix-feed” in order to be able to record 

1420MHz as well as 1665MHz. This was to be a disaster as a can with 78mm in diameter was by Chat marked as “Sweet spot”... I lost two weeks here.

The first iteration with a printed can of 135mm in diameter and 300mm in height was bad and I got a S11 value of maybe -5db at best… I now decided to demote Chat, he was not to be trusted in decision making.

I then did what I should have done in the first place, I followed the design of Apostolos Spanakis-Misirlis and built a “standard” SETI-waveguide and lo and behold, this sucker worked!

I have chosen an approach with PetG plastic for both inside and outside, I know there must be some decibels lost on S11 with me having a plastic liner around the walls but this makes it really sturdy and keeps the shape neat.

And here is the result:

> linux-headers-6.12.69+deb13-amd64

It seems to work but other than being really good at getting my test-signal I really haven’t seen or even processed anything for real.

If anyone is interested in this building technique and would like to know more, just send me a message or a reply here. That goes for telling me I am wrong too. 

Kind regards // Jim

--
--
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 visit https://groups.google.com/d/msgid/sara-list/f2f50b45-b097-4a8a-a309-0b0a3965ac97n%40googlegroups.com.

[Don Latham]

From our friend Claude:

The skin depth (δ) of RF in copper at 1420 MHz can be calculated using the formula:

δ = √(2/(ωμσ))

Where:

• ω = 2πf (angular frequency)
• μ = μ₀μᵣ (permeability of copper ≈ μ₀ = 4π × 10⁻⁷ H/m)
• σ = conductivity of copper ≈ 5.96 × 10⁷ S/m

At f = 1420 MHz (the hydrogen line frequency):

δ = √(2/(2π × 1420 × 10⁶ × 4π × 10⁻⁷ × 5.96 × 10⁷))

δ ≈ 1.73 micrometers (μm) or about 0.0017 mm

This extremely shallow skin depth means that at 1420 MHz, the RF current flows in a very thin layer near the surface of the copper conductor. This has important practical implications.

And for aluminum:

δ ≈ 2.12 micrometers (μm) or about 0.0021 mm

So aluminum's skin depth at 1420 MHz is about 23% larger than copper's (2.12 μm vs 1.73 μm), which makes sense since aluminum has lower conductivity.

Practical implications:

• Aluminum requires slightly thicker plating or surface layers for the same effectiveness
• Still very shallow, so surface condition remains critical

So copper tape and aluminum tape should be quite sufficient for forming "stuff" at the H2 line. Note the caution re surface condition.

The formula and calculations are correct

------------
Don Latham
PO Box 404,
Frenchtown, MT, 59846
406-626-4304

 

---

From: fasleitung3 <sara...@googlegroups.com>
To: sara-list <sara...@googlegroups.com>
Date: Thursday, 12 February 2026 1:09 PM MST
Subject: Re: [SARA] My take on the SETI-Feed

To view this discussion visit https://groups.google.com/d/msgid/sara-list/da9de3c0b1a59cfe14273009562185049cfe8b01.camel%40googlemail.com.

[Marcus D. Leech]

I watched Claude get bamboozled into spewing woo-woo amethyst-crystal-vibrational-consciousness astrological horse rot last night.  I am increasingly skeptical of anything
  ANY AI has to say.  Always check the results.  Which usually involves consulting the 'conventional' materials you should have used in the first place....

To view this discussion visit https://groups.google.com/d/msgid/sara-list/1023141260.23210853.1770933712952.JavaMail.zimbra%40blackfoot.net.

[Dave Typinski]

On 2/12/26 17:04, Marcus D. Leech wrote:
> ...

> I watched Claude get bamboozled into spewing woo-woo
> amethyst-crystal-vibrational-consciousness astrological horse rot last night.

Sounds like a few people that one ran into on the sci.* newgroups. Maybe the
LLM training data includes old Usenet posts and the Time Cube guy's web site?

> I
> am increasingly skeptical of anything
> ANY AI has to say. Always check the results. Which usually involves
> consulting the 'conventional' materials you should have used in the first place....

Absolutely, until the LLMs start making money. The AI industry is currently a
financial circle jerk between the LLM companies, Nvidia, and venture
capitalists. Nobody in the AI market is making much money yet, which indicates
LLMs aren't useful yet. They may get there someday, but nobody knows when or
even if someday will occur. For now, especially for scientific inquiries, it's
far more efficient to pretend LLMs don't exist.
--
Dave

[Captain Anne Flint]

Alas, my concern is finding them in those big blue Chat buttons, especially on financial websites. When I found one that didn’t seem to understand my issue at all, I typed in, “Are you an AI?” and in a VERY short time it was replaced by a human who groveled at my [virtual] feet. Although my problem was solved, I still don’t feel too good about it. 

Wende 

--
--
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 visit https://groups.google.com/d/msgid/sara-list/698E5F76.9010107%40typnet.net.

[Jimmy Fransson]

I skipped all the fancy looking methods and just taped the bottom on the outside as it will all be stabilised by the plastic.

It's ugly but to my knowledge it works. :)

The connector (SMA) is pressed against the can with a brass spring washer on the inside, it came with the connector.

I made some "RF-Room" around it simply with another version of lining, simply put... A hole. 

I will make some new images when it comes inside again, to clarify.

[Jimmy Fransson]

Sure, but I didn't know the difference between AM and FM as late as Oktober last year and have to start somewhere. :)
I did some "co-work" with a guy from SETI (Jader Monari) regarding the Hessdalen-lights in 2014 (I am a professional photographer) and was impressed by his knowledge and all those cool gadgets... But I never got to it as you know,,, work, family, life...

But as a general advice regarding AI I say that you are absolutely correct but with Chat something can start from just brainstorming and to me that's pretty neat.

But also: It will kill us all starting with our creativity and we should for our own sake just stop it all and just live in the woods in different clans.