Linear/C-pol switched feed BUD project

KF4YLM

Active SatelliteGuys Member
Original poster
Sep 20, 2017
21
27
VA
My Sunday project was to add another feed to my 10-foot Unimesh, for circular polarization reception. The recipe called for an old scalar ring and single-band polarotor, LNB, junk metal brackets, and a cheesy plastic cutting board. It was a success.

Quick look at the photos and diagram attached describe the thousand words below.


A quote from a thread on here a while back inspired me. Someone discussed multi-feed arrangements, and then posed the question, "wonder if I could do that on my dish that tracks the arc". Someone responded, "why would you do that?!" -- makes sense, you do multi-feed arrangements so you don't HAVE to turn the antenna.

BUT, let's say you want to be able to switch from linear polarities to circular polarities on the fly, without having to go out to the antenna and swap your feeds out or jam a dielectric plate in the feed?

The prime feed is a C/Ku polarotor, with Norsat LNBs. The antenna runs a VonWiess motor, and an ASC-1 controller drives the motor and the servo.

Adding the second feed was simple; I opted for a second full scalar ring to start. It was placed right up against the prime feed. I had pre-perforated steel angle brackets laying around that I cut to support the new feed. The prime feed ring was removed from the support arms of the reflector, those bolts then run through the brackets, and the feed support arms were placed on top of the brackets. (All this done wit 20MPH wind gusts, on a ladder! That, indeed, resulted in profanity!).

This arangement set the phase center of the feed off from the prime feed a good 8 inches or so. Since the target C-pol birds were further EAST, I placed the feed to the WEST of my prime feed.

Switching between my C/Ku LNBs has always been achieved with a 22kHz tone switch, 2-in/1-out. This had to scale up to a real DiSEqC switch. I had a 4-port Pansat switch, which didn't seem to want to work behind the ASC-1. I had another 8-port on hand that did work. So this was placed in the prime feed's radome.

Control of the offset feed's polarity runs in parallel with the prime. Two parallel motors seems to work flawlessly with the ASC-1. An old California Microwave LNB I rescued from somewhere ended up on the new offset feed.

Tested the new offset feed with linear polarity first. It was hearing birds OK, at about 70% magnitude (more than I expected) relative to the prime feed, and about 10 degrees further East-looking. This result was after some peaking was done on the angle at which the feed looked back into the reflector, and on the f/d plunge. The prime feed was also re-peaked, and each checked again to ensure accurate reference.

Next, a cheap cutting board from a discount chain box store was sacrificed -- a small square was cut out that would fit down the throat of the polarotor feed -- maybe 2-1/4" square or so. The measurement was eyeballed, then cut on the outside line, then edge-shaved with a utility knife for a snug fit. There were no guide rails in this feed for such a plate. Those rails exist in my prime feed, but the Ku-band corner reflector obstructs any chance of a dielectric plate being inserted there.

Side bar: So, the whole "dielectric plate" idea is to add a material with a slower wave propagation constant relative to air, which "de-polarizes" or "linearizes" (I like the latter term better!) a circularly polarized wave so it can be received with a linear polarity instrument. This method has a penalty of no less than -3dB loss over a stepped waveguide or phased probe array -- but has the advantage of being real cheap and easy. A 3dB hit was therefore deemed tolerable, even on top of the offset feed loss, on account of cost, simplicity, and feed structure mass, and resultant feed blockage. Plates are sold online for cheap -- Teflon-like materials and fiberglass seem to be the most popular. But I wondered if real cheap cutting board plastic (not even close to Teflon) would work.

The plate is inserted at a 45 degree angle relative to the linear feed positions (H/V). So observing where my ASC-1 placed the feed probe at max H-pol signal, the plate I cut was jammed -- then hammered gently -- in place at an eyeballed ~45 degrees. Close enough.


Earlier in the week, I'd steered over to SES14 and SES6 (near the extreme travel for my west-mounted screwjack polar mount) to see what happens with C-pol on linear feeds. Confirmed, as I suspected, that some transponders could be heard with a linear feed -- and indeed H/V didn't make a difference, nor did Left-hand/Right-hand! So long as that transponder's passband had NO signals in the opposite polarity, we were OK. But this is a trite handful of transponders. Data point collected!

Now, with sun set, chilly night air settling, I settled on the couch to find those birds again, this time on the offset feed with the linearizing plate. Showed up right where I expected them to from calculated position counter clicks! A few manual scans with my Vu+ Uno 4kSE revealed more transponders! And indeed, changing polarity on this linearized feed from H/V drastically changed SNR on those transponders who were previously uncopiable!! I hooked up my old AzBox to blind scan these birds and reveal more stuff!!

This is SOOOO COOL!

A note: Polarity commands on the Vu+ and the AzBox for H/V /can/ operate +13/+18V; and ASC-1 understand what to do with this. Vu+ has polarity descriptors for transponders that call out Left/Right hand CP. However, my box doesn't seem to change its output LNB voltage to instruct ASC-1 what to do. For now, I'm using the linear term on the box (H/V, for R/L, respectively).

VY 73 DE KF4YLM
 

Attachments

  • IMG_20190331_164620797.jpg
    IMG_20190331_164620797.jpg
    507.4 KB · Views: 334
  • IMG_20190331_164654554.jpg
    IMG_20190331_164654554.jpg
    137.2 KB · Views: 354
  • KF4YLM-Antenna-Config-2019-03.pdf
    179.9 KB · Views: 304
Nice work!

I just did the same project last week to get 177w, also a circular bird. I ordered a dual LNB holder for $20 instead of fabricating one and put my Corotor II+ in the centre and a C1PLL with dielectric plate on the right (eastern side). My project was to get Guam PBS, which it turns out doesn't come in. My 10 footer is half buried in a garden bush when pointed so far West and I am only getting 2 strong TPs - MTN colour bars and the AFN TP. Basically, I get nothing but I'm leaving it as is. I might change the circular LNB with an L-band unit and work on a new project, since the hard parts are done in terms of adding a second LNB inside the nose cone.

I should have done this project a month or 2 ago before the leaves filled in the bush. I guess I'll be watching Guam PBS this winter, at least.

I had set my ASC1 to constant 18v for the corotor LNBs, but now have changed it to 13/18 to accommodate the R/L circular LNB

118943-14a691fddae95e0caffac6763ee38f0e.jpg
118944-cdff3c555983d65513516f7bb7b9e969.jpg
118945-73d11cd479542ca0d25551fa1caec2da.jpg
 

Attachments

  • IMG_20190401_8257.jpg
    IMG_20190401_8257.jpg
    56.7 KB · Views: 316
  • IMG_20190401_27678.jpg
    IMG_20190401_27678.jpg
    161.5 KB · Views: 335
  • IMG_20190401_48342.jpg
    IMG_20190401_48342.jpg
    126.3 KB · Views: 313
  • Like
Reactions: martin-f and k.r.
My Sunday project was to add another feed to my 10-foot Unimesh, for circular polarization reception. The recipe called for an old scalar ring and single-band polarotor, LNB, junk metal brackets, and a cheesy plastic cutting board. It was a success.

Quick look at the photos and diagram attached describe the thousand words below.


A quote from a thread on here a while back inspired me. Someone discussed multi-feed arrangements, and then posed the question, "wonder if I could do that on my dish that tracks the arc". Someone responded, "why would you do that?!" -- makes sense, you do multi-feed arrangements so you don't HAVE to turn the antenna.

BUT, let's say you want to be able to switch from linear polarities to circular polarities on the fly, without having to go out to the antenna and swap your feeds out or jam a dielectric plate in the feed?

The prime feed is a C/Ku polarotor, with Norsat LNBs. The antenna runs a VonWiess motor, and an ASC-1 controller drives the motor and the servo.

Adding the second feed was simple; I opted for a second full scalar ring to start. It was placed right up against the prime feed. I had pre-perforated steel angle brackets laying around that I cut to support the new feed. The prime feed ring was removed from the support arms of the reflector, those bolts then run through the brackets, and the feed support arms were placed on top of the brackets. (All this done wit 20MPH wind gusts, on a ladder! That, indeed, resulted in profanity!).

This arangement set the phase center of the feed off from the prime feed a good 8 inches or so. Since the target C-pol birds were further EAST, I placed the feed to the WEST of my prime feed.

Switching between my C/Ku LNBs has always been achieved with a 22kHz tone switch, 2-in/1-out. This had to scale up to a real DiSEqC switch. I had a 4-port Pansat switch, which didn't seem to want to work behind the ASC-1. I had another 8-port on hand that did work. So this was placed in the prime feed's radome.

Control of the offset feed's polarity runs in parallel with the prime. Two parallel motors seems to work flawlessly with the ASC-1. An old California Microwave LNB I rescued from somewhere ended up on the new offset feed.

Tested the new offset feed with linear polarity first. It was hearing birds OK, at about 70% magnitude (more than I expected) relative to the prime feed, and about 10 degrees further East-looking. This result was after some peaking was done on the angle at which the feed looked back into the reflector, and on the f/d plunge. The prime feed was also re-peaked, and each checked again to ensure accurate reference.

Next, a cheap cutting board from a discount chain box store was sacrificed -- a small square was cut out that would fit down the throat of the polarotor feed -- maybe 2-1/4" square or so. The measurement was eyeballed, then cut on the outside line, then edge-shaved with a utility knife for a snug fit. There were no guide rails in this feed for such a plate. Those rails exist in my prime feed, but the Ku-band corner reflector obstructs any chance of a dielectric plate being inserted there.

Side bar: So, the whole "dielectric plate" idea is to add a material with a slower wave propagation constant relative to air, which "de-polarizes" or "linearizes" (I like the latter term better!) a circularly polarized wave so it can be received with a linear polarity instrument. This method has a penalty of no less than -3dB loss over a stepped waveguide or phased probe array -- but has the advantage of being real cheap and easy. A 3dB hit was therefore deemed tolerable, even on top of the offset feed loss, on account of cost, simplicity, and feed structure mass, and resultant feed blockage. Plates are sold online for cheap -- Teflon-like materials and fiberglass seem to be the most popular. But I wondered if real cheap cutting board plastic (not even close to Teflon) would work.

The plate is inserted at a 45 degree angle relative to the linear feed positions (H/V). So observing where my ASC-1 placed the feed probe at max H-pol signal, the plate I cut was jammed -- then hammered gently -- in place at an eyeballed ~45 degrees. Close enough.


Earlier in the week, I'd steered over to SES14 and SES6 (near the extreme travel for my west-mounted screwjack polar mount) to see what happens with C-pol on linear feeds. Confirmed, as I suspected, that some transponders could be heard with a linear feed -- and indeed H/V didn't make a difference, nor did Left-hand/Right-hand! So long as that transponder's passband had NO signals in the opposite polarity, we were OK. But this is a trite handful of transponders. Data point collected!

Now, with sun set, chilly night air settling, I settled on the couch to find those birds again, this time on the offset feed with the linearizing plate. Showed up right where I expected them to from calculated position counter clicks! A few manual scans with my Vu+ Uno 4kSE revealed more transponders! And indeed, changing polarity on this linearized feed from H/V drastically changed SNR on those transponders who were previously uncopiable!! I hooked up my old AzBox to blind scan these birds and reveal more stuff!!

This is SOOOO COOL!

A note: Polarity commands on the Vu+ and the AzBox for H/V /can/ operate +13/+18V; and ASC-1 understand what to do with this. Vu+ has polarity descriptors for transponders that call out Left/Right hand CP. However, my box doesn't seem to change its output LNB voltage to instruct ASC-1 what to do. For now, I'm using the linear term on the box (H/V, for R/L, respectively).

VY 73 DE KF4YLM

nice project and congrats on the achievement!

I've been doing pretty much the same, with an old polarotor side-mounted to the prime scalar. I did not put a scalar on the circular feed as I think the scalar works only if everything is symetrical (so, only in the prime spot) - someone correct me if I'm mistaken.

I get pretty good reception of 34.5W, 40.5W, 47.5W, one transponder on 27.5W, and can "smell" some signal on 22W and 53W (out of footprint for 53W, and trees in the way for 27.5W and further east).

As I recently won a Titanium LNBF on a contest on this forum, I will soon try this with a LNBF instead of the old polarotor, and compare. Since I manually drive the motors from an old analog receiver, it will be nice if I can eliminate the servo, and just use voltage switching.

How far East do you have a line of sight? There should be stuff on 11W, 8W and 5W, but I have too many trees and my neighbor's house in the way. I heard Caterpillar makes good preamps too ;-)
 
Antenna is in an open field, so it's clear from 139W back to 11W before a ridgeline gets in the way. Have to swap actuator positions from West to East to see anything below 27W. An H-H motor is a project on my list.

Thanks for the bookmarks... I'll try these out.

The second full scalar ring was added from convenience of mounting the feed. Scalar rings are beamforming mechanisms that are designed to increase feed efficiency by better reflector illumination. I haven't performed any modeling or other analysis to see if I'm better or worse with it...

Take a look at this paper from Paul, W1GHZ, about high efficiency prime focus feeds.


http://www.w1ghz.org/antbook/conf/high_efficiency_prime_feeds.pdf

His Antenna Book for microwave, a free PDF, is a critically important work for any and all (layman to PhD) who consider dabbling in cm- and mm-wave radiating structures, in my view.
 
Antenna is in an open field, so it's clear from 139W back to 11W before a ridgeline gets in the way. Have to swap actuator positions from West to East to see anything below 27W. An H-H motor is a project on my list.

Thanks for the bookmarks... I'll try these out.

The second full scalar ring was added from convenience of mounting the feed. Scalar rings are beamforming mechanisms that are designed to increase feed efficiency by better reflector illumination. I haven't performed any modeling or other analysis to see if I'm better or worse with it...

Take a look at this paper from Paul, W1GHZ, about high efficiency prime focus feeds.


http://www.w1ghz.org/antbook/conf/high_efficiency_prime_feeds.pdf

His Antenna Book for microwave, a free PDF, is a critically important work for any and all (layman to PhD) who consider dabbling in cm- and mm-wave radiating structures, in my view.

Wow -that's a pretty good run from 11deg to 139!
When I had a Ku LNBF in the centre of my 7.5' Unimesh dish with a Cband LNBF sidecar, I found that the modified Cband scalar didn't make any difference.
 
***

Users Who Are Viewing This Thread (Total: 1, Members: 0, Guests: 1)

Who Read This Thread (Total Members: 1)