Like
most of the projects I build, this one started with a freebie. For the
last 50 years I've been mostly building around "junk" collected for
free or very cheap. This dish was sitting in my brother-in-law's back
yard unused for years so I decided to put it to use. The goal is to
make it usable for 23/13 cm EME and not spend any cash doing it.
A 6 foot tripod was sitting out in the antenna stock pile, left over from an earlier EME project. It was built from scrap metal and recycled at least twice by me so it is really a freebie! I elected to put it up in a "temporary" fashion so that it could be easily moved if need be. (or abandoned as the case may be) The tripod sit on pads placed on the ground with large cement blocks next to the pads. Tie downs hold the tripod in place. (The cement blocks are items that I make up whenever I pour concrete. Any left over concrete in the truck gets poured into forms I keep handy for this purpose. The blocks weigh 280 pounds each and I can easily move them around with one of my tractors. They come in handy for "temporary" antennas like this one.
The following photo shows the tripod:
This shows the concrete block better:
The top bearing is a simple sleeve so a bottom bearing must be in place to hold all the weight of the dish and it's mount. An old trailer axle bearing was used. A bearing mount and an insert to adapt to the main mast had to be fabricated from scrap steel:
When I first conceived of this project I wanted to use a gear system I had on hand but it proved to be unsuitable for the task. Scrounging around in the junk box produced an old motorcycle chain and sprocket set and a ancient CDR TRA rotator. The switches and stops were removed from the rotor so it would turn without limit. With a 3:1 (actually 43:15) ratio, the little rotator is more than enough turn this dish. The rotor mount and the top bearing are all salvaged scrap. Unfortunately at this point i was forced to spend 10USD for some muffler saddle clamps and 5USD for some bolts.
With the bottom end taken care of, it was time to look at the top end. I used the original TVRO dish mount and made some brackets to adapt it to AZ/EL. (BTW, the photo below shows my moonrise horizon, plus the tower holding up my C31XR tribander, the 40 meter 4-Square, and the Davis weather station in the background My WX at w7iuv.com/wx/wx.htm):
Two more views of the elevation system:
The photo below shows the azimuth position pot. The pot is a ten turn precision type. The pulley gives about a 4 to 1 ratio so the resolution is very good. With the setup I have, I see about +/- 0.5 degree accuracy. The readout is the same unit I described in the May, 1983 edition of the old Lunar Letter Magazine.
Here is the dish on the mount. The square black box contains a single turn precision pot and a pendulum for elevation position.
This image shows the dish with an old Chaparral C-band feed in place. There is a target dawn on the center of the dish and the sun throws a nice shadow from the feed horn allowing me to check positional accuracy. The tripod feed mount will be replaced with a new system when the 1296 feed is completed. In the background, you can see the 8x8 2 meter EME array. The tower is down in the maintenance position, normally it's up at 52 feet. The mast at the right is the 90 foot 160 meter vertical.
With the dish mounted and steerable, it's time to turn my attention to the feed horn. The feed is something that I have been agonizing over for a long time. From the research I've done it is clear that the "Super VE4MA" feed is the best performer. However, it is also difficult to construct with the limited resources I have available. The OK1DFC septum with a choke ring is a close second. Although it is not within my present capabilities to fabricate the OK1DFC as drawn, I might be able to modify the mechanical aspects of it so that I can build it and maintain the tolerances I want to achieve.
I've been bending sheet metal at home for various projects for 50 years now and I am well aware of the dimensional tolerances I can maintain. When I build 1296 antennas, I usually keep to +/- 0.5 mm. Rolling a tube for the VE4MA or bending up a "U" channel for the OK1DFC, I would be lucky if I could keep to +/- 0.5 INCH! However I can cut sheets pretty accurately and usually make chassis and such by using angle stock to make the corners.
So I started out with the hope of putting together an OK1DFC septum using the fabrication techniques I'm most comfortable with. My goal was to keep to +/- 0.5 mm and do it without spending any money!
I started with an old aluminum sign rescued from the scrap metal yard. (I buy these real cheap when ever they are available and keep "in stock")
The hardest part of salvaging these signs is cleaning the paint and vinyl off the front surface. Next I cut out all the parts:
I might mention that I think the drawings for both the VE4MA and the OK1DFC are not very good. If you build these things all the time and make it EXACTLY like the original, the drawings are fine. But I ran into trouble because some of the parts are not dimensioned or at least not dimensioned in a way I'm familiar with. I wasted some time and material because of this. I would recommend cutting out all the pieces from paper or cardboard first and fitting them together with tape or glue to see how it all fits together if this is your first attempt at feed construction.
The next part cost me money. I had to go buy aluminum angle stock and aluminum pop rivets. About 20 USD. Here is the pieces put together with pop rivets:
Note that the back plate is made from two pieces. Looking at the OK1DFC drawings, I see his plate is a single piece with no apparent mechanical or electrical connection to the septum. Other builders have hard connected the septum to the back by soldering, welding or other means. I chose to extend the septum and use two back plate pieces with angle stock so that the septum was connected securely. One of the problems I had with the drawing package was that I could not find a diameter for the feed probes. This caused me some grief and I had to experiment with diameter and length to get better than 30 dB return loss.
Next is shown the finished feed where you can see the pop rivets and stainless steel machine screws holding the halves together.
And the septum and probes:
In retrospect, I'm pretty sure I would not build it this way again. For one thing the angle stock and the pop rivets cost more than all the hardware for the dish mount and positioners. Another issue was the amount of time I spent building it. It was very labor intensive and took much more time than I anticipated. There are 204 pop rivets and 58 stainless steel screws holding it together. Match drilling all those holes took the most time. I was not able to meet my goal of +/- 0.5 mm but did meet +/- 1.0 mm. It is still a question in my mind whether or not the losses involved in the horn being an irregular polyhedron (as it would be if I were to bend it up) or the losses caused by the protruding rivets is greater. We will see.....
Moving on to the feed horn supports, it took a lot of time to come up with something that I could build and be happy with. Goals were cheap/free (as always) and minimal blockage. I don't yet know if I will put a choke ring on the septum or not, so am trying to minimize the blockage assuming I won't. I had to buy some steel 3/4 inch square tube (about 10 USD) but had everything else. I started three different brackets sets before I came up with something I could build and be happy with. I don't have any detailed photos of the mount yet but here is a "prototype" held together with "C" clamps:
And a close up:
After carefully reading and analyzing the data in W1GHZ's papers, it was obvious that blockage on a very small dish would be significant. In light of that data, I wasn't sure if I wanted to use a choke ring or not. I decided to build a simple choke and try it. I had some old aluminum siding from a discarded commercial tool shed. The material is very thin and has a pattern embossed into it. It's not good for much so I use it for "test" projects where I don't care about fit and finish much. The good part is that it is easily worked with a pair of tin snips and a hand nibbling tool. The outer ring was made from a single strip of material and lap joined with pop rivets. The back plate was laid out and cut by hand to the pencil marks on the material. You can see the two pieces ready to be assembled below:
In the photo below you can see the assembled choke. I used self tapping sheet metal screws to fasten the back plate tabs to the outer ring. I ran out of pop rivets, didn't want to buy more, and besides I was tired of drilling matched holes for them. In this case the screws work well, although they protrude into the inner choke area.
Another angle of the choke assembly:
And yet another:
Again, I wasn't sure if this would work and even then I wasn't sure if I knew where to put it in relation to the end of the septum feed. I bent those rearward tabs so that they would contact the outside of the feed horn, positioned it where the W1GHZ papers said it should be and held it in place with duct tape!
Another shot:
Since I finished this at mid day, the sun was in a good place so I made some sun noise measurements before and after installation. The cold sky / sun noise went up by 0.5 dB! Careful measurements showed something I thought was pretty significant. The absolute sun noise only went up a tiny bit but the cold sky noise went down more! This tells me that indeed blockage is limiting the forward gain improvement but the pattern is really better and hepled the G/T a noticeable amount.
I spent the best part of two days repeating those tests and getting good repeatability. One time the measurments was way low only to find out the duct tape had loosened and the ring slipped! More duct tape and the sun noise was right back where it should be.
Echo testing with 40 watts at the feed was dissapointing, with echoes only seen a couple times and weaker than the VK3UM calculator would suggest.
Last night I was able to con K2UYH into providing me a test signal and I heard him on his first TX for my first 1926 EME signal. I called him (JT65C) and we completed after a bit. PY2BS then called me and we completed also! Thanks to Al and Bruce for my very first 1296 EME QSO's!!
I guess the thing works!!
See the rest of the station at http://w7iuv.com
A 6 foot tripod was sitting out in the antenna stock pile, left over from an earlier EME project. It was built from scrap metal and recycled at least twice by me so it is really a freebie! I elected to put it up in a "temporary" fashion so that it could be easily moved if need be. (or abandoned as the case may be) The tripod sit on pads placed on the ground with large cement blocks next to the pads. Tie downs hold the tripod in place. (The cement blocks are items that I make up whenever I pour concrete. Any left over concrete in the truck gets poured into forms I keep handy for this purpose. The blocks weigh 280 pounds each and I can easily move them around with one of my tractors. They come in handy for "temporary" antennas like this one.
The following photo shows the tripod:
This shows the concrete block better:
The top bearing is a simple sleeve so a bottom bearing must be in place to hold all the weight of the dish and it's mount. An old trailer axle bearing was used. A bearing mount and an insert to adapt to the main mast had to be fabricated from scrap steel:
When I first conceived of this project I wanted to use a gear system I had on hand but it proved to be unsuitable for the task. Scrounging around in the junk box produced an old motorcycle chain and sprocket set and a ancient CDR TRA rotator. The switches and stops were removed from the rotor so it would turn without limit. With a 3:1 (actually 43:15) ratio, the little rotator is more than enough turn this dish. The rotor mount and the top bearing are all salvaged scrap. Unfortunately at this point i was forced to spend 10USD for some muffler saddle clamps and 5USD for some bolts.
With the bottom end taken care of, it was time to look at the top end. I used the original TVRO dish mount and made some brackets to adapt it to AZ/EL. (BTW, the photo below shows my moonrise horizon, plus the tower holding up my C31XR tribander, the 40 meter 4-Square, and the Davis weather station in the background My WX at w7iuv.com/wx/wx.htm):
Two more views of the elevation system:
The photo below shows the azimuth position pot. The pot is a ten turn precision type. The pulley gives about a 4 to 1 ratio so the resolution is very good. With the setup I have, I see about +/- 0.5 degree accuracy. The readout is the same unit I described in the May, 1983 edition of the old Lunar Letter Magazine.
Here is the dish on the mount. The square black box contains a single turn precision pot and a pendulum for elevation position.
This image shows the dish with an old Chaparral C-band feed in place. There is a target dawn on the center of the dish and the sun throws a nice shadow from the feed horn allowing me to check positional accuracy. The tripod feed mount will be replaced with a new system when the 1296 feed is completed. In the background, you can see the 8x8 2 meter EME array. The tower is down in the maintenance position, normally it's up at 52 feet. The mast at the right is the 90 foot 160 meter vertical.
With the dish mounted and steerable, it's time to turn my attention to the feed horn. The feed is something that I have been agonizing over for a long time. From the research I've done it is clear that the "Super VE4MA" feed is the best performer. However, it is also difficult to construct with the limited resources I have available. The OK1DFC septum with a choke ring is a close second. Although it is not within my present capabilities to fabricate the OK1DFC as drawn, I might be able to modify the mechanical aspects of it so that I can build it and maintain the tolerances I want to achieve.
I've been bending sheet metal at home for various projects for 50 years now and I am well aware of the dimensional tolerances I can maintain. When I build 1296 antennas, I usually keep to +/- 0.5 mm. Rolling a tube for the VE4MA or bending up a "U" channel for the OK1DFC, I would be lucky if I could keep to +/- 0.5 INCH! However I can cut sheets pretty accurately and usually make chassis and such by using angle stock to make the corners.
So I started out with the hope of putting together an OK1DFC septum using the fabrication techniques I'm most comfortable with. My goal was to keep to +/- 0.5 mm and do it without spending any money!
I started with an old aluminum sign rescued from the scrap metal yard. (I buy these real cheap when ever they are available and keep "in stock")
The hardest part of salvaging these signs is cleaning the paint and vinyl off the front surface. Next I cut out all the parts:
I might mention that I think the drawings for both the VE4MA and the OK1DFC are not very good. If you build these things all the time and make it EXACTLY like the original, the drawings are fine. But I ran into trouble because some of the parts are not dimensioned or at least not dimensioned in a way I'm familiar with. I wasted some time and material because of this. I would recommend cutting out all the pieces from paper or cardboard first and fitting them together with tape or glue to see how it all fits together if this is your first attempt at feed construction.
The next part cost me money. I had to go buy aluminum angle stock and aluminum pop rivets. About 20 USD. Here is the pieces put together with pop rivets:
Note that the back plate is made from two pieces. Looking at the OK1DFC drawings, I see his plate is a single piece with no apparent mechanical or electrical connection to the septum. Other builders have hard connected the septum to the back by soldering, welding or other means. I chose to extend the septum and use two back plate pieces with angle stock so that the septum was connected securely. One of the problems I had with the drawing package was that I could not find a diameter for the feed probes. This caused me some grief and I had to experiment with diameter and length to get better than 30 dB return loss.
Next is shown the finished feed where you can see the pop rivets and stainless steel machine screws holding the halves together.
And the septum and probes:
In retrospect, I'm pretty sure I would not build it this way again. For one thing the angle stock and the pop rivets cost more than all the hardware for the dish mount and positioners. Another issue was the amount of time I spent building it. It was very labor intensive and took much more time than I anticipated. There are 204 pop rivets and 58 stainless steel screws holding it together. Match drilling all those holes took the most time. I was not able to meet my goal of +/- 0.5 mm but did meet +/- 1.0 mm. It is still a question in my mind whether or not the losses involved in the horn being an irregular polyhedron (as it would be if I were to bend it up) or the losses caused by the protruding rivets is greater. We will see.....
Moving on to the feed horn supports, it took a lot of time to come up with something that I could build and be happy with. Goals were cheap/free (as always) and minimal blockage. I don't yet know if I will put a choke ring on the septum or not, so am trying to minimize the blockage assuming I won't. I had to buy some steel 3/4 inch square tube (about 10 USD) but had everything else. I started three different brackets sets before I came up with something I could build and be happy with. I don't have any detailed photos of the mount yet but here is a "prototype" held together with "C" clamps:
And a close up:
After carefully reading and analyzing the data in W1GHZ's papers, it was obvious that blockage on a very small dish would be significant. In light of that data, I wasn't sure if I wanted to use a choke ring or not. I decided to build a simple choke and try it. I had some old aluminum siding from a discarded commercial tool shed. The material is very thin and has a pattern embossed into it. It's not good for much so I use it for "test" projects where I don't care about fit and finish much. The good part is that it is easily worked with a pair of tin snips and a hand nibbling tool. The outer ring was made from a single strip of material and lap joined with pop rivets. The back plate was laid out and cut by hand to the pencil marks on the material. You can see the two pieces ready to be assembled below:
In the photo below you can see the assembled choke. I used self tapping sheet metal screws to fasten the back plate tabs to the outer ring. I ran out of pop rivets, didn't want to buy more, and besides I was tired of drilling matched holes for them. In this case the screws work well, although they protrude into the inner choke area.
Another angle of the choke assembly:
And yet another:
Again, I wasn't sure if this would work and even then I wasn't sure if I knew where to put it in relation to the end of the septum feed. I bent those rearward tabs so that they would contact the outside of the feed horn, positioned it where the W1GHZ papers said it should be and held it in place with duct tape!
Another shot:
Since I finished this at mid day, the sun was in a good place so I made some sun noise measurements before and after installation. The cold sky / sun noise went up by 0.5 dB! Careful measurements showed something I thought was pretty significant. The absolute sun noise only went up a tiny bit but the cold sky noise went down more! This tells me that indeed blockage is limiting the forward gain improvement but the pattern is really better and hepled the G/T a noticeable amount.
I spent the best part of two days repeating those tests and getting good repeatability. One time the measurments was way low only to find out the duct tape had loosened and the ring slipped! More duct tape and the sun noise was right back where it should be.
Echo testing with 40 watts at the feed was dissapointing, with echoes only seen a couple times and weaker than the VK3UM calculator would suggest.
Last night I was able to con K2UYH into providing me a test signal and I heard him on his first TX for my first 1926 EME signal. I called him (JT65C) and we completed after a bit. PY2BS then called me and we completed also! Thanks to Al and Bruce for my very first 1296 EME QSO's!!
I guess the thing works!!
See the rest of the station at http://w7iuv.com