AZ/EL Antenna Build

Discussion in 'Satellite and Space Communications' started by AB6RF, Nov 21, 2020.

ad: L-HROutlet
ad: l-rl
ad: Left-2
ad: Subscribe
ad: L-MFJ
ad: abrind-2
ad: Left-3
  1. AB6RF

    AB6RF Ham Member QRZ Page

    I've been wanting to add LEO satellite capability to my home station, but instead of taking out the credit card and going to town ordering new fancy stuff from HRO, I’m going to try to home-brew the most of it.

    I've been collecting the parts for this project, and recently I scored three old but unused Alliance U-100 rotators, and then an ugly but functional Yaesu G-800 rotator.
    The U-100 makes a good elevation rotator, and after I rebuilt the G-800, it is ready for taking the azimuth job.

    Today I made an adaptor plate to mount the U-100 on top of the G-800.



    Earlier I already made a 5VDC / 28VAC power supply for the rotator system (still need to add 24VDC rail for the G-800).

    Next step is to make the elevation position sensor.
    I want to build the position sensor inside the U-100 rotator to keep it out of the elements and keep the setup "clean" and robust.
    I've been already thinking how to do it, and the parts are starting to come in.

    Mikko, AB6RF
    Last edited: Nov 21, 2020
  2. VE3CGA

    VE3CGA XML Subscriber QRZ Page

    looks good, I did almost the same thing a few years ago. The sysytem is still working after hot & humid summers, rain, cold winters.
    I have some info on my qrz page, but used the k3ng software (modified of course) and built a plexiglass cover for the rotors and drive/sensor electronics
    Only 24vac and serial linesa go to the remote electronics
    I'm using electronic sensors for both az & el
    works all over the place (vucc) including 14 countries

    good luck
  3. AB6RF

    AB6RF Ham Member QRZ Page

    Thanks for sharing the info about your controller.

    I see you went with a “split architecture”, with a second Arduino and the motor controllers up in the tower.
    That’s the one detail I’m most wondering about. So far I’m thinking to keep all the electronics in the shack controller box, and just use potentiometers for the position sensing.
    But the I2C magnetometers and accelerometers are sooo much more elegant and “modern”, just can’t run the I2C for tens of feet down to the shack.
    I might yet change my mind and split the controller into two boxes, like you did.
  4. VE3CGA

    VE3CGA XML Subscriber QRZ Page

    The reason for the master/slave setup, was mainly due to the use of the I2C sensors. The main sends RS232 (ttl level) out to the slave where the commands for azel are converted to the motor movements and compared to the mag sensors. The remote corrects the difference and sends the sensor inputs back to the master.
    I'd like to change the RS232 to a 2.4Ghz link like I use for my wifi ant switch.
    Also I didnt like the idea of calibrating analog 0-10v over 20' especially with 110W of RF that close.
    Originally I set up the rotor with a photo interupter counting gear revolutions, but that would be similar to the c band dish actuators and would require calibration or some memory to store a position count when shutting things off.

    The electronics are not really in moisture proof sealed containers, but are shielded from rain & snow. The circuit board was sprayed with clear laquer and is inside a hammond plastic box. Theres no sign of any moisture. Different story for the sensors, they are in small plastic boxes and are sealed with liquid electrical tape on the lid.
    Last summer I hit the wire for the elevation sensor with the tree trimmer pole, ripped the cable out of the box, so I was very suprised there was no signs there had ever been any moisture in anything out there. The bottom of the plexi- rotor cover is open and I used some rubbery plastic coffee can lids as weather resistant washers over the boom as it enters and exits both sides of the plexiglas cover, they work perfectly just friction fit with the hole saw cut for the boom pipe.

    theres a groupio forum for Radio Artisan and that where you'l find the K3NG code. Pretty complex but fairly well documented and configurable to just about every type of interface possible. I would strongly advise the master/slave setup and RS232 the 2 boards. I think I have 30' of cable for the serial and 24VAC for the rotor. The IC9100 and the rotor fed from satpc32. I was going to change the serial to RS485 levels if distance was going to be a problem but I think a better way to go would be to use the 2.4Ghz xceivers, they are 95' from each other on the wifi antenna switch and have already survived the 2nd season of snowy wx

    I think I have about $150 invested in the project and 2/3 of that was the rotor
  5. AB6RF

    AB6RF Ham Member QRZ Page

    Great info, thanks.
    My controller will also be based on the K3NG code, already made one controller using that code for a simple AZ rotator.
    It uses potentiometer for the sensor, and it's working great, so that's what pushed me towards using the same approach also this time.
    I thought it was really fun to dig into that code and modify it to do exactly what I wanted.
    The K3NG code is a real gem, it's an amazing resource, and totally free!

    Have to say, I really like those I2C sensors, and the split architecture would make lot of sense. Also it would require less cables running up the mast.

    I really need to think about that split approach.
    I have a relay board and countless Arduino boards already sitting around that could be made into the remote unit.

    A question about your antennas.
    Looks like you're using linearly polarized single-band yagis, but slanted about 45° from the horizontal.
    If you don't mind, why did you go with the 45° slant, as opposed to say vertically mounted yagis?
  6. VE3CGA

    VE3CGA XML Subscriber QRZ Page

    well, no reason in particular but
    I was a little worried about getting RF into the electronics if antennas were broadside or perhaps desensing in the rig
    So far theres been no interference or lock ups. I only run enough power to hear myself on the downlink and the other day I was hitting AO-7 with 5W almost to LOS as it was heading north-north west
    I was going to find out if there was a particular circular polarization to use but never did. It wouldnt be too hard to add elements on each yagi for that purpose. Some fading at times but hasnt been severe enough to change anything
    I still do some 2M weak signal on band opennings so the thought was to keep some horiz

    The sensors came from amazon and were very cheap.
    I didnt go the relay route, instead went with zero crossing optos to drive triacs, so the entire slave unit is solid state.
    If you are going to do the mag sensors, I used the cables from 2 old microsoft ball type mouse units. The wire is small, very flexable even in freezing temperatures and its shielded with enough conductors for power, gnd and I2C. I ran it thru some tygon flexable fuel line and fed that thru some plastic bulkhead connectors, remember to keep metal away from the sensors, thats why the AZ sensor is about a foot higher than the motors. The EL sensor doesnt seem to mind the aluminium boom under it. I think recent developments with the code may allow for a 6 way sensor to do both, but you'd have to check
    That entire antenna has to be moved on to a tilt over telescoping mast- maybe next spring so I can do some 2M work again
  7. NG1H

    NG1H XML Subscriber QRZ Page

    A few minor mechanical tips:

    1) Make sure your payload is almost balanced. A little imbalance keeps the pressure on one side of the gear and makes the system less likely to bounce around due to backlash.
    2) If possible incorporate a ramp up/down feature in the motor controls. Less wear on the gears and motors and lowers the bouncing around time after you come to a stop after a slew.

    Keep updating your progress. This is an interesting project.
    N6MST and VE3CGA like this.
  8. AB6RF

    AB6RF Ham Member QRZ Page

    Thanks to both VE3CGA and NG1H, those are good suggestions.

    That suggestion about making the cross boom "almost balanced" is good one, didn't think of that.
    Currently my plans for the control box use relays, not sure they are fast enough to implement the slow-down ramp for the DC azimuth motor. I'll have to see how much the whole thing wobbles with the ramp-down.

    I should get some more parts from McMaster today so I can continue adding the position sensing into the EL rotator.

    I don't have a solid plan for the antennas...
    For 440 band I don't currently have anything, so I wonder if I should just get the M2 436CP16 that they also use in their LEO-pack.
    For 144 I could just use an existing 4-el liner pol yagi I already have. Not ideal but it would get me started.
  9. KS1G

    KS1G Ham Member QRZ Page

    I started with the WA5VJB Cheap Yagis, also had pieces of a HyGain Oscar array from back when we had AO13 I was able to reuse for a while. Linear antennas - you'll give up some pass time to QSB and polarization fades but they were easy to build and got me (re)started. I could have tried building a 2nd set of elements and phasing harness to circularize them, but opted for the LEO pack and they work very well. I know of a folks who didn't bother with circular at all - they have V and H available and 2 remote control SPDT relays (one for each band) to switch as needed.
  10. AB6RF

    AB6RF Ham Member QRZ Page

    Made some progress today adding the position sensor to the U-100 EL rotator.
    Mind you, I don’t have a machine shop or any fancy tools, this is simply a bench vise, hacksaw and a file job.


    My initial plan was to use a small timing belt to connect a multi-turn potentiometer to the rotator main shaft. I have these NOS Bourns 1K 10-turn potentiometers.
    To adapt the pot actuator shaft to the little pulley, of course required an adapter shim.



    But the belt puts a side load on the potentiometer shaft and that might not be so good in the long term, so I got couple plastic gears from McMaster.

    Cutting the center out of the larger gear gives me a “tooth ring” that can be slipped on the rotator shaft, to directly drive the potentiometer.


    But that skinny tooth ring doesn’t leave enough room for the potentiometer body, it interferes with the rotator shaft.
    So I cut another ring out of 1.5” PVC pipe to stretch out the main gear diameter.


    Looks ok, time to put down the tools and go back in the house.


    Mikko, AB6RF
    Last edited: Nov 24, 2020 at 6:16 AM
    KS1G and VE3CGA like this.

Share This Page