[amsat-bb] Re: Non-mechanical Azimuth/Elevation Feedback Advice

[Greg D]

Yes, indeed the fine art of pulse counting works very well.  My rotor 
controller does the same with the "clicker" kind of rotors, where the 
contacts were intended to drive a solenoid ratchet switch mechanism for 
positioning.  Instead of the Arduino, I used a Basic Stamp (this is 
pre-Arduino heritage), but the pulses are 6-10 degrees apart.  500 per 
360 degrees would be too fast for this little squirrel.

Good luck with your projects.

Greg  KO6TH


Ing. Pavel Milanes Costa wrote:
> Yes, me to...
>
> I do have a old Japanese tracking camera mounting also with 24 AC 
> motors here...  360 degrees azimuth with stop switch on 0 and 360 
> degrees and about 0 to 120 degrees of elevation (0 ~ horizon, 90 ~cenit)
>
> Pretty good construction, a ham here in Havana have one the uses to 
> move 2 phased yaguis of 5 elements for 2m... the brake system is 
> primitive but works fine...
>
> Pointing is by eye and as in your case no means of feedback... 
> recently after completing some TODO-IN-THE-FUTURE projects with 
> Jal/JalLib (Just Another Language and libs for PIC devices) and pics I 
> get hands on this project
>
> A couple o months or so (maybe more) there was a thread on the list 
> about this...
>
> I find the arduino project from K3NG, but no arduino here, only 
> Microchip PIC (PIC18F4520) so I have to re-invent the wheel...
>
> I was thinking on gray encoding the position with 7 bits, but this is 
> A LOT of cables to run..
>
> Then I see on the specs of the arduino a feedback mechanism of pulse 
> counting...  and find on the internet a project from a polish? ham 
> that uses this mechanism wit a ATMEGA MCU but with a trick... (I have 
> the link on my work place... link will be tomorrow)
>
> I explain, he uses several magnets glued on the edge of an 
> intermediate gear in the reduction mechanism... then get a magnetic 
> switch (or a hal sensing device if you like) close to the edge where 
> the magnets are...
>
> This mechanism do about 500+ pulses for each 360 degrees... pretty 
> neat an simple... with only 2 cables...
>
> The only thing is you have to reset the mechanical position of the 
> antennas at each power-on to reset the internal count in the 
> pic/arduino/atmega MCU...
>
> I'm on the ideas-on-the-boad part of this project for making it 450 
> degrees instead of only 360 and how to deal in software with the 0 to 
> 90 degrees restriction on the elevation part...
>
> Just another idea to the brain storm...
>
> 73 de CO7WT.
>
> PS: control will be serial emulating a  RS232B protocol from Yaesu...
>
> El 07/08/13 09:58, Zach Leffke escribió:
>> Hello Everyone,
>>
>>                  I recently came into possession of a Pelco PT170-24P
>> tracking pan tilt pedestal designed to support large security 
>> cameras.  My
>> intent with this new acquisition is to repurpose it as a "low cost" 
>> (got it
>> on Ebay for ~$75 + S&H) alternative for an antenna tracking pedestal for
>> amateur satellites.  It uses 24VAC induction motors to move the 
>> azimuth and
>> elevation assemblies, pretty much just like the G5500s that I'm sure 
>> so many
>> of us are familiar with.  It definitely cannot support the same 
>> amount of
>> weight as the G5500, but I'm looking to construct a small, portable
>> satellite ground station node and this thing is plenty beefy enough to
>> handle a couple of Arrow style antennas.  Here is the problem, it 
>> provides
>> absolutely no feedback.
>>
>>
>>
>> My question to the group is does anyone know of a non-mechanical 
>> method for
>> getting relatively accurate feedback for azimuth and elevation.  I'm 
>> looking
>> for an all electronic means that I can mount somewhere outside of the 
>> actual
>> pedestal assembly (like perhaps on the cross-boom) that will be able to
>> provide measurement of the az/el (or pan/tilt, or yaw/pitch, whatever 
>> you
>> want to call it) position.  I'm using an arduino microcontroller for the
>> tracking controller.  Originally I intended to find a way to mount
>> potentiometers in inside the unit and simply use the ADCs on the 
>> arduino to
>> read the position feedback voltage from the pots, however, there is 
>> barely
>> enough space to mount an elevation feedback pot inside the unit, and 
>> there
>> is virtually no space for an azimuth feedback pot.  Hence I'm looking 
>> for a
>> non-mechanical method.
>>
>>
>>
>> My first thoughts for the elevation feedback was to use the old
>> potentiometer plus nice heavy weight method mounted out on the boom.  
>> This
>> idea doesn't appeal to me very much as other factors can now affect the
>> position feedback (such as high winds).  I then thought of something 
>> along
>> the lines of an accelerometer.  I also tossed around the idea of a 
>> 2-axis
>> gyro for both Az/El.  My issue is I have limited experience working with
>> these types of sensors, and was hoping to get advice from everyone in 
>> this
>> group.  I know for example that the gyro will provide rate of motion 
>> around
>> an axis and thus I have to integrate over time to get the actual 
>> position.
>> This becomes cumbersome because now I have to keep track of time in the
>> Arduino while executing movement commands (certainly do-able, just more
>> complicated than reading an ADC voltage).  Additionally, I believe these
>> devices suffer from drift and require frequent calibration (although 
>> there
>> may be a scheme of starting from a known position, say at one of the 
>> limit
>> switch contact points, for each pass that might work).  I also toyed 
>> with
>> the idea of an electric compass for azimuth feedback, but I'm worried 
>> about
>> distortion of the magnetic field near the pedestal due to the AC 
>> induction
>> motors or when the antennas are radiating.  In theory the motors are 
>> housed
>> inside the metal pedestal enclosure and thus are shielded from the 
>> outside
>> world, but I can just see it now, nice steady feedback when the 
>> pedestal is
>> stopped and as soon as I execute a motion command the azimuth feedback
>> starts dancing all over the place.  Since the motion stop command is 
>> based
>> on achieving the target position, system instability is sure to 
>> occur.  Even
>> if I solve the AC motor EMI problem, I still worry that when 
>> transmitting
>> the fields could potentially be distorted if near the antenna 
>> (remember my
>> goal is a compact design) and taint the position feedback.
>>
>>
>>
>> Any ideas from the group would be greatly appreciated.  I'm looking 
>> for a
>> "sparkfun" type solution here and if anyone has experience working with
>> accelerometers, gyros, electric compasses, etc. I would love your 
>> advice on
>> which might be the way to go for the position feedback.  If you think 
>> I've
>> hit on a good idea above and should go with it please let me know.  
>> Again
>> I'm using an Arduino, so analog voltage feedback, I2C, SPI, and UART 
>> serial
>> are all on the table for communicating with the sensors to get the 
>> feedback
>> info.
>>
>>
>>
>> Thanks in advance!
>>
>>
>>
>> Sincerely,
>>
>> Zach, KJ4QLP
>>
>>
>>
>> _______________________________________________
>> Sent via AMSAT-BB at amsat.org. Opinions expressed are those of the author.
>> Not an AMSAT-NA member? Join now to support the amateur satellite 
>> program!
>> Subscription settings: http://amsat.org/mailman/listinfo/amsat-bb
>>
> _______________________________________________
> Sent via AMSAT-BB at amsat.org. Opinions expressed are those of the author.
> Not an AMSAT-NA member? Join now to support the amateur satellite 
> program!
> Subscription settings: http://amsat.org/mailman/listinfo/amsat-bb