[amsat-bb] Optical shaft encoders

Thu Mar 24 13:43:41 UTC 2016

John,
I'm testing this board with my custom rotators and controller.  The board
has been working quite well, with consistent reading for both az and el.
The az readings are even accurate when the el is at 90 degrees.

That said, I won't be using it as a primary position sensor.  Optical
encoders with sufficient resolution are still superior.  I haven't tested
this board in the presence of RF, so I'm not sure what it will do on
transmit.  The board also can't tell you if the rotators are pointing at
360 degrees or 0.  A POT, or something similar, is still needed to
determine that.  The i2c bus needs to be as short as possible, so whatever
is reading it needs to be mounted on the rotators.  Positioning the
controller with the rotators will also improve the reliability of the
optical encoders.  This board works well for calibrating the rotator to
magnetic north and zeroing the el, as well as validating the az & el
positions reported by the optical encoders.  How well the validation will
work on transmit is yet to be determined.

If all you want to do is read the position outside the controller circuit,
then I would think it would work quite well connected to an Arduino mounted
on the rotator.  An ethernet shield would allow for a simple web page that
could display the rotator's position.  Reading the values off the board is
quick and doesn't require very much resources.  Calibration is the main
issue with these types of sensors.  It's much easier to deal with on this
board.  The on-board micro helps a great deal.  The calibration data is
lost when the board powers up.  Until it is calibrated, the values are
useless.  Once calibrated, previous saved calibration values can be passed
to the board when it powers up.  The initial calibration does call for
specific movements outside the range possible by just mounting it to the
rotator and moving it around.  The initial calibration would need to be
done close to the actual mounted position on the rotator.  Once the rotator
and sensor are in their final positions and the rotator az and el moved
around a bit, the on-board micro will update it's calibration values from
the initial calibration.  These final calibration values would be the
values to save and re-apply on power up.  Since the on-board micro
continues to update the calibration, it might be a good idea to update the
saved values periodically.

Steve
KD8QWT

On Wed, Mar 23, 2016 at 5:58 PM, John Toscano <tosca005 at umn.edu> wrote:

> Bob et. al.:
>
> Would something like this:
>   https://www.adafruit.com/products/2472
>
> ...be useful for the purpose of determining absolute position without
> relying on mechanical linkages of gears, belts, pulses, potentiometers,
> etc.? It uses 3-axis magnetometers, 3-axis gyros, and 3-axis
> accelerometers, and a Cortex M0 ARM to do all the hard computations that
> combine the three types of data into a viable position indication. It is
> inexpensive enough that I will probably install one when I finally get
> around to raising a tower at my new QTH, and using it only for readouts at
> first. I have been using and will resume using Yaesu rotators when the
> tower goes up, so no urgent need to add safety limit switches etc. In many
> years of operation at my prior QTH, the rotator never got "confused" about
> its position except when somebody put tension on the cable plugged into the
> back of the control box and loosened some of the connections between plug
> and socket. I do realize that even the G1000SDX model I was/will be using
> is not hefty enough for some applications, but again it served me well for
> many years in spite of temperature swings from -40F to +105F. (Yeah, it
> turned a bit slower at the begining of a contest when the temp was -40F and
> the wind chill was -60F. OK, wind chill doesn't apply to inanimate objects,
> but -40F actual is pretty darned cold. Plus the ice and snow added on
> didn't help.)  :-)
>
> 73 de W0JT/5
> EN34js -> EL09vu
>
> On Tue, Mar 22, 2016 at 8:53 PM, Robert McGwier <rwmcgwier at gmail.com>
> wrote:
>
> > This is an awesome conversation and we are extremely grateful.
> >
> > Zach is trying to build a robust repeatable super ground station.
> >
> > Bob
> > On Mar 18, 2016 6:30 AM, "Daniel Cussen" <zg3410 at gmail.com> wrote:
> >
> > > >for those with Alfa Radio HR model rotators, maybe you have
> same/similar
> > > issues?
> > >
> > > I do not have a HR model, but I do have a similar system and we are
> > > seeing similar problems.
> > >
> > > First off there is no end stop safety switches. This means if the
> > > control box becomes confused it can damage coax cables and move
> > > elevation to positions to damage antennas etc.
> > >
> > > Second, there is no absolute position sent, meaning if it gets
> > > confused it cannot reset itself as only the amount of movement is
> > > sent, not the actual current position. Over time this means errors
> > > accumulate and grow.
> > >
> > > The basic Yaesu G5500 has both safety protections meaning it is
> > > unlikely a confused control box will result in damaged antennas or
> > > coax.
> > >
> > > >So first off, Bob is basically asking if anyone has built a custom
> > > optical shaft encoder to >replace the magnetic hall effect sensors in
> the
> > > High Resolution Big-Ras rotators.  >Machining, circuit design,
> > > performance.....?
> > >
> > > I do not know, however I have nearly added safety switches to a
> > > similar model to turn off the motor to protect the coax cables. This
> > > is also very important if the relays stick in the control box.
> > >
> > > Others have removed or replaced position sensors with more accurate
> > > absolute positions sensors. In particular the HH-12 is used by a lot
> > > of large EME stations as it is both accurate, cheap and absolute
> > > position. If you connect it using a rubber hose it will protect itself
> > > from damage if you try turn it too much.
> > >
> > > http://www.vk5dj.com/hh-12.html
> > > Mounting it to the Big-Raz is unknown, but here is a similar project:
> > > http://e-kutz.eu/seite10.html
> > > Here is a complete controller with two sensors:
> > > http://f1frv.free.fr/main3o_AZ_EL_Display.html
> > >
> > > >The noise voltage was 1 or 2 volts peak to peak when measuring the
> lines
> > > with an o->scope.
> > >
> > > This is not good.
> > >
> > > > The shields of the cables are connected together at the connector on
> > the
> > > rotator (8 pin >MIC connector) and at the connector on the MD-01
> control
> > > box.  The shield is also >jumpered to a good station ground at the
> > control
> > > box.
> > >
> > > To reduce noise the recommendation it to only connect the shield AT
> > > ONE END, and not both ends. Normally only at the shack end. I think
> > > this prevents ground loops.
> > >
> > > >So that's about it in a 'nutshell.'
> > >
> > > So in my case we have a similar sensor. All the HAMTV ground stations
> > > in Europe (6) are using Prosistel Az/El with uses hall magnetic
> > > sensors with thousands of pulses per second. We too are seeing
> > > positions change with the motor stop, so much so thousands of pulses
> > > must be read while stopped.
> > >
> > > The solutions we have used so far:
> > > 1) Multiple screened cables as you suggest grounded at one end
> > > 2) Adding filters to the motor wiring to reduce motor noise/cross talk
> > > 3) I experimented with stepping up rotation feedback from 0/5V to 0/20V
> > > 4) I am working on a replacement control box, where we can modify the
> > code
> > > 5) I am adding safety switches to protect the coax.
> > >
> > > The real solution would be for the manufacturer to use absolute
> > > position feedback. There is complete controllers available, if you can
> > > manage to connect their sensors to your existing system using
> > > belts/gears/cogs etc.
> > >
> > > Another option I think you should consider is using your own control
> > > box. We found that the Prosistel supplied control box was flawed in
> > > some ways. The open source control box is already designed to take
> > > pulse inputs, and seems to work with thousands of pulses per second.
> > > You can even just hook the inputs in parallel to see if the problem is
> > > the control box or the sensor outputs. All you need is an arduino
> > > (mega preferably) and the correct version of the code. It will display
> > > a second opinion of the position, so you can determine if the control
> > > box has issues too. We found our control box misses some pulses, we
> > > think it is busy updating the LCD or talking to the computer and
> > > misses pulses.
> > >
> > > Other features of this is a master/slave option, meaning the
> > > controller can be mounted at the antenna, meaning only short cable
> > > runs to the position sensor.
> > >
> > > http://blog.radioartisan.com/yaesu-rotator-computer-serial-interface/
> > >
> > > In the long term absolute position sensors combined with safety
> > > switches are the only real  solution. For HF even a bog standard
> > > potentiometer would probably work better. Some suppliers use 3 or 10
> > > turn potentiometers to allow 360 degrees or 1.5 turns rotation.
> > > _______________________________________________
> > > Sent via AMSAT-BB at amsat.org. AMSAT-NA makes this open forum available
> > > to all interested persons worldwide without requiring membership.
> > Opinions
> > > expressed
> > > are solely those of the author, and do not reflect the official views
> of
> > > AMSAT-NA.
> > > Not an AMSAT-NA member? Join now to support the amateur satellite
> > program!
> > > Subscription settings: http://www.amsat.org/mailman/listinfo/amsat-bb
> > >
> > _______________________________________________
> > Sent via AMSAT-BB at amsat.org. AMSAT-NA makes this open forum available
> > to all interested persons worldwide without requiring membership.
> Opinions
> > expressed
> > are solely those of the author, and do not reflect the official views of
> > AMSAT-NA.
> > Not an AMSAT-NA member? Join now to support the amateur satellite
> program!
> > Subscription settings: http://www.amsat.org/mailman/listinfo/amsat-bb
> >
> _______________________________________________
> Sent via AMSAT-BB at amsat.org. AMSAT-NA makes this open forum available
> to all interested persons worldwide without requiring membership. Opinions
> expressed
> are solely those of the author, and do not reflect the official views of
> AMSAT-NA.
> Not an AMSAT-NA member? Join now to support the amateur satellite program!
> Subscription settings: http://www.amsat.org/mailman/listinfo/amsat-bb
>

-- 
"In a time of universal deceit, telling the truth becomes an act of
rebellion."
--George Orwell