Display Degrees of rotation.

[Bobw55]

I know nothing about programming a PIC or circuit design. However, give me a schematic and some breadboard and I am good to go. I want to build a display to show me the position of my antenna rotor. I am planing on using the encoder from a computer mouse. (still doing the math for how many pulses=degrees of rotation. The display would need to remember it's position when powered off. I plan on using a simple push button to reset or home the display. My question for the group is where can I find some info on a schematic or circuit design?
I would like to build this as cheaply as possible.
Thanks
Bob
http://melvinweather.com/

[Archangel]

Hello Bobw55,
here are 2 links, the first is a "special" Google link which provides you the ability to search just the forum or the entire internet. the second is one that search turned up.
http://www.google.com/custom?hl=en&cof=AH%3Aleft%3BS%3Ahttp%3A%2F%2Fwww.picbasic.co.u k%2Fforum%3BL%3Ahttp%3A%2F%2Fwww.crownhill.co.uk%2 Flogo.gif%3BLH%3A37%3BLW%3A174%3B&domains=picbasic.co.uk&q=&btnG=Search&sitesearch=
http://www.picbasic.co.uk/forum/showthread.php?t=7349

[aratti]

I am planing on using the encoder from a computer mouse

Old days mouse encoders are not suitable for your application, since they have very low resolution. they will have no more than 48 pulses per turn. Additionaly you will need two channels so that you will control also direction (I am sure you will need to turn CW and CCW your antenna)

You will need also a reference pulse that will be activated when direction is at 0 degree, so you can adjust possible error in the count.

You will need at least a 360 pulses per turn, yielding a resolution of 1 degree (I am not an expert in directional antenna but one degree seems to me not enough). With an ecoder at 360 pulses per turn two channels, using a technique called "quadrature" You could achieve a resolution of 1/4 of degree, that can be doubled with a mechanical transmission yielding two turns of the encoder per turn of the antenna.

Pic mcu can handle the encoder input and display the direction of your antenna on a LCD.

The hardware is not very difficult, apart the shielding of the signal cables, the software is also not very difficult but not for a beginner for sure!

Al.

[Bobw55]

I do not plan on using the pule disc from the mouse, just the encoder from a newer one. It will be driven via a gear that would provide about 3 rev per single rev of the antenna giving at least 1000 pulses. As far as a zero point, I was figuring on using a push button for that part. I would like the display to be accurate as possible even thought +- 5 degrees is close enough.
I am still planning all of this.

Bob

[cncmachineguy]

a good source for encoders is US Digital (http://usdigital.com/) They are very resonable on pricing and also nice folks. I have been buying from them for about 10 years. They have a wide variety, so I am sure they have something you could use. As for cost, keep in mind what it will cost to scratch something together using an old mouse. You could make it work with 48 pulses per rev, but again, it depends on what you want to spend. I would want at least 1 degree per rev, so for that you would need a 7.5 to 1 mechanical gizmo to spin the mouse encoder 7.5 times faster then the antenna. (Lego's come to mind)

The encoders I have used from US digital all have TTL level outputs, directly readable by the PIC. I am not sure if you can expect a clean transmission from the roof to inside the house. If not, you could have a pic at the antenna send RS232 signals to the house, then read them with a puter of build a unit for inside to display the position.

[cncmachineguy]

This little unit looks pretty nice (http://usdigital.com/products/encoders/absolute/rotary/shaft/ma3/)
It absolute so no need to store the position, 0-5V analog output, 20 mA draw. only $40!!

[Bobw55]

I think the sensor from the mouse will work. It is the disc that I have to change out. If I can elongate the motor shaft to accommodate the disk, I can have all the pulses I need. It will create the problem of having to figure out how many per one rev of the antenna. Not too much room in the housing accept under the motor. I have also thought about using the hall sensors that are inside a CD drive motor. For me, finding a way to get sensor information from the motor is the easy part. Selecting a PIC and LCD display will be the hard part..
Keep the ideas coming!!!

Bob

[Darrel Taylor]

Just a crazy thought ...

Is the rotor's motor AC or DC?

If it's DC, maybe you could count the commutator pulses from the motor current, and not need an encoder at all.

[Bobw55]

It is 2 phase AC. I thought about replacing it with a DC motor. It would be easier to have the display unit control the motor as well. This is a small TV antenna style rotator. Because the control unit uses TIME to allow for movement, the pointing accuracy can really vary, and you are mechanically locked into a max of 360 degrees of rotation. Once I figure out a digital display, I am going to remove the mechanical stop. I would like to under rotate by 90 degrees and rotate past 360 by 90 degrees (540 degrees total).

Bob

[aratti]

How much power do you pump into the antenna ?

Al.

[Bobw55]

I only run 50 watts through LMR400 cable. I thought about RF interference, But that can be solved with shielded cable and some roids. Going to grab the rotor off the tower today so I can see where I can mount some sort of sensor.

Bob

[Bobw55]

Well managed to get a pulse disc attached to the end of the motor shaft today. Will have pulses a plenty. Starter question would be:

1) What would I need minimum to just count the pulses and display them assuming I am using a sensor from a mouse?
2) I can mount up to a 2 1/2" disc with a fare amount of divisions on it Was thinking 20 to start.

Sorry to ask so many questions.

Bob

[Dave]

Bobw55 , I have to ask, What kind of rotor are you trying to retrofit? There should be some kind of encoding device already inside. The old alliance rotors had a switch that produced about 50 or so pulses. Cant you just use that one for starters?

Dave Purola,
N8NTA

[Bobw55]

I am using a Phillips rotor and control. There is no feedback in the motor unit. When the controller energizes the motor, it is relying on a constant motion of the motor (1 rpm). The controller has a 2 digit led display to show the position by 10 degree increments, but this is time based only. In the cold and wind and the added weight of the antenna, the position gets off by 20 degrees on a regular basis. It does have the means to re-initialize the rotor and controller by driving the motor CCW for one minute, which in theory drives the motor against a mechanical stop.

I just love a good head scratch-er!

Bob

[Dave]

Bobw55, Wow Bob, I have never seen one like that... It makes sense though. At 60 hertz with the right gearing... I don't know of any way to get the position back from something like that without being concerned about the weather proofing aspect. I might consider some kind of string pot with the string wrapped around the rotating mast. That way you wont have to calibrate it but only once, plus it will give you a linear output. It's always hard to retrofit something that is already built... If you have it down, I would disassemble it and see if there isn't some way to get a pot connected internal... Personally I perfer a pot as a feed back source because of the absolute aspect. Just my thoughts....

Dave Purola,
N8NTA

[Bobw55]

Well I managed to make and fix a pulse disc to the back of the motor. The mouse encoder reads it, just not with any accuracy. Used the encoder from the mouse in my drill and it works fine.
Dave: I thought about using a multi-turn pot, there is room to mount one, it is finding a gear that would mesh up with the final output shaft. I know the high end rotors use a pot.

I also have a drive motor from a CD drive that has 3 hall sensors mounted about 10 degrees apart. I could use, not sure how to wire it so you can see the change in direction.

Too much time on my hands.

Bob

[mark_s]

Bob,

Maybe this would work, uses a hard drive motor as an encoder.

http://www.fleurey.com/franck/pmwiki.php?n=Main.HDDEncoder

[Bobw55]

Mark,
I have come across that link a few days ago. I think I may use the CD drive motoronly due to it being mounted on a pcb that would lend to easy mounting to the tail of the rotor drive motor, plus it would only require 3 wires to run it. Will try to post pictures of what I a working with later then maybe all of this will make sense.

Bob

[Bobw55]

Thanks for all the replys.
I ended up using a hall sensor and the magnets from a CD drive motor for the sensor part. I used a 16F1827 PIC, a 2x16 LCD and other hardware. I now have a rotor controller that will allow me to go from -90 of North to 360 rotation plus an additional +90 of north 540 total degrees of rotation. To give me a HOME position I used a IR emitter detector with a reflective strip to light an LED on the controller panel. I can increment/decrement the "GOTO" poition by either 1 or 10 degrees at a time. I am really happy.

Bob

[Archangel]

Thanks for all the replys.
I ended up using a hall sensor and the magnets from a CD drive motor for the sensor part. I used a 16F1827 PIC, a 2x16 LCD and other hardware. I now have a rotor controller that will allow me to go from -90 of North to 360 rotation plus an additional +90 of north 540 total degrees of rotation. To give me a HOME position I used a IR emitter detector with a reflective strip to light an LED on the controller panel. I can increment/decrement the "GOTO" poition by either 1 or 10 degrees at a time. I am really happy.

Bob
Hi Bob,
Lot's of Hams on this forum. Your project looks useful as all getout to me, might make a great article or entry into one of this forums contests. Food for thought. If you do not intend to make a commercial product, perhaps you would write it up with code and submit it.

[Bobw55]

I may have to consider that.
I only have a hand drawn schematic.
Here is a picture of the almost completed project. I still need to finish the labeling.

5352

[Ramius]

Hi Bobw55,
As a ham (WB6NSN) I can appreciate the need to know where the antenna is pointed. One other possible solution is to use a compass module such as the R1655. Here is a link with everything: http://www.robsonco.com/Dinsmore/Untitled_5.html

For me, I slipped the sensor into a plastic tube and sealed it.
The accuracy is 1/10 of 1 degree.

5353

[Darrel Taylor]

Hi Bobw55,
As a ham (WB6NSN) I can appreciate the need to know where the antenna is pointed. One other possible solution is to use a compass module such as the R1655.

What a great idea!

[Dave]

Ramius , My only concern would be the close proximity to the antenna during transmit. The flux coils may be over driven....

Dave Purola,
N8NTA

[Ramius]

Hi Dave,
Actually the R1655 sensor is actually 2 HAL effect transistors and a permanent magnet so there are no coils. The compass works next to a 5 watt 1.2GHZ transmitter with no effect. On the compass PC board the sensor is mounted right on top of the A/D converter to keep the connections as short as possible just to avoid any noise since the compass is in an R/C submarine with a lot of DC motors and there is a bypass capacitor on the same board with the sensor and A/D. I used two each 3-wire shielded audio cables between the compass and the PIC. I believe you would be just fine and you could also include some ferrite beads if you wish. Hope this helps, Ed

[Bobw55]

I had thought about using a compass sensor. Still like the idea. Most I found were on the expensive side, not to mention I was just getting my feet wet messing with PICs and programming. Should I ever decide to redo things that may be the way to go. Would like to see some pictures as to how you mounted yours.

Bob