You should be able to cycle several times to see if it gets out of step.

The old satellite actuators display the count.

The hall effect sensors were very sensitive to adjustment.

Also, I don't know if you are running at the same speed as original counter.
If it's like the sat. actuators there will be a narrow window of pulses per minute that it will respond to.

They ignore others as noise.

It is getting out of step, I was writing it off as over run and such, but I wonder if it is the sensor. The original rotor motor ran in the 6000-6500 rpm area. The motor I am using is 1526 rpms, but I used an 8 pole magnet instead of the 2 pole that was on the original. So the sensor should be seeing 6104 rpms.

 

If you changed the motor but did not change the gearing, then the pulse/distance has not changed just by using a different rpm motor?
IOW, why what was the need to change the pulse count?
This is if the sensor is related to distance.
What was the nature of the 6000rpm motor? P.M. servo?
Max.

 

If you changed the motor but did not change the gearing, then the pulse/distance has not changed just by using a different rpm motor?
IOW, why what was the need to change the pulse count?
This is if the sensor is related to distance.
What was the nature of the 6000rpm motor? P.M. servo?
Max.

Maybe I need to explain a little better what I did. I bought an antenna rotor and control. The rotor motor itself is a very small 19v DC motor that turns in the 6000-6500 rpm area, this is what the rotor control measures with a Hall Effect pickup to track location.

The stock little motor has very little power and was of no use for what I am doing, but the control system was just what I wanted. So I used my own AC motor and built a copy of the Hall Effect board that was on the original motor to use on my motor, that is what is in the picture above. I hook my board to the original control board. Since the original motor was 6000+ rpm I used an 8 pole magnet on my motor to match as close as possible to the same rpm (1526rpm x 4 = 6104rpm). So as far as the control board knows it is turning at the RPM range it expects to see.

 

Here is a partial drawing to show the simplification when using a 3 state buffer [ MM74HC241 ] to input data to acomparator. Programming by DIP SW, solderless breadboard, or strip board. 4 binary programmed positions , 21 cm, 35 cm, 75 cm, & 183 cm are shown for illustration. The buss lines for the comparators could also be made from strip board, although I have never used strip board.

 

Maybe I need to explain a little better what I did. I bought an antenna rotor and control. The rotor motor itself is a very small 19v DC motor that turns in the 6000-6500 rpm area, this is what the rotor control measures with a Hall Effect pickup to track location.

The stock little motor has very little power and was of no use for what I am doing, but the control system was just what I wanted. So I used my own AC motor and built a copy of the Hall Effect board that was on the original motor to use on my motor, that is what is in the picture above. I hook my board to the original control board. Since the original motor was 6000+ rpm I used an 8 pole magnet on my motor to match as close as possible to the same rpm (1526rpm x 4 = 6104rpm). So as far as the control board knows it is turning at the RPM range it expects to see.

I think you were wise to do that, as the sw may be written to expect certain timing. At least that's how I would do it.

I believe it would be helpful to run all the way to one end and mark position exactly.
Then cycle to the furthest point you can program and mark where it coasts to a stop.

Repeat this a few time manually adjusting to the same start point each time.

The only difference you should see are the count losses.
If it won't stop exactly in the same place each time I would put the blame on miscounting.

If you have a large coast time, the coast error could make test meaningless, but it's a place to start. IMO

 

The satellite controllers work with a much slower count as the shaft holding the magnet wheel runs the same speed as the screw drive. ≈60 RPM.

With a dc motor, braking, and low inertia it will stop on the same magnet every time.

It has to as these were expected to work for months at a time. About 4 counts accumulated error and the CX would be calling.

I just wish I could find a Toshiba for you. They were great. Actually, I have one to give away, but can't find a remote.

Edit:
This seems to be the latest stand alone, remote controlled, 36vdc, pulse input, programmable, positioner.
http://www.sadoun.com/Sat/Order/Motors/VBox7-positioner.html

 

I have been doing a lot of reading and learning the last couple of days and think I have a handle on what I want to do. It is based on what Bernard suggested except I am going to use micro switches instead of a code strip. Basically I have 7 switches, one at each point I want to stop along the track. Each switch will store a 4 bit binary number.

Input another 4 bit number with the use of an IR remote (haven't figured that part out yet.

Use a comparator to compare the two numbers, this will trigger left, right, or stop based on < > =

Now the problem. I am trying to design this in Multisim, but for the life of me I can't seem to get a diode to work like I think it should. I always get reverse voltage through them for some reason. See in the simple circuit I attached how their is 12v at probe #2, why is that?

 

It is good to do some independent thinking, but I think it's back to the drawing board. A switch closure only conveys one bit of information, no left or right, just stop. You can use a remote control to select which switch is to be obeyed. Quick unstuddied remark: It may be possible to use segments of code strip, say covering 2 in. of 16 bits for each of the 7 locations. Remote would select which switch is active, but also need directions as to start & which direction to go. When selected SW is reached, motor stops, code reader activated, directing motor to move to center of code strip & stop. Only one 4 bit comparator is needed so resolution could be upped to 1/8 in.??

 

Seems OP is on the right track.
If all the extra wiring is ok.

Each switch can be input to controller separately and assigned a number.

Actually two switches in parallel might be better. With motion stopping between.

One for each direction. Coasting to a stop after contact. Switches could be adjusted for coast time. Or they could be far enough apart to delay stop in sw. Then it could be fine tuned in sw without mechanical adjustments.

 

Now the problem. I am trying to design this in Multisim, but for the life of me I can't seem to get a diode to work like I think it should. I always get reverse voltage through them for some reason. See in the simple circuit I attached how their is 12v at probe #2, why is that?

I don't understand the context. " in Multisim"

But if it's a high impedance reading. Then add a high value resistor to ground. 1meg or so at point 2.

There will be voltage measured at that point until diode is switched off.
You shouldn't need it in actual construction.

 

It is good to do some independent thinking, but I think it's back to the drawing board. A switch closure only conveys one bit of information, no left or right, just stop. You can use a remote control to select which switch is to be obeyed. Quick unstuddied remark: It may be possible to use segments of code strip, say covering 2 in. of 16 bits for each of the 7 locations. Remote would select which switch is active, but also need directions as to start & which direction to go. When selected SW is reached, motor stops, code reader activated, directing motor to move to center of code strip & stop. Only one 4 bit comparator is needed so resolution could be upped to 1/8 in.??

I have it working in multisim using switches running to a BCD to 7 segment decoder to verify it works. Using the switches to send power to 4 inputs you can make a 4 bit number. Inputs A B C D

#1 makes A high B&C&D Low - so their are 4 bits of info

#2 makes B high
#3 makes A&B high
#4 makes C high
#5 makes A&C high
#6 makes B&C high
#7 makes A&B&C high

The comparator reads the 4 inputs as the binary number 1-7

I have verifies that the comparator will power left of right and stop when the numbers are equal.

Problem I am having is getting the switch to send power to multiple inputs without allowing power back through when only one input is being powered. A diode should work but I can't seem to get them to act like they should.

 

I don't understand the context. " in Multisim"

But if it's a high impedance reading. Then add a high value resistor to ground. 1meg or so at point 2.

There will be voltage measured at that point until diode is switched off.
You shouldn't need it in actual construction.

Multisim is the name of the simulation program that I am using

 

Multisim is the name of the simulation program that I am using

Try the pull down.

 

But if it's a high impedance reading. Then add a high value resistor to ground. 1meg or so at point 2.

There will be voltage measured at that point until diode is switched off.
You shouldn't need it in actual construction.

This did work perfect, I don't understand it but all switches out put as I expected them to.

Only thing I have to figure out is getting the 4 bit binary number into the comparator by way of an IR remote now.

 

What about coasting?

Do you think the error of approaching switch from different directions will be minimal?

I haven't read all. Still ac motor?

 

What about coasting?

Do you think the error of approaching switch from different directions will be minimal?

I haven't read all. Still ac motor?

Yes I will beable to keep all of my A/C motor and motor control like I have it now. Just replace the rotor control with this new circuit. Coast will not be an issue and is very minimal. It is the multiplying effect of multiple coast over time that was the issue. Using switches the will be no multiplication effect. All my motor control require is a switch in polarity to reverse direction. So all the circuit needs to do it put out negative DC for one direction and positive DC for the other, nothing for stop.

I plan on running a double switch. One will be used to tell the location that it is at, the other will latch the location in memory. It will hit the location switch first riding on a ramp for maybe 1" or so. Next the latch switch will hit, it will have very steep ramp to activate it so it is only clicked for an instance, this will eliminate the problem of approaching from different directions. The location is only read when the latch is hit, so that will be the point where it will stop also.

Once the latch switch is hit the location is held in memory until it is hit again. This way even when in-between locations it will know where it last was.

 

Got it. At first I thought two switches redundant. As it can "remember" last location.

But the way you are doing it, with the location switch, it will not have to remember. So even at power up it will know where it's at.

 

Got it. At first I thought two switches redundant. As it can "remember" last location.

But the way you are doing it, with the location switch, it will not have to remember. So even at power up it will know where it's at.

I got everything in the simulation program now and all seems to work perfectly.

My last hurtle is trying to do something with get and IR code in the mix. I have found some Sony decoders that would probably work but I am already using Sony codes for another piece of equipment in the room, so I can't go that route.

 

Do you have 7 more Sony codes available?

 

Do you have 7 more Sony codes available?

I am sure I do but would rather not. And I found this which will solve all my problems I think.

http://www.tauntek.com/tinyir2-learning-ir-remote-control-receiver.htm

Also doing some more looking I found what is called a 16 Key encoder. If I read it correct you power one of 16 inputs and it out puts the binary number of that input. This would make wiring the switches easier. Also Can use along with above ir reciever.