Ideas for Positional Switching Accuarcy 1

[aschartner1]

Hi Everyone,

I am working on a home project and I have hit a wall and wanted to get some opinions from everyone else. Essentially what I have is a carousel (eight separate compartments) that is rotated by an electric motor turning right in the center. The way I have it now is the electric motor comes on for a set duration of time when the relay triggers. The time duration is supposed to be set to move the carousel ideally from the center of one compartment to the center of the the next. I need to find a more accurate way to do this though as it varies too much as more and more weight is removed as with each rotation a compartments contents are dropped. I was thinking some type of positional sensor of some sort? Magnetic maybe? I am getting confused and have been staring at this too long. I could implement a micro controller but I am trying to keep it simple and the cost low. I appreciate any insight anyone can offer.

Andrew
(A frustrated home project engineer)

 

[aschartner1]

To add to the previous post and sum it up simply, I would ideally like for when the relay comes on for the motor to turn 45 degrees then to stop regardless of whether or not the relay is still on. Then the next time the relay comes on it move another 45 then stop and so on and so forth.

 

[OperaHouse]

The basic method is to have a detentin the table a microswitch falls into. This unpowers the motor and it stops. The timer onlu needs to be om long enough to get the micro switch out of the detent.

 

[aschartner1]

That is a good thought. That makes me think what about reed switches? The only thing with a reed switch I wonder is when it reaches the magnet and the motor shuts off when the relay comes on again later will the motor start up and go to the next reed switch or will it be stuck because it is still over that reed switch that initially stopped it?

 

[OperaHouse]

A reed switch is not going torun a motor without support, at least toolong. You only need one switch and eight detents, Thought you were a mechanical guy. They like detents. The detent should be a sharp dropoff and long rise.

 

[aschartner1]

Just trying to widen my horizons to electrical/controls and turn to the dark side! The thing with detents that maybe I am overlooking is when it moves from say point A to Point B into the first detent and then stops, when the relay comes back on to move it to the next detent will it be able to move it out of that first detent it is currently in?

 

[OperaHouse]

My suggestion is for a sequential index. The detent switch is in series with the motor. lets say each index takes 10 seconds and the detent space off is the same as 2 seconds. Timer contacts are parallel with detent switch. If timer turns on for 3 seconds, the motor will have enough time for microswitch to get out of detent. It will continue to run till it reaches next detent. You have not given enough information for me or anyone else to suggest a solution. I gave a possible solution to get the conversation started. You have some splainin to do if you want more help.

 

[aschartner1]

Thank you for the help this helping me a lot more now and makes more sense. It is a food dispenser I am playing with making. Essentially I planned to have each container have a set amount of whatever food and figure the time required to turn the carousel so that the container crosses over a hole and dispenses it. The problem I ran into is as the carousel turns, the weight restriction becomes less and the motor turns a little further than it did before. Eventually the error compiles and it ends up too far off. That is when I realized I needed an accurate way for doing positional rotation but I was confused with using switches. I assumed with switches that when I turned to the point I wanted to stop and the switch tripped, then it would stay tripped off reagrdless of the relay cycling on and off. I thought I would need to manually push it through that switch region and that of course is counter productive to what I am trying to do. I am going to try that and I will bug everything if I fail at it. Thanks for the help, my apologies for not being thorough.

 

[jraef]

It may not be as simple as just picking the right sensor. Accelerating, decelerating and indexing a platen that has a widely varying mass is skirting around the edges of a motion control engineering situation. You will need to have the motor provide enough torque to accelerate the peak mass, but that will then likely have too much torque for accurate indexing when the mass is at it's nadir. If you are trying to do this by directly coupling the platen to the motor shaft, you are exacerbating the problem.

Before going the complicated route of stepper or servo motor control, I would start by using a small high speed motor and a worm gear to drive the platen, because they have the lowest backlash. The high gear ratio will provide you with the necessary torque to accelerate the platen when full, but the low speed will mean that slop in the motor turn-off time will not have as much effect on the positioning.

My guess as to the nature of your machine: vacation pet food dispenser... which, not to burst your bubble, already exist. And what they do is move the COVER, not the food tray. Less mass and it doesn't vary.





"Will work for salami"