Shop OBEX P1 Docs P2 Docs Learn Events
Linear acuator limit switch questions — Parallax Forums

Linear acuator limit switch questions

This device has two wires to drive it, just reverse polarity to go the other direction. It shows a limit contacts at #14 a Limit Switch Bar at #7. Since the device stops working in each direction when it hits the end of travel switch area, and since it can still reverse, this means they are using some diodes in the limit scheme to stop the flow of current in one direction only. I want to use the device in conjunction with other motors and this one needs to complete and be sure it completed before proceeding. I was thinking the power to the actuator for one direction could be a DPST relay that connects both GND and +V. The device will have a known travel time, so the Prop could tell it to move by setting the relay, then wait until slightly after the known travel time to turn off the relay. Once the device hits the limit I am assuming that no more current will flow in that direction and then it could be possible to switch in a +V on the same + side we were just using on the relay and check of a voltage on the other wire of the actuator. If the device has arrived then the path will be completed and the prop can see the voltage ( maybe use a divider and still use 12V). If the voltage is not present on the test then shut down and warn of a fault. There is no other easy way to rig up another switch or sensor to detect the travel completion in this application and I thought why not use the built in diode/limiting scheme. Since I don't have that device I was wondering if this logic makes sense.

lin.jpg 71.8K

Comments

  • I'm not really sure of what you are trying to accomplish and you don't have a schematic of your switch and motor layout. A reversing relay would reverse the travel and limit switches would stop it at either end. You could add an encoder to the lead screw or motor to track the position. But you will need to start the motor and run it to a home limit switch so you would know where your starting position is like a CNC setup.
  • Yeah that is not possible, this is a concealed device I found online. There is no chance of adding any encoder to it and that is to complex for this application. This may not be the best device and I am looking at another option.
  • A linear potentiometer might work that is attached to whatever this is moving. Check out the rotary
  • idbruceidbruce Posts: 6,197
    edited 2017-04-15 19:13
    T Chap

    If you remove the end cap #19 and can see the end of the lead screw, you could drill and tap the end of the lead screw and install an extension shaft for an encoder.

    Let's assume that it is a 1/2" lead screw. You could drill and tap it for 10-32 threads. Then insert a high precision shoulder bolt with 10-32 threads, 1/4" shoulder, and a 2" shoulder length. Cut the head off, drill the end cap, and install a rotary encoder. Mount a slotted optical switch to the end cap.

    Doesn't sound too difficult, providing you can see the end of the lead screw.

    https://mcmaster.com/#94035a189/=177m0x7

    EDIT: On the other hand, if the lead screw is hardened, then I would definitely look for an alternative.

  • kwinnkwinn Posts: 8,697
    If all you want to do is drive the actuator all the way from one side to the other without stopping in between then a DPDT relay will do the job. If you want to stop between the ends then some method of removing power such as another relay or transistor is needed. If it is the latter case an H bridge driver may be simpler.
  • No what I was trying to find out is whether there was some way to use the existing diode limit switch scheme to test whether the device was fully extended or fully retracted. I have abandoned this idea for now.
  • kwinnkwinn Posts: 8,697
    T Chap wrote: »
    No what I was trying to find out is whether there was some way to use the existing diode limit switch scheme to test whether the device was fully extended or fully retracted. I have abandoned this idea for now.

    Zero current tells you if it is at one end or the other. Polarity to the two wires should tell you which end.
  • Phil Pilgrim (PhiPi)Phil Pilgrim (PhiPi) Posts: 23,514
    edited 2017-04-16 03:54
    Here's a circuit diagram to do what you want:

    (In the next post. Dang forum software!)

    The two diodes in parallel with the limit switches are part of the assembly. The other four diodes form a bridge that provides an output that's high (at motor voltage) when the motor is running, zero, otherwise.

    -Phil
  • kwinnkwinn Posts: 8,697
    By adding a couple of diodes to the double pole relay as per the attached diagram you should be able to sense the +12V on the side opposite to where the +12V is coming in to the motor when the relay is open. The polarity of the diodes on each side of the motor must match for the circuit to work.
    850 x 616 - 15K
  • ercoerco Posts: 20,244
    All good ideas. Could also use any type of current sensor in the 12V positive supply wire to the actuator, before the reversing relay.
  • FWIW: A part of what I do is to take existing "dumb" linear actuators and make them "smart" and there are usually space limitations.

    "Yoyo" / "Stringpot" Encoder
  • I was looking for something similar to an electronic deadbolt for a certain application where I need a pin to move out .75", but the program needs to verify if the pin has moved and if the pin is in the retracted or extended state. I have used linear actuators before and they would serve the purpose but would require some effort to get the position out of it without adding external hardware. I found the perfect solution, an electronic deadbolt with a built in lock status output. I had looked before but didn't see one with the status output, then found some yesterday that do include it. All great ideas above, nice info for future use with the linear actuator.
    589 x 190 - 15K
    aa.jpg 14.6K
  • a
    Mickster wrote: »
    FWIW: A part of what I do is to take existing "dumb" linear actuators and make them "smart" and there are usually space limitations.

    "Yoyo" / "Stringpot" Encoder

    Yep, I tried to add a linear slide pot and no go. $30 w/out feedback, $130 w/ feedback. I don't like the idea of an encoder on the motor shaft, I would want absolute position without having to home anything.

    There are still some options I'd like to try. I ordered a flex sensor, I'm going to try to add that. If the actuator is all the way to the "left" there would be no flex, as it gets closer to the right end stop, there would be max flex. I hope it works.



    1210 x 908 - 271K
Sign In or Register to comment.