0
$\begingroup$

I have a motor attached to a spool, attached to a string. The string is attached to a rod that is able to move backwards and forwards freely but when the rod is pulled back too far it compresses a spring. I can pull the rod in by winding the spool and I know when to stop because of the IR sensor. The locking mechanism then locks the rod in place and holds the rod until I want it to be released. See diagram below.

Setup ......

The problem is I know how far to pull the rod back because of the IR sensor, but I don't know how far to release the rod. If I want the rod to fully extend, I could just put another IR sensor at the end and unwind the motor until the IR sensor at the end detects nothing, but the rod might hit something at a random point along its travel (I am making this actuator for this specific purpose, to hit something). So I don't know how far to unwind the motor, If I unwind the spool too much without the rod moving, there will be no tension on the string and the string might get knotted as it will bundle up and not get pulled tight. If I don't unwind the motor enough, there will be too much resistance on the rod and the rod will not fully extend as I want it too.

Is there a mechanism that lets me pull the rod in and then have the string disengage from the motor shaft so there's no friction so the rod can extend freely? I've looked into this and normally a clutch is used to engage and disengage from a motor, but this is the motor I'm working with and commercial clutches are for much larger motors. I've also found posts that try and do the same thing but they having provided any useful answers. If there's a way that I can detect the motion of the rod to tell when it has stopped, that would be useful too as I can unwind the motor until the rod stops moving. Hope it makes sense, thanks

$\endgroup$
2
  • $\begingroup$ a clutch may be an unnecessary complexity ... why would unwinding the string not be enough? $\endgroup$
    – jsotola
    Nov 14, 2023 at 16:39
  • $\begingroup$ @jsotola, how would I know how much to unwind? Lets say I use what I suggested in the post, an IR sensor at the end to detect when the rod has reached the end of the stroke. If I tell the motor to keep unwinding the string until the rod reaches the end (detected by IR sensor), what if the rod hits something along it's travel and the motor just keeps unwinding the string? Then the string will get caught on the spool again and the rod will get pulled in even though it hasn't reached the end of the stroke $\endgroup$
    – amir
    Nov 14, 2023 at 22:31

3 Answers 3

1
$\begingroup$

I would run the string over a roller-lever microswitch to detect string tension. This is the same principle of operation as many "dancer-controlled" unwind systems used in industrial machinery. When the string is taut the switch is closed and the motor can run. When the string is slack the switch will release and the motor can stop.

The problem with this approach is that you may only be able to get it to work well for a specific rod speed. Your post doesn't explain what controls or limits the speed of the rod.

$\endgroup$
1
  • $\begingroup$ What controls the speed of the rod when extending is the spring that's under compression, and that should be the only thing controlling speed when extending (there should be no resistance from the motor). What controls the speed of the rod when retracting is the motor that's pulling the rod in $\endgroup$
    – amir
    Nov 14, 2023 at 22:34
0
$\begingroup$

enter image description here

Figure 1. Converting the drive to a closed-loop capstan arrangement.

If mechanical constraints allow it, looping the string around a forward pulley ensures that it never goes slack. This prevents any over-run on the unwind. Multiple turns around the motor shaft ensures adequate grip much like the capstan arrangement on a ship. Some sort of tensioner may be required.

If you really require zero (or very low) resistance during rod extension then have a look at Sprague or one-way clutches for the capstan. These work rather like the freewheel on a bike's rear wheel but typically without the clicking of a ratchet.

$\endgroup$
14
  • $\begingroup$ So when the motor is unwinding, it's also pulling the rod forward using the pulley? Is that right? But the reason why I'm using a spring to extend the rod and why I'm not using the motor to extend the rod is because if the rod hits something along it's travel, the motor will be trying to force the rod forward and the rod won't be able to move which will most likely break the hardware $\endgroup$
    – amir
    Nov 15, 2023 at 15:42
  • $\begingroup$ No, the spring pushes the rod forward feeding the string around the closed circuit. This causes the capstan to rotate on the one-way clutch and the motor shaft doesn't move. You could try it with the one-way clutch and without the loop but it might overrun a bit when the rod stops and there would be some slack in the string. The motor then only ever has to turn in the rewind direction. $\endgroup$
    – Transistor
    Nov 15, 2023 at 15:50
  • $\begingroup$ Can you explain more on how the one way clutch would let this work without the loop, I understand one way clutches let rotation in one way but not the other. And how does extending the rod wind the string around the capstan. Thanks $\endgroup$
    – amir
    Nov 15, 2023 at 17:43
  • $\begingroup$ Extending the rod unwinds the string around the capstan while the one-way clutch allows it to freewheel. Starting the motor engages the one-way clutch and winds up the string. $\endgroup$
    – Transistor
    Nov 15, 2023 at 19:11
  • $\begingroup$ Wouldn't you only be able to wind it? How can you wind and unwind the string if the string is connected to same one way clutch? $\endgroup$
    – amir
    Nov 15, 2023 at 20:56
0
$\begingroup$

Here's a handful of ideas of ways you can detect loss of tension in the cable:

  • I really like the previous comment about the dancer sensor. I think that would work well here. I attached a sketch. sketch of tension sensor
  • monitor motor current
  • use a conductive cable and run the wire through a short tube. If you ever see continuity, you lose tension
  • add a reel encoder
  • the spring should provide a reliable speed, so you might could use a timer in your control program and if you don't see the "extended" IR sensor after X seconds.

Next, are you wanting to launch the rod using the spring? For that, a very simple trigger mechanism could be designed. If the rod is recoverable, just put a spring loaded pin that is incerted which links the cable to the rod, then have an escape hatch for the pin to spring out of once you have reach a certain level of spring compression. *****this might not work because of the potential forces involved on the pin which might cause binding.

  • attach the cable to a small electromagnet which travels the length of the tube. Then, when you insert the rod, energize the magnet to hold the rod to the cable, release the magnet when you are ready to let go. The electromagnet would need to be smaller diameter than the rod and the spring should only engage with the rod. The electromagnet should pass through the center of the spring.
  • cut the string (lol)
$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.