Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Solenoid design: Can a short solenoid have a long stroke?

  1. May 28, 2012 #1
    Hey, I'm wondering if a short solenoid (50mm) would be able to pull a 300mm core through it if it was open at both ends? Or would the soleniod stop acting once the core had come to the far side?

    I've attached a pic of what I'm trying to do (forgive the paint drawring).

    I'm basically trying to have a solenoid that I can control the position of by varing the voltage (by pwm).

    1.) Can anyone tell me if this wont work?

    2.) Where can I get some basic info on how to calculate the force I'll get out of it? I looked here: www. ledex.com/basics/basics.html but it wasn't amazingly helpful for formulas, and I can't seem to find any on this site.

    3.)The page I referenced in 2.) said that
    does this mean that a tapered core would provide more force as it got thicker? What does it mean that the effective air gap changes to become a fraction of the actual stroke? Does the air gap have a relationship to the stroke that can be achieved?

    Thanks heaps for any help, these forums are fantastic.


    Attached Files:

  2. jcsd
  3. May 31, 2012 #2


    User Avatar

    Staff: Mentor

    I'd be surprised if a solenoid could be depended on to move a core more than about 85% of its own length. I think you'd need to incorporate a lever arrangement for greater movement.

    Where did I get the figure of 85%? It's just a guestimate. :wink:
  4. Jun 1, 2012 #3
    If that's the case, then it makes things difficult. What about the idea of a tapered core? Wouldn't that create a higher field strength as it got thicker and the air gap reduced?

    Has anyone tried anything like this before?
  5. Jun 1, 2012 #4


    User Avatar

    Staff: Mentor

    Are you planning to incorporate position feedback, to ensure the core is located close to where you'd hope it would be?

    If you explain the goal, someone might be able to suggest a more practicable solution.
  6. Jun 1, 2012 #5


    User Avatar
    Science Advisor

  7. Jun 3, 2012 #6
    It's difficult to explain, so bear with me. It's going in a product for commercial sale so I need an inexpensive solution. I've got a weight that I want to 'bounce' at different rates and heights controlled by a microcontroller. The weight varies from consumer to consumer so it needs to be able to accomadate a variation without adjustment. The weight could be anywhere from 8-12kg per unit with multiple units working togeather.

    To do this I was going to use a spring to take up most of the weight (whilst allowing for variation). This way the solenoid only has to overcome the relatively small resistance of the spring to move the weight (new picture attached).

    The distances I gave in my first post are not realistic. The real dimentions are more like 20mm solenoid and 80mm stroke. No positional feedback nessicary

    I've investigated a few different options for this project, but all seem too expensive. Some of the other criteria are a compact design and as low a noise as possible.

    Rack and pinion would still have to have a large and expensive motor to shift that kind of weight, so is out for a few reasons.

    I looked at linear motors and linear servos but both are hideously expensive for that type of force range (in the hundreds usually).

    I found some hobby servos that were really inexpensive, but the biggest couldn't really shift the weight, there were doubts as to how many repitions they could do and they had the potential to be fairly noisy.

    I'm open to suggestions, but I was really hoping that this would be an inexpensive solution that was low cost, fairly compact, and almost noisless compared to other ideas.

    I had thought that using a tapered core may overcome the problem of 85% movement. If a thicker core/lower air gap increases feild strength, then wont the solenoid tend to want to pull in further?

    LOL, I guess I'm asking the same question again; maybe there just hasn't been anyone stubborn enough to try this before. If it just can't be done, then feel free to tell me - I guess I'm just hoping there might be an expert in solenoid design out there somewhere.

    Why is there no option for spell check in replies on this forum? I know I must have spelt stuff wrong, but what?

    Attached Files:

  8. Jun 3, 2012 #7
    It may be possible to use a 50 mm solenoid and a 50 mm or smaller core. There are some clocks with pendulums that use this approach.
  9. Jun 4, 2012 #8


    User Avatar

    Staff: Mentor

    If you are not incorporating feedback, how will you (or your microprocessor) know whether the weight is bouncing to the desired height? With time and temperature, friction and spring constants will change, yet your system will have no way of adjusting to changed conditions.

    Maybe you could incorporate rubber stops at the various desired heights, each controlled by its own small solenoid to switch it in and out of position. Then the main solenoid could provide a much more energetic pull than is ever necessary, [practically] guaranteeing the weight would be accelerated enough to reach any of the end stops.

    A strong field may allow you to rely on inertia to pull the core into the solenoid from one end and carry it through to almost exit at the other. Maybe travelling close to double the length of the core. Better than that and you'd be looking at treating the core as a projectile in a solenoid gun, to have it overshoot to x4 the length of the winding.

    I can see how a tapered core could theoretically be made to work, but I wouldn't hold much hope of it being anything more than a demonstration, and needing precision workshop skills and frictionless suspension, and probably unable to bear much load relative to the weight of the core itself. Positional feedback would be essential if you wanted a dependable length of travel.

    Using an electromagnet and a spring, you may be able to use mechanical resonance to gradually increase the bouncing of your weight. Would that suffice? — to build up the amplitude of the bouncing in stages, taking many cycles before it reached the ultimate x4 amplitude?
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook