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

Looking for mechanical clutch solution

Tags:
  1. Jun 4, 2015 #1
    Hi,

    I made an electronics project but I'm having some mechanical problems. Particularly, I'm having issues finding a good clutch for my problem.

    What I'm trying to do: physics1.jpg
    A motor is connected to shaft 1, a human can manually rotate shaft 2.
    I want that when the motor isn't doing anything, a human can easily rotate shaft 2. In case it matters, the human will rotate the shaft at most 3 rotations in either direction. When the motor rotates, both shaft 1 and shaft 2 need to move.

    The solution I came up with: physics2.jpg
    1) gear 1 and gear2 are not engaged, hence gear 2 can rotate freely when the motor is not moving
    2) The motor is turning, the gears interlock and so gear2 turns
    3) continues until back in position 1

    There are a few problems such as you need to be sure that when motor is at rest, there are no gears interlocking. However I might be able to find a solution for this. My biggest problem however is that if the motor's batteries die when it is engaged, there is no way that the user will be able to rotate shaft 2. Any solutions are always welcome =) Thanks!
     
  2. jcsd
  3. Jun 4, 2015 #2

    billy_joule

    User Avatar
    Science Advisor

    The gears on your proposed solution will not last long. And the output shaft will not have constant velocity.
    What is the power transferred? RPM? Does it operate in both directions? How often will it need to be rotated manually?
    Have you looked at centrifugal clutches? The type found in go carts & mini bikes can be had for ~$20 new. Remote control car scale clutches can also be found cheaply.
     
  4. Jun 5, 2015 #3
    Thanks for the reply! It will rotate at a maximum of 100 RPM (freerun) probably and a maximum of 2Nm (stall) torque. Yes it operates in both directions. I have briefly looked at centrifugal clutches but haven't found any that are small enough (must fit in 3cm^3). Also, when motor turns at lets say 1.5Nm, it might only do 10RPM or so. Not sure if this is fast enough for centrifugal clutch.
     
  5. Jun 16, 2015 #4
    Considering that you would like to have a Human easily operate shaft two I complete isolation between shafts may be in order. Humans are very low power producers.
    I would be interested to hear further a few items. 1) How much power can you afford to lose to parasitic losses such as in using a fluid coupling? 2) Will the shafts be coming to rest for engagement and disengagement? 3) Does the engagement need to be smooth are can you afford some clunkiness? 4) What modes of connection do you need to achieve? ie. either shaft only, both shafts together, either shaft locks out the other, etc.
     
  6. Aug 31, 2015 #5
    Sorry for late reply, I must of missed it somehow.

    1) as little as possible since it will be battery powered and the motor will only have like 5W
    2) yes, there will never be an engagement when one of the shafts is rotating
    3) preferably smooth, but not necessary
    4) either shaft2 should rotate or shaft1 and shaft2 should rotate together.

    Also I was thinking about maybe finding a dc motor where I would need a small reduction so that a clutch would not be necessary. Someone told me that a 1:100 reduction should be small enough to be able to turn shaft + motor by hand without much problem. This would eliminate the need for a clutch. Is this a viable solution? Thanks!
     
  7. Aug 31, 2015 #6

    Nidum

    User Avatar
    Science Advisor
    Gold Member

    The easiest of all clutches to make is the axial dog clutch .

    Almost as easy is the cup and cone clutch .

    Another easy one is to use worm drive and engage/disengage the worm .

    Tell us more about the whole project - there may be completely alternative arrangements possible .
     
    Last edited: Aug 31, 2015
  8. Aug 31, 2015 #7

    Baluncore

    User Avatar
    Science Advisor

    Why disengage the motor?

    You could put a round knob on the motor shaft and turn that be hand when needed.
     
  9. Sep 1, 2015 #8
    I'm making a DIY version of the Kwikset Smartcode. So basically I need a motor to lock/unlock my door while still working flawlessly with the key. However I want a different clutch system. They use a dog clutch with 180 degrees free turning to allow the key to rotate. However this means that the motor will mostly rotate 180 degrees before engaging. I'd like something that doesn't need this 180 degrees of play. (If this isn't clear I can upload an image).

    From my research on line, it seems that I need a dogclutch with centrifugal engagement. The dog clutch because it would need high torque and the centrifugal engagement because I can't physically engage/disengage the clutch myself. However, this seems like it would have an abrupt engagement/disengagement. The engage/disengage worm also sounds interesting. Would this mean that I need to physically move the shaft of the worm with a different motor to engage/disengage it? Do you have any link to info on how this would work? Thanks!
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook




Similar Discussions: Looking for mechanical clutch solution
  1. Clutch Question (Replies: 2)

  2. Gears and Clutch (Replies: 0)

Loading...