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

A question about DC Motors, Clutches, and Brakes.

  1. Feb 15, 2012 #1
    I am making a functional model friction crane. I am in the design and material gathering phase right now, but I have a question about DC Motors and how they would behave in a certain situation.

    I plan on having a DC motor spinning a shaft with a clutch on it. When the clutch is engaged, their would be a considerable considerable load added to the motor, in the direction it is already spinning. How would the motor react? Would it maintain its RPM, or would it speed up?

    This is how I plan on booming the crane down, and hoisting the cable down. I will have brakes on the crane as well, but I don't know how they will be implemented as of yet. I will post more details as I have them, but would love an answer to this question ASAP!
  2. jcsd
  3. Feb 15, 2012 #2


    User Avatar
    Science Advisor
    Gold Member

    Hmmm... doesn't sound like a good idea. If engaging the clutch will add a force turning the motor in its powered direction you'll have two potential problems. Firstly a DC motor will typically spin up to some maximum speed.

    (Note: typically the produced torque at a given supply voltage will be about linear from max=stall torque down to 0 torque at max speed.)

    If you add a "negative load" i.e. push the motor to spin faster its speed may exceed the mechanical tolerances, burning out bearings or physically breaking components.

    If your motor can handle that then the 2nd problem is that it will act as a generator producing its own voltage. If you're spinning the motor in reverse you'll recharge your supply if it can take a reverse current (say a battery) but if you're overspinning it you're effectively pushing more current through the supply in a short circuit situation. It will draw a lot of current and you may burn out your supply or get serious ohmic heating of the motor.

    Here's a link I found on DC motor basics: http://lancet.mit.edu/motors/motors3.html
  4. Feb 15, 2012 #3
    Hmm, that is what I figured. I suppose I will have to make it a live boom and hoist, and only have the motor power hoist up and boom up. I will have to use the boom/hoist brake to regulate the speed of descent when going down on either of those functions. Thank you for your input.

    Is their a way to have a controlled descent in a situation like I have described?
  5. Feb 15, 2012 #4


    User Avatar
    Science Advisor
    Gold Member

    You can use electronics to use the motor as a brake. The 2nd problem I gave is then turned to an advantage. If you disconnect forward power and then connect the motor to an electrical load (say a high power resistor with good heat dissipation) then the current will induce a back torque. A resistor is basically the electrical version of a brake. You should view the motor as a coupler between current flow and mechanical motion. (an inefficient one, i.e. with "slippage" )

    Likewise a capacitor is the electrical analogue of a spring and an inductor as the analogue of a flywheel.

    [EDIT: BTW I'm good on theory but haven't practical experience. I'm sure this problem has come up many times, Electrical Engineers probably have classes and even minor concentrations on the subject. A quick google of "overspeed protection dc motor" turned up may hits which may be helpful.
  6. Feb 15, 2012 #5
    Instead of motors, clutches and brakes, just use hobby servos.
  7. Feb 16, 2012 #6
    I fly Electric Helicopters, so I am well aware of servos and their capabilities. :)

    I don't want to use servos on this project. That would turn this Friction crane model into something else entirely. I am already cheating a little by using electromechanical clutches instead of mechanical friction clutches, but the concept is still upheld as they are still clutches.

    The idea is to display as much of the innards as possible. I want the mechanics to really shine when this thing is running. I am a heavy equipment operator by trade and have fantasized about making little functional models for a long time.

    I ordered some clutches today from a surplus store. They were so damn cheap I couldn't resist! This is where I got them.


    I ordered two of each to start off. Now, a question on Electromechanical clutches... Would the grip of the clutch be directly proportional to the amount of voltage supplied? If it is, I could try and use these clutches as a brake/power down for boom down and hoist down.

    Also, is their a simple way to regulate voltage to a motor/clutch in the form of a switch/lever? I would rather a lever instead as it would be a smooth transfer from one voltage to the next.

    I appreciate the help guys, I was a little lost before I found this forum, lots of ideas, very little know how!
  8. Feb 16, 2012 #7


    User Avatar
    Science Advisor
    Gold Member

    Not proportional. I don't think these clutches are designed to transmit partial torques. They will tend to "grab and go". But there's got to be some transition curve so experiment with them to see. Typically these are just electromagnets holding metal plates together so the normal force will be proportional to the square of the current (which in turn will be more or less proportional to voltage). The slip torque will then probably be the same.

    Also there will be a bit of motion before the plates come into contact. So my intuition tells me that below a certain voltage you'll see little force. At a threshold voltage the plate is pulled into contact and you'll have a sudden jump in transmitted torque which will then increase along a parabolic curve as voltage further increases.
    Hmmm... obviously you could use a lever to switch the clutch on and off. To continuously vary the voltage you can have the lever turn a variable resistor which in turn controls the voltage (or more properly in this setting the current). That will probably need at least some form of transistor circuit, you can look up variable output voltage regulator circuits.

    One problem with what I think you're trying to do here. You won't have the same tactile feedback as you get with a lever directly connected to a friction clutch. There won't be the same "feel" of physically pushing against the friction surface to control the amount of transmitted torque. I think you'll at best get a simple "on-off" behavior without some very sophisticated feedback or contol mechanism...

    Something like a force gauge on the lever and a digital micro-controller pulsing the clutch or dynamically controlling its voltage with another sensor detecting the slippage. Something basically like anti-lock brakes.
  9. Feb 17, 2012 #8
    Okay, how about stepper motors? You could have much better control of speed and direction changes, yet still use the clutches.
  10. Feb 18, 2012 #9


    User Avatar
    Science Advisor
    Gold Member

    I think that is still too far removed from the intent of his model. A friction crane uses a friction clutch as a PTO from a rotating shaft.


    I think if you use a flywheel then the issues I raised with forward power on a motor won't be too much of an issue. No more than the actual friction cranes.

    Consider a DC motor driven flywheel with a governor calibrated to apply friction above the motors zero load speed. Under load the motor works to lift. Under no-load the motor just turns the shaft. Under negative load the governor slows the shaft.

    I think that's the most elegant solution and closest to real friction crane.

    For a governor you can use a centrifugal clutch (adjustable) attached to a fixed shaft. Above the critical speed it begins to engage and brakes the system.

    [edit] Actually forget the flywheel! The clutch-governor system should do the trick!
  11. Feb 18, 2012 #10
    What's the size and desired capacity of this crane?
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook