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

Servo vs ? for lift/winch/hoist

  1. Mar 20, 2015 #1

    As many others here, I am starting a project that involves some engineering and will at some point, if I go ahead with it, require an engineer to design and certify. But for now I'd like to understand the principals involved and dust off some of my 1st year physics and see how much of it I can solve with some help from experts.

    The details so far:

    -hoist controlled by a microprocessor that can vertically position the payload with 1/2"-1" accuracy
    -payload of 24 kg max
    -wire rope hoist line, 1/4" diameter (my calculations indicate this gives a safety factor of 30 or so and electric hoist applications require SF of only 7)
    -hoist line length is 100 m and mass of that length of 1/4" wire rope is 16 kg
    -desired lifting/lowering speed is 1 m/s
    -desired acceleration is .5 m/s/s, when lifting that works out to net acceleration of 10.3 m/s/s
    -winch drum diameter is .254 m when drum is loaded with wire rope

    So the force required to hit that acceleration during lifting:
    F = ma, F = 40*10.3 = 412 N

    The torque I'll need from the motor to achieve that force when the drum is fully loaded:
    T = Fd, T = 412*.254*.5 = 52.3 Nm

    I also have these formulas for selecting a motor:
    Ts = stall torque
    Wn = no load speed

    so T = Ts – (Ts/Wn)*W
    W = (Ts – T) (Wn/Ts)

    That's as far as I am now and I'm now running into the issue of which type of motor is best for this application. I spoke with a supplier of motion control equipment who advised using a servo. That's all fine with me since they are made to keep track of position. However, because the velocity of the payload will vary as the amount of wire rope left on the drum changes I am considering having a sheave with an encoder, remote from the motor and drum that the wire rope would run over. This way the actual velocity of the wire rope could be read. What I'm not sure of is whether a servo can use info from an encoder that isn't attached directly to the motor. Can anyone provide some insight here? And since I wouldn't be using the servo's inherent position tracking ability, would I be better off using a regular motor with VFD?
  2. jcsd
  3. Mar 21, 2015 #2
    Would this question be better off in a different forum section? Electrical eng maybe? I wasn't sure so just posted in Mech Eng.
  4. Mar 23, 2015 #3
    You can get a free refresher course of motion control dynamics by websearching & finding the "Smart Motion Cheat Sheet" in PDF.

    Your positioning accuracy is not too demanding. You need to determine 'peak torque' which incudes not only torque to raise the load against gravity (and friction, and stiction, and sludge in the rails, and moon phases, and...) but the added torque needed to accelerate the load from zero velocity. So you've got some significant mass moment of inertia calculations in front of you. Your unit is sized for peak torque plus any safety factors.

    A servo could work, but is overkill IMHO. I don't know what your budget is, but my first look at this would be with a small gearmotor with a small AC induction motor and variable frequency drive. Many of those vendors (SEW Eurodrive pops into my head first) provide nifty online sizing calculators. If the sizing calculator does not have this "elevator" model, then the manufacturers usually provide good Engineering Guides to help you run the calculations and size the unit. You'll need to gain a bit of understanding about the different types of gearboxes in order to select the best fit for the load application. If I recall, certain of the SEW VFDs could accommodate the rope wrap-up diameter change of the winch spool because their units are used for elevators very often.

    Properly applying a gearmotor requires some analysis of the frequency of use, too. Many frequent start-stop cycles will require a larger unit to dissipate the heat generated.

    I was always told: "Torque is cheap...use plenty of it." Get lots of reserve torque capacity. Your system will only draw enough current to provide what torque is demanded of it.

    And put brakes on that thing.
  5. Mar 25, 2015 #4
    Thanks tygerdawg. That was a really helpful reply.

    Would using a wormgear in the drive mean I can get away without using a brake? Or is redundancy in braking a must?

    I'll get in touch with our local SEW Euro guys and see what they say.
  6. Mar 26, 2015 #5
    Worm drives are OK, but have less mechanical efficiency that other gear trains. If I remember, they're good for shock-load applications. Not having a brake requires that the gear ratio of the worm gear arrangement has sufficient backdrive resistance to keep it from moving under load. There's a formula out there somewhere for this, but I don't remember much about it. Use a brake to be sure. Especially if this is a safety-critical application.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook