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

Help with gear ratios please

  1. May 19, 2010 #1
    Hi All,

    I have a project which involves pulling a sled (wheeled, on rails - low friction) weighing approximately 15kgs over a set distance and time. The distance is approx 60cms and the duration is 3 days.

    To minimise cost I intend to anchor a motor(ac or dc)/gearbox combination and use a take up spool to "reel in" the sled.

    I've calculated that a 2cm diameter spool will need 9.5 revolutions, a 3cm spool 6.5 revs and a 4 cm spool about 5 revs - over 3 days!

    Where I need help is what motor and gearbox/es will I need to achieve this?

    The time and distance doesn't need to be exact, I reckon distance can be +/- 5cm and time can be +8 hrs.

    To save cost I'd like to be able to leave out a controller, ie run the motor straight from battery or mains - preferably battery.

    I'm starting from scratch so anyone with suggestions for off the shelf components please let me know.

    Otherwise I guess I need to know what speed motor at specific voltage and what gearbox/es to attach to it.

    Sorry for my ignorance!

    Any help/advice much appreciated.


    Slim
     
  2. jcsd
  3. May 19, 2010 #2
    Very simple gearing question. With such a slow speed and no load you can use pretty much any electric motor you want. (low power low torque)

    You'll have to search for electric motor data as I don't know any off the top of my head.

    The only limiting factor is the size of the gear/sprocket you can use. I'd suggest you look at using a chain and sprocket for something like this. This would allow you to simply purchase ready made gears and use multi stage gear reduction.

    EDIT: The reduction ratio is likely to be huge, so sprockets may not be the best idea. A worm wheel gear may be a better bet. Hmm whats the rolling radius of the wheel?
     
    Last edited: May 19, 2010
  4. May 19, 2010 #3
    Chris,

    Many thanks for your input. I had considered the worm gear option but it would have left me with similar issues and added engineering complications and expense.

    I envisage using non stretching fishing line on the take up spool, in fact that's pretty much exactly what I'm after - a motorised fishing reel anchored at 90 degrees to the direction of the sled.

    What I really need to know is what components to put together to make it happen.

    Many thanks,


    Slim
     
  5. May 19, 2010 #4
    You are never going to get a motor to rotate slowly enough (I think, im not that up on motors). It will need to be geared down to reel it in at the right speed.
     
  6. May 19, 2010 #5
    Chris,

    Totally agree. It's the gearbox side of it that I'm struggling with.

    I imagine I will need a motor with gearbox going through another gearbox. That's what's making my head hurt!


    Slim
     
  7. May 19, 2010 #6
    Thats why you need a worm gear solution. You need something to move incredibly slowly which means a very large reduction gearing.

    Ok assuming your fishing reel set up has a diameter so that 1 revolution means 1cm linear movement. You need the reel to be rotating at 0.0139 rpm.

    The lowest rpm ac motor I can find is 10rpm (I didnt look that thoroughly). So from the motor to the reel mechanism you need a reduction ratio of 719:1. This is an incredibly large ratio for anything other than a worm and wheel.

    You may even need 2 worm wheel gears in sequence.
     
  8. May 19, 2010 #7
    Hi Chris,

    Thanks again.

    I've tried to work out ratios for normal gears and the figures, as you suggest, are crazy. I guess you're right about the worm gear solution but the problem still stands. What the heck do I buy to achieve the desired result? I can deal with the engineering to make it work but I need to know what to get.

    Never thought this would be so difficult!


    Slim
     
  9. May 19, 2010 #8
    Hi Chris,

    Thanks again.

    I've tried to work out ratios for normal gears and the figures, as you suggest, are crazy. I guess you're right about the worm gear solution but the problem still stands. What the heck do I buy to achieve the desired result? I can deal with the engineering to make it work but I need to know what to get.

    Never thought this would be so difficult!


    Slim
     
  10. May 19, 2010 #9
    In the long ago this task would be solved with 'clockwork', a wind-up spring and a balance wheel oscillating to meter its unwinding.

    Basically you still want a clock. (The old kind with hands.) (In fact, the old kind that are not run by battery, because the very cheap battery style tend to have very 'slippy' hands and possibly not much torque. Look for a plug-in clock.) You want to attach a small diameter disk or drum to the hour shaft.

    One of the problems to solve is whether your rails are perfectly level, because if not the cart might roll ahead a little. You could put a slight tilt, and then either allow the clock to pull it up, or better yet, lower it down.

    Well, enough about crazy ideas. Here is something you buy for US $15 with 1300:1 reduction and a motor attached. If you get two, one can drive the other whose output can be a drum.

    http://www.hobbyplace.com/robotics/gearbox.php

    Jim papadopoulosj@uwstout.edu
     
  11. May 20, 2010 #10
    Nothing crazy about them, they are the best ideas in this thread yet. The only problem I can forsee is that the clock mechanism doesn't have the strength to keep the rollers moving.

    Those robotics motor/gearbox sound pretty good, you'd need to couple 2 of the gearing systems together though as most the supplied motors appear to be 10500rpm @ working load.

    15gcm is plenty of torque after gearing it down (I just hope the plastic bits dont shatter to bits when you first turn the motor on). It may be worth finding out if you can use a variable input to alter the rpm for a short time.
     
  12. May 20, 2010 #11
    Great stuff guys,

    Now we're getting somewhere! The more ideas the better but at least I've got a potential supplier for the components. Controlling those revs could be an issue, as pointed out.

    Any more thoughts very welcome but thanks to all so far.


    Paul
     
  13. May 20, 2010 #12
    I'm not facile yet with this interface, so please pardon the lack of quoting.

    Let me tell exactly how a clock would do the job. Imagine a CUCKOO clock with weights that fall over a couple of days. Let's say a weight is 1 kg (it's probably less). You can replace the clock weight with the cart on tilted rails or cart pulled by a rubber band, so that it is motivated by a force of 10 N or whatever. The clock merely holds it back, and takes motive power from it. So your cuckoo could announce the travel of every few cm!!!!

    [Or you could have a 2kg falling weight pulling an equivalent of 1 kg uphill.]

    I am not explaining this in detail, hopefully the concept comes through.

    Jim
    bikengr@netnet.net
     
  14. May 20, 2010 #13
    I'm loving the cuckoo clock concept. I'm tempted to try it just to see the look on people's faces!


    Slim
     
  15. May 20, 2010 #14
    This is a job for a servo motor, aka stepper motor. They aren't cheap, but you only need one for this one-off project, right? If you can possibly get your hands on a surplus or scavenged one, along with its driver, you can just send a pulse from the driver every x seconds, it rotates a fraction of a degree and you get your furlong per fortnight snail's pace.

    Actually, looking around a bit, they are pretty cheap these days. This is not my area of expertise, but it seems worth looking into.
     
  16. May 22, 2010 #15
    I'm embarrassed I didn't think of stepper motor. Shows that my mind is still pre-1980, at that time you had to be a real expert to wire up and program a controller. Now it's probably trivial.

    The term SERVO used to be reserved for a high torque, low inertia DC motor with shaft position or speed feedback. That is certainly another option -- if you have electronics capable of generating a slowly rising voltage signal, the motor would just position itself at a position corresponding to the voltage.

    Mechanical Guy -- Jim Papadopoulos

    PS I guess as I'm opening my mind to other opportunities, there are surely chemical means to do the job, for example have something move as it dissolves or melts. Or a kind of water clock, drips lowering a large floating weight.
     
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook