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

Meaning of "fast on the shaft?"

  1. Dec 3, 2015 #1
    Hello all,
    I'm trying to put together this little puppy: upload_2015-12-3_12-31-43.png
    which is found here: http://507movements.com/mm_226.html
    Now, there are a few terms thrown around that I could use help clearing up. The descriptions use the terms "fast on the shaft, F" and "loose on the shaft, F" to describe the position of things. I need to be sure that "fast on the shaft" means that the mechanism is actually attached to the shaft e.g. like a wheel and axle, so that they must turn together, and that "loose on the shaft" means they will rotate separately.

    Then I'm also curious, in this diagram, about the position of frame A with respect to bevel gear D. You'll probably need to read the description to follow it easily, I think. Apparently A and D rotate together, and A rotates with F. So what sort of connection is there between D and A? I can't think of a linkage that would allow D to rotate axially, and keep it connected to A, without interfering with shaft F. Any clarification? Or better yet, a name for this device? Help is always appreciated!
    Thanks in advance.
    Sam Gallagher
     
  2. jcsd
  3. Dec 3, 2015 #2

    CWatters

    User Avatar
    Science Advisor
    Homework Helper

    Correct.

    This looks like a patent drawing so it only has to demonstrate the principle of operation. It doesn't have to be a 100% representation of what you would actually get if you purchased one from the inventor.

    The text implies shaft F goes right through from one end to the other but it doesn't have to. For example shaft F could just be welded to the right hand side of frame A. The bit of shaft F that supports gear E could be a separate piece of shaft. In other words a gap in shaft F could allows the shaft of D to go right through frame A (into/out of the page). That would allow the bearing/mounting for D to be on the far side of A. On the other hand Frame A could be a hollow cube with gear D mounted on the front face which has been omitted from the drawing to allow gear E to be seen.

    The key to understanding the operation is that shaft F and gear C rotate in opposite directions so their relative velocity is twice that of the input.
     
  4. Dec 3, 2015 #3

    JBA

    User Avatar

    I strongly believe the term "fast on" is the equivalent of "attached to".

    From what I can tell by looking at the document as a whole, this is simply a collection of possible motion transfer mechanisms but of which none may have ever been used in any actual application.

    For the specific mechanism you are trying to assemble, while in fact gear E may turn two times the speed of gear A, since gear E is not connected to any of the existing shafts, and is within the spinning frame assembly (and therefore cannot be engaged with any output gear) then this assembly is simply a curiosity and of no value as an actual power transfer assembly. One modification that might allow it to do so would be to add an additional tubular shaft attached to that gear that surrounds shaft F and extends through the left end of the rotating frame to deliver the gears rotation out through the left end of the assembly. I think the only thing that is unique about this unit is that it would do a 1:2 ratio using the same gear for every location.

    I am curious as to why you are assembling this unit. There are much easier ways to make a 90° transfer gear with a 1:2 ratio than going through all of the motions of this mechanism.
     
  5. Dec 4, 2015 #4

    Merlin3189

    User Avatar
    Gold Member

    I apologise for putting this in the thread, but I can't see how to add attachments to conversations.

    I've drawn my understanding of these gears.
    Gears.png
    I've simplified the ugly Frame A and shown how the D pinions are attached.
    I've doubled up the pinions D and B because I don't like the asymmetric arrangement of the original. This simply provides balance to reduce bending moments on the shaft and to reduce vibration in operation. But you can remove one D and B if you want.
    The shaft F is shorter than in the original: this was just an error which I couldn't be bothered to correct! It might be better to extend it to to seat in a loose bearing inside the output shaft, just to aid alignment of the two shafts? But with a cast base and the bearings for the shafts done in one through operation, I don't think it helps.

    I have shown thrust faces on B but not elsewhere. I couldn't be bothered. The bearing for B is oversimplified, so it doesn't add anything anyway.

    As someone else said, I can't see any use for this or any advantage over a simple 2:1 gear. But you never know. And it may be the principle is useful, even if the specific embodiment isn't?
     
  6. Dec 4, 2015 #5

    SteamKing

    User Avatar
    Staff Emeritus
    Science Advisor
    Homework Helper

    This mechanism reminds me of a couple of automotive differential gears joined together. Shaft B transmits a torque into the mechanism which is rotated thru a couple of 90-degree turns and comes out shaft D (or vice-verse; it isn't clear from the one diagram). Or perhaps shafts B and D are both outputs and this mechanism permits a speed differential between the two.
     
  7. Dec 4, 2015 #6

    JBA

    User Avatar

    Regardless of its usefulness, I give you full credit for your obvious improvement over the original design.
     
  8. Dec 6, 2015 #7

    rbelli1

    User Avatar
    Gold Member

    From the description it looks like an academic exercise. It is disproving the hypothesis that you can't double the speed with equal sized gears. It may very well be a terrible idea to do it that way but it is possible just the same.

    BoB
     
  9. Dec 6, 2015 #8
    Hey all,
    You guys are a huge help to me. Academic exercise is the right word for the whole thing, and I'm building it for a 3D CAD class I'm taking at university. I really appreciate all the help! This is a simple thing to play with that can stay on my desk. I'm crunched for time though, and I'm no mechanic, so who knows if the thing will actually work when I assemble it. (We're supposed to have all the parts 3D printed, so the best I can do for bearings is lubricate that stuff up. To my knowledge. It'll sure be interesting.)
    Thanks again.
    Sam Gallagher
     
  10. Dec 6, 2015 #9

    CWatters

    User Avatar
    Science Advisor
    Homework Helper

    So having all the gears the same size might be an advantage... you only have to put one into the cad system.

    Do post some photos when it's done.
     
  11. Dec 8, 2015 #10
    Well at long last the modelling is done. I'm new to mechanical assemblies and didn't realize just how involved and detailed you had to be when making things like caps and frames! Here is the model in Creo (my school's unfortunate program of choice):
    upload_2015-12-8_13-12-28.png
    The major outline of it comes directly from Merlin's sketch, so thank you so much! After that, it was all about measuring things and coming up with ways to connect all the pieces. Some used tapered "plugs" and others used a thread, depending on where I thought the most strain would come from. The handles are ugly squares with weird texture, just to make sure it wasn't too perfect. (ha). Thanks for all the help. It'll be sent away to the printer soon, I'll just be making sure the thing works together well in CAD first.
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook




Similar Discussions: Meaning of "fast on the shaft?"
Loading...