"...we choose to do these things, not because they are easy, but because they are hard... (JFK)"I haven't read all of this thread so forgive me if this has already been covered. Why not just use an anti-backlash gear?
Seems to cover it
I haven't read all of this thread so forgive me if this has already been covered. Why not just use an anti-backlash gear?
http://www.hpcgears.com/products/ant...lash_gears.htm
"...we choose to do these things, not because they are easy, but because they are hard... (JFK)"I haven't read all of this thread so forgive me if this has already been covered. Why not just use an anti-backlash gear?
Seems to cover it
Last edited by BillTodd; 04-21-2009 at 04:59 PM.
Bill
Great stuffInspired by this thread and the .pdf found by aystarik, I built a dual stage 90:1 cycloidal reducer.
Bill
Zincboy/Zoidberg-
Awesome work! I think I may have to try one of these at some point.
Zoidberg - what CAD software are you using? Would you be willing to post a model of the latest drive? One other question - would you need a bearing between the two outermost housings on this latest model? Depending on the size of this model, I have some IKO preloaded crossed roller bearings which could work very well to mate the two housings together.
Keep up the great work.
Keith
I'm using solidworks! But the i used rhinoceros 3D to make the profile (with rhinoscript).
I can export in most formats.
If you want i can draw something "custom" for you! Just tell me the specifications!
Bearings, number of teeth, diameter, axle diameter and so on
Nice design Zoidberg. I like the fact that the eccentric shaft is supported on both ends. In my design, the input shaft can have some unwanted eccentricity due to the lack of support. This causes the input to not be as smooth as I would like.
I am glad you were able to use the script. If I had Rhino I would have used it instead of acad When I load the .dxf files into acad, I convert them to a pline and enable curve fitting as well. I am sure Rhino works a bit better in that regard.
I measured the backlash in my reducer today. The output has about 0.004" (0.39deg) of play. This is largely expected as I machined the cam 0.001" small and the outer pins 0.001" oversize. I was expecting more binding and things not to work quite as well as they did. Next time I will try it with no tolerance
I also measured the input movement required to move the output and it came out as 37deg. I would expect around 35deg due to the 90:1 ratio and the output play so I count this a success. Note that I could not backdrive the assembled unit (even when I put the big wrench on there. I suspect that my efficiency is not as high as it should be.
Two things to watch for that I have found:
1. The input shaft needs to be well supported and any unwanted eccentricity will cause rough running.
2. The tolerance of the cam/pin interface is very tight if you want low backlash and having a "correct" cam profile is critical.
I think my next step will be to try to create a reducer with actual bearing rollers instead of the fixed pins in the current design. It will increase the size quite significantly, but I should be able to get the backlash right down
I use the Rhinoceros 4.0 Evaluation, its good for 25 saves/exports. If you want to try the scriptI am glad you were able to use the script. If I had Rhino I would have used it instead of acad When I load the .dxf files into acad, I convert them to a pline and enable curve fitting as well. I am sure Rhino works a bit better in that regard.
If you use rollers you should try to preload them against the cams like Onvio Dojen, Check the picture on page 3 and 4!I think my next step will be to try to create a reducer with actual bearing rollers instead of the fixed pins in the current design. It will increase the size quite significantly, but I should be able to get the backlash right down
very nice drawings (again)
I actually have a request, and backslash isnt actually a problem because it will turn one way only. A friend of mine is building a home telescope and he needs a very slow, constant movement to counteract earth movement and take long photo shots.
He found a 1RPM motor but he needs a lot less than that, something like 1/100.
Im doing it with gears, but this would be cool to try too
[edit] I don't quite understand how these cycloids work yet, but im googling as I type this
Started to draw on a dojen style preload!
Added balancing!
zoidberg, can you share some of the latests posts reducers solidworks models? I still don't understand how these work, I would like to try to animate them in solidworks, I succesfully animated spur gears, these should be doable too. I specially would like to understand the one posted by ZincBoy Thanks!
what version of solidworks do you use? i can make a simple assembly with mates...
Made an animation...
I have SW2007 but a friend of mine have a 2009 seat. Im starting to understand based on the video, I still don't get where the camshaft is used, can you animate a cut view?
Also, how do you keep the reduced speed side from falling apart?
Im still lost, got the cycloid and the pins, now I need to understand how the cam moves and how to determine the cam shape based on the cycloid shape.
I want to try this in my CNC mill
Loved ZincBoy video, I saw planetary gears in action (from a screwdriver) but never one like these
NEATman: I have started on a design that uses your bearings!
The housing is Nema 34. But i think i will draw up something that mounts directly to the motor! It would need some seals for the output!
Zoidberg-
Looks great!
I agree, seals would be needed. Perhaps some sort of automotive or truck drive shaft seal, or perhaps just an inexpensive seal from McMaster carr?
I think I understand what you mean about mounting to the motor - along the lines of removing the first bearing in the housing and just rely on the bearing in the motor? The eccentric could then be added to the motor shaft, and a coupling would not be needed.
Keith
BTW, I'll send you the bearing model tonight when I get home.
Zoidberg-
How about using roller bearings instead of ball bearings on the Dojen style design. It would save a considerable bit of space in the design, while having a higher load capacity. They also tend to be less expensive too.
Keith
NEATman
Basic research is what I'm doing when I don't know what I'm doing. Wernher von Braun