Hi everyone!
Some of you may remember my previous post in which I created a rocker bogie drivetrain using FRC parts, ( https://www.chiefdelphi.com/media/photos/45330 ) but eventually came to the conclusion that it would be “…absurdly expensive and time consuming…” to make. I’ve decided to make it, and found that the most expensive element of the design was in fact the 8 versaplanetary gearboxes. Upon further examination, I discovered that the most expensive part of the gearbox is the base kit, coming in at a whopping $50. The gears themselves only cost $15, and an output shaft is $5, so I figured that by 3d printing the input, ring gear, and output stages, I could save myself over $50 per gearbox. (Yayy, money!)
While creating a ring gear profile to mesh with VP gears, however, I found that the teeth were tiny (the flat part on top of the tooth was about 0.4mm across). Since I don’t own a 3D printer and my team’s lab is closed for the summer, I haven’t really had a way to just print a ring gear and find out if it would work. I could get a ring gear printed at some 3D print hub, but in order to get a material and printer that are precise enough it would cost around $60 per print plus shipping.
In short, probably not. Even if your printer has enough resolution to print a coherent tooth pattern, it likely wouldn’t hold up to any substantial load.
Please just don’t. Yes there are some high strength materials out there that will work, but the reality of the situation is that if using plastic were feasible/reliable, vex would have just used plastic. The teeth on these gears are tiny, as are the rest of the parts, and 3d printing technology up to almost the highest end just isnt great for this stuff. Maybe you could try using PETG, or Shapeways can 3d print some crazy stuff nowadays, or even some of that CF inlayed 3d printing, but my bet is that even aftet the investment in those materials, the system printed as is will fail, and your investment would be close to if not surpass the cost of some versaplanetaries.
Tl;dr- don’t 3d print this stuff. It may work once or twice but in the long run these parts will fail if they’re 3d printed without a serious design change to make them work as such.
This will probably not work with the loads that a drivetrain will see. The tooth profile is small enough that a small amount of flex in the printed ring gear will let the teeth skip.
That said, if you have a printer, and you already have the other parts of a gearbox sitting around, why not print a ring gear and give it a little test? It’s not a very expensive part to make, and it’s always fun to test things and see if they work / why they won’t.
I wonder how something like this would perform long term using a stronger filament (PETG, Nylon, Polycarbonate, etc.) with the appropriate lubricant. Assuming you used it with appropriate loads and checked the wear over time, I could see these being useful for certain applications. I feel like the hardest part would be adapting it for FRC motor shafts which tend to be quite small (perhaps fit it for standard pinion gears on the motor shaft rather than couple it directly to the shaft itself).
I disagree with what everyone here is saying. I have 3D Printed Versaplanetarys and they work perfectly. The biggest problem is you dont have one easily accessable to test scaling for the gears, because the biggest problem when 3D Printing mechanical parts is the plastic not being exactly the right size. I used 97.5% Scale with a .2mm nozzle, and I was able to get a versa that worked just fine, and for the robot you want to make, if theres not huge loads and you are driving just a drivetrain, imho you will be fine.
Edit: I used normal old PLA. Also there a tons of planetary gearboxes on Thingiverse.com
You 3D Printed the actual VEX Versaplanetarys? Don’t you have issues with the the gear teeth being that small with plastic? Or perhaps you mean planetary gearboxes more generally?
Our team has had some good success with 3D printed gearboxes. The points of failures are often where it transitions from one material to another. (3D printed gear to metal gear.) The gearboxes that have failed on us were also used in applications with a lot of shock load and relatively high output speed. I think a planetary with a slower output speed like in the case of a drive train would do pretty well. Especially if the drive train is traction limited. (i.e. wheels spin instead of stalling motors.)
(You might want to beef them up a bit and I would still not suggest them being used in an actual competition setting. But it definitely sounds like an excellent affordable way to prototype extravagant designs.)
That’s odd, we’ve seen the opposite. As you add more reduction to a gearbox, the torque increases, and the plastic teeth are more likely to shear off under load. We did this a few times to the second stage gears on our turrets (which had all 3d printed gears). The gears tend to survive better in the first stages of the gearbox, and then kerplode quickly in later stages.
We did okay with our feeder gearbox, which had 32DP gears to run the ball feeding belts for our shooters. Even so, we did have one occasion when a feeder shaft binded and we ended up sheering teeth off with the motor pinion (AM 9015 motor, 15t steel pinion). It works fine under normal conditions, but as soon as the motor stalls, plastic starts breaking.
Hi all! Sorry for taking so long to reply, I’ve been super busy. The reason I was asking about printing to match the VP contour was mostly a time thing; putting out 20 sets of gears and 8 sets of casing would be nothing short of tedious.
I really liked Brian M’s scaling solution, and the optimist in me is saying that it will suffice. I say this because my plan is to have nothing more than a drivetrain by the end of this project (this is for school) as my goal is to show that there are affordable, simple solutions to complex problems.
I agree with that you said here. I’ve been considering creating my own tooth profile for this (with bigger teeth) so maybe they wouldn’t shear as easily. I was mainly choosing to print the VP profile for prototyping to prove that at least the concept was sound (and to save time) but I’m beginning to think that I’ll need to make some design changes.
I use the original (Sometimes called the MKI) Prusa i3, Its the older version of the famous Prusa i3 MKII and is just amazing. I use the free slicer Cura to create the Gcode, and in Cura you can really easily just set the scaling.
There is no doubt in my mind that 3d printing can work to solve your problem, but I think youre asking the wrong questions. Could you 3d print functional stuff for a versaplanetary? I think not, others think so, but why are you limiting yourself to use of stuff in that setup. If you want to design something to work with 3d printing, design it for 3d printing. Don’t force yourself to use a modified version of a product, as a designer its your responsibility to understand the materials you’re working with and resources you have, and create the best product for the application within your constraints. Generally speaking, designers already have lots of constraints from cost to manufacturing, blah blah, but using a VP shape setup is not a real constraint and limiting yourself to that form will come back to bite you down the road.
Once you truly understand the forces at play here and exactly what you want to achieve with this project, I’d bet that with some creativity you can design something that will work better than 3d printing VP stuff (even scaled or whatever), cost less, and still handle itself well under some load because you’ve designed it with your 3d printing capabilities in mind.
Designing stuff is hard enough as is, please don’t limit yourself to what’s already on the market, and please please please don’t take that stuff and just remake it in worse material with less accuracy.
By designing your parts to actually be 3d printed you will ultimately get a much more robust and overall better design, and go further to “show that there are affordable, simple solutions to complex problems”
The video that tr6scott posted is very relevant. I saw that a few weeks ago and was amazed at what a relatively small 3D printed gearbox could do. Using a larger planetary like that could serve you very well.
If you want to save a bit on the gears themselves, my harmonic drive could work: https://www.chiefdelphi.com/media/photos/45278?langid=2
But it requires a metal output to really shine under the torque loads, and a fairly good 3D printer like an ultimaker. Harmonic drives are great because they spread the load among several teeth, so they’re a lot stronger when printed.
I’d be interested to see how those tests go. When and how the plastic fails.
Stalling motors is definitely bad for printed gears. I will agree that plastics can only take so much torque before shearing. In OP’s case I think they could get away with printed parts.
