I have created models of the Changeup Balls that are optimized for 3d printing.
The main use for these models is for prototyping while teams are waiting for the official game objects.
“Ball_Simplified” is optimized for rapid 3d printing and can be printed without support. By tweaking the settings it can be printed under 6 hours (depending on the printer).
“Ball_Accurate” can be used to assemble a ball of same weight as the actual competition ball, but will require supports and a much longer printing time (at least double) . Nylon screws and nuts must be used in order to get the weight to be accurate.
Both these models are 1/2 of the complete ball. Two of these halves need to be printed and then require five 1.5" screws and 5 nuts to hold the two halves together.
The files can be found on GrabCAD at the following URL:
I printed two types of models with no supports - both suffer that top indent is trying to print in middle of air - not ideal, but prints something
One requires gluing together and the other requires screwing together…
I glued both models together.
I think it is adequate for prototyping - mass is around the right amount for game object. Wall thickness? well pretty solid for prototyping while waiting for real objects to arrive.
Each half print requires 4-5 hours to print - so 8-9 hours to have finished product… Worth it? nope not for more than one test object - better things to print time wise.
Printer used Bare3D.us printer - rock solid design and keeps printing - designed by 44/QCC2 Leland Crowther - great engineering.
It looks you printed the “ball_accurate” file, this one requires support for the exact same reason you discovered. I was surprised though that it could print the rest of the ball fine without support, I was expecting several other spots to need support.
The second file “ball_simplified” should be able to be printed fine without supports, but the weight wont match the actual ball
I stopped the print because there was no advantage from printing second file - took twice as long to print and twice the material for no functional gain.
Here is photo showing how the file did not solve the problem of no support material:
despite blurry photo, you can see how layers are not fused correctly in the inner support layer. So no gain.
My advice if you need object, go quick with fast print, no support and decide if you need to super glue two halves together or fasten with screws. End of the day, you will have a game object to implement your mechanisms… Better to order real objects. Right mass, right rigidity …
The files were not named properly, I reuploaded them with the proper names. Thanks for catching that!
The simplified version should 100% be able to be printed without support. Most 3d printers should easily be able to handle slopes of 45 degrees without any issue.
The simplified version with no support when compared with the accurate version with support prints faster, I confirmed this by comparing the prints after slicing in cura.
Here’s a screenshot of the two:
Ball_Simplified
Ball_Accurate
This is using the default settings for “fast print” with the only difference being the “generate support” for Ball_accurate being ticked. With optimization these times could be reduced.
with 10% infill the properties for Ball_Simpified with no supports becomes this:
which has the same amount of material as the Ball_Accurate with supports.
So far I have done two (60% scale) change up ball test prints. I swapped to a 0.8mm nozzle on my ender 3 and am printing with petg(and later on pla). I got the settings tuned in and they are quite strong. For the full size balls it takes 5 1/2 hours per cube so in total 12 hours. But they are very strong and are within 2 grams of the actual ball. Pictures tmr. Eventually going to design a goal ring(or whatever the name is.)
