# Improving ball feeder design

I designed this ball feeder for my robot however I'm experiencing frequent ball jams as the ball is stuck against the wall of the hole entry or just a ball being moved left and right not entering the.

I was thinking of maybe adding a wall to one side of the opening to guide the ball into the opening. However since the balls are on top of the moving disc they could fall on th other side and get stuck there.

I should mention that this is implemented using a single servo motor so only 0-180* movement is possible.

I'm open to any suggestion either ne design or improvement.

• can you have it wiggle or vibrate a little? maybe tweak shape and especially the edges of the entry pocket to make the contact angles on the way in shallower? Commented Mar 6, 2021 at 0:35
• A few questions for you. The feeder is to pus the ball in the hole, or the other way round (to receive the ball from the hole. Also were are the jams occuring? Also the side view is a bit confusing.
– NMech
Commented Mar 6, 2021 at 6:12
• I have thought of a couple of solutions, but the best one is subject to the plate being flat (parallel to the bottom), is that possible? Are you using the slope to guide the ball as it falls from the top? Also, is there a cover for the circular portion?
– NMech
Commented Mar 6, 2021 at 9:28
• @PeteW I can generate a jerking motion but it does not help that much. Commented Mar 6, 2021 at 10:22
• @NMech the feeder should push the balls into that opening. Basically balls are dropped on top of the swivelling disk. It should be possible yes to make the bed flat and parallel to the bottom plate. I'm using the slope yes to guide the balls downwards towards the hole. Imagine a hollow bucket sitting on top of it and balls being dropped in. Commented Mar 6, 2021 at 10:25

The simplest idea for me would be to add on the bottom (and the top cover) small inserts of the following shape (dimensions will vary depending on the size of the ball, etc)

They would guide the ball towards the opening with the right velocity.

You only need to provide adequate clearance for the rotating part, which from your side view should not be a problem.

The inserts do not need to be large, but you need to take into account curvature so that the ball is driven nicely in.

Other things that you can do is:

• provide a small draft angle in the opening.
• Add a similar insert from the other side of the opening (the shape is not the same, it should help guide the ball towards the middle, so that it gathers the necessary momentum to be driven out.
• Since you are printing you can also add the following draft on the inserts (apologies for the poor drawing - I was trying to sent it before you print)

• also, you could do with redesigning the rotating part. in your case is symmetric, which I don't think is optimal for this scenario.
• This looks promising I will try printing it today. Commented Mar 6, 2021 at 16:11
• There are many improvements/finetuning. The main issues will be from the top cover to make sure that the ball will end up in n the tray. But that is subject to the constraints above.
– NMech
Commented Mar 6, 2021 at 16:25
• Since you are printing see the final image for the profiles of the inserts
– NMech
Commented Mar 6, 2021 at 16:44

I think you need change the structure, as show image of ball feeder.