torso

I wanted it to be colourful. The blue/dark blue combination was the first idea (with black or possibly DBG drum), but it didn't feel colourful enough. I like dark pink and have wanted to build something in that colour, but there weren't enough curved parts available. Looking at the available colours for those parts, magenta stood out as the closest. It reverses automatically after about 40 seconds. My previous module (cloud) is even more solid (but smaller). They go well together.

I'm glad you like it! I also can't help but associate the looks with some kind of kitchen machine.

As can be seen in the movie, this screw can't be reversed - balls are always transported forward regardless of rotational direction. A very simple mechanism. The time consuming part has been making it sturdy. Had to rush the design at the end to finish before an exhibition, so some things will be adjusted later. I intend to post a picture of how it's constructed when the 3D model is finished. The switch is purely mechanical using regular transmission parts. Driven by a medium motor.

I've always pushed the axles apart. It's not particularly rigid. Maybe it would help to fix the half-stud offset and make it centered over the tipping platform. Another error I know about (seen once) is that sometimes a ball will not start rolling when at the top. As soon as the arm starts moving again, the ball drops straight down. That it sometimes only pushes up four balls is OK. Whenever I've measured the throughput, it has exceeded the specification. There are several improvements planned for this module, but for now it's good enough. The most needed improvement would be to get it off baseplates and onto something that is easier to move and pack.

I'm using a 5600K (daylight) studio light when building. Separating old and new grays under that lamp is easy. I can even build black things. I used to do that until I encountered an instruction manual where they changed direction between steps. Now I align them according to my own preference.

I do the opposite - avoid introducing rotation where it isn't needed. Rotation will often introduce misalignment due to rounding errors.

Great progress. The LBG slopes added in step 36 are not turned correctly though. Good catch. I've added it to my model. To my knowledge it should be theoretically buildable. Didn't spend much effort on making it easy though.

I have now aligned things better. You can get the LDraw model and synchronization instructions here: http://torso.me/cardanlift That would be awesome. I'm waiting.

So, there's still interest in this module. I've made some tweaks to it after posting that video. Now there are no known reliability issues. I made a 3D model while designing it, as I find that easier than building with bricks directly. I'll align the last few things and post it. Not going to make instructions though.

You can buy balls from Bricks & Pieces. Search for part 72824.

Well done! Looks like it's running reliably. I like the uniqueness and simplicity. If you're planning on making a GBC conformant in basket, it's not just a matter of making it bigger. You will also need a mechanism to line up the balls, because that will not happen automatically.

There's not enough gear reduction to run the motor at full speed, so I don't think that would be advisable. I've been toying with the idea, but for now I prefer the flexibility of stand-alone modules. I also have several motors and can afford new ones when needed, so it's not really a problem.

That was a comment I did not expect.

Finished my second module. Simpler mechanics this time, but I haven't seen this particular variation anywhere.

I've only noticed that new parts will repel each other. It's difficult to make nice stacks while sorting.