[GBC] The Akiyuki Project (2010 - 2020)

[jesuskyr]

Now I am working on Catch and Release module. It's simple and effective:

[Blakbird]

Instructions for Akiyuki's Tilted Rotors module are now tested and complete and are available here along with a Brickstore parts list.

• 89 pages
• 1223 parts

It seemed like it would be a simple matter to create 3 identical rotors each with 4 identical platforms. Wow, was I wrong. First of all, each rotor is different. Each has shorter arms and therefore the ball gradually spirals toward the center. Not only does each rotor have different platforms, but even within a given rotor there are as many as 3 platform types. After agonizing for many hours, I realized that at least the placement of the platforms had to be 90 degree symmetric, and a top view should prove that the platforms are aligned to allow a ball to roll from one to the next.

They good good statically, but what about dynamically? When they are tilted, do they interfere? To answer that question I had to make an animation:

The animation shows that everything appears to align perfectly. (Yes, I realize that the torque links in the animation are fixed and interfere with eachother. I only wanted to prove that the model would work, not make an accurate kinematic simulation. I've already done that for Sheepo's KA-32 rotor head.) Furthermore, once the equations for the animation were complete, it was an easy matter to render it again but this time with the camera revolving around the mast at the same speed as the rotors. This does a better job of showing the motion from the point of view of a ball.

I couldn't help but notice how cool the stackup of frames was when I was making the animation, so I saved that as well. It does a good job of showing the overall envelope of motion:

The initial build was difficult because I had quite a few errors in the instructions. All have since been fixed. Mechanically, this is a very reliable module but as Akiyuki himself admits, "The ball spilling is not zero". In fact, it was spilling so many balls that I was considering using some LEGO nets as a catching device so I could collect the dropped balls in an external hopper and prevent the gearing from jamming on them. However, it appears that the secret is the rotation rate. Only gravity causes a ball to drop from the waiting indexer into the first rotor. If the speed is not right, the ball will either fall too soon or too late and miss the platform. Similarly, the transfer from the first to second rotor involves crossing a gap and this only works reliably with the right speed. I found that the 7.2V from the rechargeable battery box was just right. A full 9V box is a little fast and you can't slow it down. A train regulator at full speed is also a bit fast, but if you throttle it back one notch then it is perfect. Once you get the right speed, it only drops about 1 ball in 100. This seems related to the little divot on the ball slowing the drop very slightly if it is sitting just wrong.

Since the indexer is lifted by the passing of a rotor arm, it does not need to be timed. Although the lift supplies one ball per cycle to the indexer, the timing is somewhat important. If the ball is provided while the indexer is already lifted, the ball can easily fall. A ball needs to be supplied when the indexer is down. This is only timing to set up.

There is one very important thing to mention about the build. Not every turntable is created equal. Before I started building, I had about a dozen turntables laying around. I found that some turned very easily while other turned with considerable difficulty. I don't think the model would have worked with the latter. I chose the 4 best turntables from my stack. I also found that if you take the turntable apart you will find a couple of small spherical bumps on the gray part. These cause friction and also stabilize the platform. I found that if you shave these bumps down a bit it reduces the friction and makes the model run much better. My model works fine on an M-motor and doesn't struggle at all. Also make sure that you use the 1x4x3 transparent panels without the side supports for the outlet ramp or the balls will not roll.

This model would NOT work running backwards and in fact would probably destroy itself so it contains a ratchet to prevent that from happening.

[MajklSpajkl]

Wow, Blakbird, you sure work at an incredible pace, thanks for all the hard work. Hopefully I'll be able to build those modules in full one day, for now they will be a great learning examples.

I was keeping myself busy for a couple of evenings with reverse engineering the Akiyuki's Pneumatic module. Some bits I changed, but not to much to really change the design, I think. Unfortunately I have really poor skills in LDD, none in MLCad so, if needed, I can provide some photos (still have to take them :-)), to make it easier for someone who has a way with digital building.

Anyway, I am really thrilled about this module, this is my first pneumatic loop ever, so now I understand the principle a bit better. Here's a little video from my cell, so I apologise for rather poor quality. I only used Akiyuki's video as a source and I almost felt, as he was trying to hide some parts in order not to make it to easy for us to figure it out :-)

Edited December 10, 2015 by MajklSpajkl

[Blakbird]

Anyway, I am really thrilled about this module, this is my first pneumatic loop ever, so now I understand the principle a bit better. Here's a little video from my cell, so I apologise for rather poor quality. I only used Akiyuki's video as a source and I almost felt, as he was trying to hide some parts in order not to make it to easy for us to figure it out :-)

Nice! I like that you used a LEGO compressor. How many cylinders did you use in the compressor? It would be nice to be able to display this at events without a shop compressor.

[MajklSpajkl]

Thanks! I was always planning this one with a compressor. I used two new pumps from 42043 Arocs. They're placed at 180 degrees, so the presure is as constant as possible. PF L does the job well without any gearing. The compressor was built very quickly and I wasn't trying to make it compact, there is plenty of room :-)

About the module: I had some problems, as the bucket was to high in lower position. The gears of lifting mechanism and the damper spring must be sinchronised correctly in order to get the bucket low enough. Also tried it without the damper and it gets quite brutal on the basket :-)

Edited December 11, 2015 by MajklSpajkl

[me00rjb]

I keep telling myself to avoid this thread; it’ll only end in tears! But I keep coming back, and I know what’s going to happen in the end!

[nychase]

I keep telling myself to avoid this thread; it’ll only end in tears! But I keep coming back, and I know what’s going to happen in the end!

My tears just started....

Its about 1600USD for the models up to tilted rotors....and you can forget about the balls needed.

To do us all a favor I have ordered 3 different types of 14mm "beads" from china. I hope to test them thoroughly to be able to recommend which can be ordered as a perfect proxy for the unattainable soccer/basketballs. The prices are of course a joke in comparison and I think it will help everyone build and enjoy these models. I do have an arsenal of the soccer/basketballs anyway so the test should be apples to apples.

[Theo van Vroenhoven]

To do us all a favor I have ordered 3 different types of 14mm "beads" from china. I hope to test them thoroughly to be able to recommend which can be ordered as a perfect proxy for the unattainable soccer/basketballs. The prices are of course a joke in comparison and I think it will help everyone build and enjoy these models. I do have an arsenal of the soccer/basketballs anyway so the test should be apples to apples.

What about a kickstarter project that produces the perfect replicas?

There should be a company in China that can make these balls according to specs?

If they can reverse engineer computers overthere they can certainly reverse engineer some plastic balls... ;)

[nychase]

What about a kickstarter project that produces the perfect replicas?

There should be a company in China that can make these balls according to specs?

If they can reverse engineer computers overthere they can certainly reverse engineer some plastic balls... ;)

I am not actually opposed to this though I wouldn't know how to find a vendor in china. I am certain that we could sell 20,000+ of them but I'm no expert on the legalities of reproducing a lego part (certainly megabloks knows a thing or two). Perhaps the answer is to create a ball of the same weight/size with no moldings (defining it as a soccer ball or basketball). It will be important to have the molded ball be multi colored (perhaps like a beach ball) because having the lines helps to show the movement and rotation of the balls and that is not something I would want to give up. It will also be important for the multi colored aspect to be molded and not painted since the paint will certainly wear off with active use.

Thoughts?

Edited December 11, 2015 by nychase

[Blakbird]

I am not actually opposed to this though I wouldn't know how to find a vendor in china. I am certain that we could sell 20,000+ of them but I'm no expert on the legalities of reproducing a lego part (certainly megabloks knows a thing or two). Perhaps the answer is to create a ball of the same weight/size with no moldings (defining it as a soccer ball or basketball). It will be important to have the molded ball be multi colored (perhaps like a beach ball) because having the lines helps to show the movement and rotation of the balls and that is not something I would want to give up. It will also be important for the multi colored aspect to be molded and not painted since the paint will certainly wear off with active use.

Thoughts?

Sounds like a pretty good idea to me. Make it so.

[freakwave]

I am not actually opposed to this though I wouldn't know how to find a vendor in china. I am certain that we could sell 20,000+ of them but I'm no expert on the legalities of reproducing a lego part (certainly megabloks knows a thing or two). Perhaps the answer is to create a ball of the same weight/size with no moldings (defining it as a soccer ball or basketball). It will be important to have the molded ball be multi colored (perhaps like a beach ball) because having the lines helps to show the movement and rotation of the balls and that is not something I would want to give up. It will also be important for the multi colored aspect to be molded and not painted since the paint will certainly wear off with active use.

Thoughts?

Are you member of a LUG? The balls are on the list of possible items to be ordered via LUGBULK. Definately a better deal than Bricklink and as the name suggests, you can order in large quantities...

[nychase]

Are you member of a LUG? The balls are on the list of possible items to be ordered via LUGBULK. Definately a better deal than Bricklink and as the name suggests, you can order in large quantities...

No....no I'm not, I'm a loner here in NYC...although probably joining a LUG in milwaukee when I move.

Does anyone want to adopt me for now? I would be willing to buy 2000+ balls depending on the price. Also, is it the original soccer/basketballs or the newer blank designs? As I mentioned the newer designs pose an issue in that they don't really show the movement the balls go through (rotations specifically) as they run through the GBC's.

[MajklSpajkl]

Unfortunately they are blank. This year there are only white and orange (football mold). If I remember correctly, last year they were also available in white with purple heart on them - from friends series. They're about a quarter € per piece...

[Mogwai]

Hallo , kann ich hier auch in Deutsch schreiben ?

Mein Englisch ist nicht so gut aber ich probiers mal .

I have made the Cup to Cup Modul but without the loader .

Found new Cups and they are working fine with the standard balls and Zamor Spheres

The Original Cup/Dish cost 6,- Euro per piece

How can i upload Photo´s ?

[Mogwai]

I have found the Option for upload

here the Pictures of : Wheel to Wheel

[MajklSpajkl]

@Mogwai: I am wondering if fake engine parts would do the job...although both solutions are rather bulky...

[Mogwai]

@Mogwai: I am wondering if fake engine parts would do the job...although both solutions are rather bulky...

Why fake engine parts ? it`s normal Wheel 18 x 14 with Axle Hole, Fake Bolts and Shallow Spokes nr.55982

I search a solution for the very expensive LEGO Part 71128 - Dish 2 x 2 x 2/3 Light Reflector (6,-€ per piece)

and try this Wheel. It´s very nice work and a big plus is to work with Zamor Sphere too . This is evtl. a way for

People they have not Soccer/Basketball/Beach Balls and can build this module . At the Moment i try to build a

loader which work with the two balltypes . Sorry for my School english

Edited December 14, 2015 by Mogwai

[nychase]

Found new Cups and they are working fine with the standard balls and Zamor Spheres

The Original Cup/Dish cost 6,- Euro per piece

How can i upload Photo´s ?

Don't forget you're going to need a few more for the basketball shooter (and I don't know how you would sub those). Honestly though they aren't 6 euros, I just bought a few for around one USD. Remember that you can use the most worn-out ugly ones because you only want the shape, the condition of the chrome is irrelevant. Bad chrome versions are sold all the time at a steep discount.

Unfortunately they are blank. This year there are only white and orange (football mold). If I remember correctly, last year they were also available in white with purple heart on them - from friends series. They're about a quarter € per piece...

I will reserve judgement until I know for sure if the beads will work. I was able to find beads that don't have the hole drilled so they shouldn't get stuck like the others.

Edited December 14, 2015 by nychase

[PeterF]

it`s normal Wheel 18 x 14 with Axle Hole, Fake Bolts and Shallow Spokes nr.55982

Great alternative! I have those lying around anyway. Just like the zamor spheres btw.

I might try to build this now.

Oh, and worry about your English. It's fine.

[Tombacus]

First of all I really want to thank all of you here who have been doing the hard work to reverse engineer and document Akiyuki's wonderful machines. So far I have built the ball factory and the invisible lift, I am in the process if building cup to cup and I am starting to collect parts for the bucket wheel tower. Yes... it is an addiction...

My question to the group is this. Do you think anyone out there would be interested in buying a full set of the parts needed to build these wonderful machines? It would be easy for me to collect more than one set of the parts as I am doing my own builds and I could offer you guys a full set of parts. Would anyone find that helpful?

[power1369]

Hi there.

For those that have help the Ball Factory, would you be able to tell me if the (2) 48288 plates are required or can they be replaced with something else?

Thanks!

[nychase]

Hi there.

For those that have help the Ball Factory, would you be able to tell me if the (2) 48288 plates are required or can they be replaced with something else?

Thanks!

Totally interchangeable with This part. This is still available from lego (at least in the US) and costs 1.70.

[power1369]

Awesome thank you very much !

[edp]

The 15395 2x2 inverted dome is a reasonable substitute for the reflector dish. It sits a bit higher than the reflector dish which needs to be accounted for but otherwise it is not too bulky. I built a simple shooter module based on Akiyuki's mechanism and it worked well, again with modifications to accomodate the added height.

[jesuskyr]

Hello^^ I just finished Akiyuki's Catch & Release(Certainly only cad -.-).

He has showed the two versions of Catch & Release in Youtube.

I followed the newer version:

This is my interpretation about his module:

I upload this file at my bricksafe folder : http://www.bricksafe...i/catch_release

Edited December 15, 2015 by jesuskyr