Ankoku

This is one of the more famous MOCs that LEPIN stole: They lost a copyright trial brought by LEGO: https://www.bbc.co.uk/news/business-42275566 and recently got arrested: https://www.bbc.co.uk/news/world-asia-48076505 But there will be more of them. Also, you get those who just sell stolen instructions etc. on eBay.

Did this guy ever post this here?

Worked it out. They seems to be a new pattern (etching) style of balls in both orange and white. With the new style, the pattern is no longer uniform. I am guessing it relates to bringing the balls up-to-date with current soccer ball designs.

As far as I know, LEGO doesn't sell printed soccer balls any more?

Yours should be touching less than mine with a 1 stud difference. Either that, or mine is 1 and a half shorter. Honestly can't remember. I think it is just half though.

When it comes to GBC balls, do you know what the difference is between the two variants LEGO has for each colour? e.g. white has: 6261261 & 6018511 and orange has: 6261264 & 6023209 Since LEGO has them as difference things, I am curious as to whether there are any physical differences?

I posted the superbird28 vids in the Akiyuki thread. He mentions it in the Blog for the factory. Akiyuki has version 1 and 2. The differences between the two are quite large. Would I say that superbird28's are 0 and -1? Not really. They are very different beasts, especially since they don't use balls. I find the most interesting bit with superbird28's designs, is that he basically has a main driver area which controls the timings of everything. Having all that in a single area hasn't really been replicated since. A lot of the factory is there in version 1, but he uses different mechanisms for achieving some of the movements and counter balances stuff more. The original version with more liftarms!

I went for a 1/2 stud shorter arms, just to stop them pressing against the ground.

That feels as though it ought not to be needed.

Yeah, I had a relatively simple GBC in Stud.io which has 154 Unique items and 3278 parts. It lags when you try and add a piece and I have had it completely lock up on occasions. It is a more powerful machine that most people have, so I feel as though Stud.io has some serious programming issues they need to overcome. There is no excuse for models this simple.

The two images are the same image.

You could do your own solution for the feedback loop. e.g. with the new design, you aren't limited by height or the base plates, which are the two constraining elements of the original. It makes much easier just to wedge on the belt like your version has. That isn't possible on the original version with the corrections as there just isn't space. Which is my I am going to raise the whole thing and remove the base plates.

It is good that they offer that feature. If you include both used and new and go with the traditional colours, that 47.4% costs about £200. Although that is across 13 sellers, so postage costs ahoy :P Very doable though.

I suspect the difference in parts is huge. e.g. there are not many liftarms in the original version, where as the new one is mainly liftarms. There were lots of standard bricks in the original, but few in the new version. The bonus is, the more obscure stuff is replicated. e.g. belts, buckets, curved bricks, that sort of thing. If I make this version, I won't cannibalise the old one for it. Will probably just share the buckets between the two.

I suspect the hoses will be in the parts list and the connectors will be in the right places etc. It is just that rendering the hoses in a curved manner is a nightmare in the software. It just isn't worth the effort.

The "will-be-added-in-two-weeks.com" is a bit of a dangerous game to play. It ends up with browsers redirecting to random sites.

Kudos for bookmarking the correct time in the video where it appears!

Got a video of it in action?

I am hoping they do a: "Retiring Soon" or an announcement of some kind before stopping production. Give people a chance to stock up. I wish the PoweredUp stuff was better than it is. You get the feeling it is going to be very dependent on 3rd parties for cables etc. I have had a number of issues with 2nd hand LEGO motors.

For motors, I prefer them new and preferably from LEGO themselves.

Okay, so here are some of the changes I have been working on, which I will called the 2012/09 version, as that is the Akiyuki version that it is based off. The one in the 2012/09 GBC Loop video on his channel. So, I changed the wedge that directs the balls. It doesn't show in this image, but basically, it is identical to what you see in video. No idea why it was changed, but this works far better. The upper section was also raised one plate. This pulls the purple tube closer to the rotating unit. It also make it easier for the ball to get into the cup. I have no idea if this is closer to the Akiyuki original or further from it. I have not bothered to check. It just makes sense and works way better. To bring the upper tube even closer, I have used a Technic, Axle and Pin Connector Toggle Joint Smooth. This stops balls causing increased resistance as they get stuck between the rotating unit and the tube. Oh, I also changed the Technic, Axle and Pin Connector Perpendicular 3L with Center Pin Hole to 2x Technic, Liftarm 1 x 3 Thin for the cup holding arm. This is a no brainer. Smoother, more secure. You can also see the double universal joint at the bottom. As ever, 2 is better than 1 for smoothness. It wobbles about a bit, which is ugly, but I can live with it for now. Can't see much here, but this is the new gear box. The use of the 12T and 24T gear together caused a lot of resistance. It was to create a 4:1 ratio. So I replaced it with 2x 8T, 2x 24T and 2x 16T gears. It is 2 additional gears, but it runs with a lot less resistance. This really improved things. Bracing is messy, but it is more secure than the old gearbox. Of my changes, this is the most satisfying both in performance and solution. Here you can see the change to the to-and-fro arrangement. Akiyuki has changed this time and again. Which leaves me wondering as to whether this arrangement has issues which I do not yet know about. It is the version you see in the 2012/09 video. The 16L axle is immaterial. The important bit is the use of 2x Technic, Axle and Pin Connector Perpendicular Double Split which differ from the instructions. The one nearest the gears is no different to how it was in the instructions. Different piece, same end result. Although this only needs 1 piece, rather than 3. It is a change seen in later versions of the Ball Factory as well. I think Akiyuki just liked using them at some point. Where things differ though, is the use of it near the gearbox. This moves the axle further down at this end. As far as I can tell at the moment, it stops the to-and-fro from locking up. It didn't do it all the time before, but often enough. Also, this is easier to set up, as the two pieces sit close to the ends of the axle. So here is a video of the machine before the changes. Although I had already added the additional universal joint for instant improvement. As you can see though, it is slowing at times. This is mainly a gearbox issue. Now, this is the current version. It isn't perfect as the abundance of balls in the hopper can make the spiral slow slightly, but other than that, especially at the end, without balls, you can see it running smoothly. I am still looking to change the hopper to stop the balls from hanging around unable to make it up the spiral. I don't have the bits for it though. I like how the machine and motor sounds a lot more now. Should have cleaned the phone lens before hand :P Anyway, I am pleased with the progress so far. Need to order a bunch of pieces to make it look better.

Should we be worried?

Okay, so in the end, I went with a: And a normal pin. This sorts out the upper part of the lift. Having some issues with the to and fro mechanism and noticed that it is very different to the original: Do we know why it was changed?

Ended up using this: Along with this:

Well, the new gearbox for Cup-to-Cup Type 1 is a big improvement. Far less resistance now. My main problem now is the way the tubes for the spiral lift are connected. One a 3L pin you always have a ridge on either one side or another. If it is on either side, it can cause enough space for a ball to sink down a little bit or even drop through. This additional resistance can cause motor strain and a rather visible slowing of the cup-to-cup part. For now, I have replaced the 3L pin with a 3L axle, which has improved things. Raising the lower part or upper part of the lift 1 plate and having the pin ridge on that side might solve the issue, but I have already raised the height of the lift by 1 tile and increasing it again may create more problems than it solves. Anyone got any ideas? I have a before and after video of smoothness. Will post them at some point.