Blakbird

[GBC] The Akiyuki Project

Recommended Posts

there is a video here:

from Alan Burchill showing Zig Zag Steps module from Akiyuki.

So may be not the exact same mechanism as Akiyuki but something that could serve this track.. (from Alan Burchill video, I have feeling behavior is quite similar)

IMHO replicating this module from Alan Burchill's video would be replicating Alan Burchill's GBC module and not Akiyuki's one since the mechanisms are different (see

at 1'05")

Share this post


Link to post
Share on other sites

IMHO replicating this module from Alan Burchill's video would be replicating Alan Burchill's GBC module and not Akiyuki's one since the mechanisms are different (see

at 1'05")

You're probably right Courbet. However, the video you shared seems to be a version 2 of Akiyuki Zig Zag Stairs module,

if you look at the original one from Akiyuki (

), there are more similarities with the one Alan Burchill did

Share this post


Link to post
Share on other sites

You're probably right Courbet. However, the video you shared seems to be a version 2 of Akiyuki Zig Zag Stairs module, if you look at the original one from Akiyuki (

), there are more similarities with the one Alan Burchill did

Alan Burchill: using cam, stairs do a little hop then stay level for 3/4 turn of the crankshaft.

Akiyuki (both videos) : stairs constantly moving up or down.

Furthermore, amplitudes in both movement is very different.

Share this post


Link to post
Share on other sites

Blackbird, I have got a last minute correction to the Pinball module: at the bottom of the conveyor belt, the pairs of DBG 1x14 technic bricks should be secured with an additionnal technic brick (1x6 will do) and 4 pins for each side.

It's obvious when you try to move this thing!

Edited by Courbet

Share this post


Link to post
Share on other sites

Zigzag Stairs, version 1 - a puzzle

I thought I'd have a go at reverse engineering version 1 of the Zigzag Stairs. I thought it looked small and wouldn't be that hard to work out. When reverse engineering, I tend to model things digitally, working my way through a model bit by bit and then build the physical model. While I have got some way through the digital build of the superstructure and the steps, I decided it would make more sense to physically build the movement mechanism and then digitally model it. The reference material is sparse, to my knowledge consisting of only one

. This has proven slightly problematic as there are insufficient views within the video to be able to determine the actual build.

Notes about the images below;

  • the images show "bare bones", sufficient to illustrate the build
  • the parts have been coloured different colours to try to make it easier to understand. The 1x2 thin liftarm is blue, the 1x3 thin liftarm is black, the 3x3 thin liftarm is brown, the "dead-point" eliminating connecting rod is green

My first attempt (shown below) consisted of 4 axles running within the frame. The length of the liftarms that were used and the half bush was based on what I could make out from the video. The first attempt is illustrated below. It does not work. When the arms are at the mid-point, they reach the "dead-point" and the "uncranked" end reverses direction.

640x444.jpg

Doing a little bit of reading, I determined that I could resolve the problem by inserting a 90 degree offset into play and this would eliminate the issue with the "dead-point". The second attempt (show below) works nicely. However, the Zigzag Stairs model requires adjacent steps to be travelling in opposite directions. This build attempt has the outside rods moving in time with each other and the inside rods moving in time with each other. What is desired is that the first and third rods are in time and the second and fourth rods are in time.

640x442.jpg

My third attempt (shown below) remedies the incorrect rod pairing. Now the first and third rods move in time and the second and fourth rods move in time. However, this build cannot be achieved without breaking the bounds of Akiyuki's model. Firstly the 3x3 thin technic lift arm comes into contact with the 1 x 10 plate on the end, so the entire mechanism has to be moved back one hole in the Technic bricks. Secondly, the vertical "space" used by the mechanism when rotating is too large. To allow sufficient vertical space for the mechanism to fully rotate, requires an additional row of bricks under the Technic blocks.

640x429.jpg

I've been unsuccessful with either coming up with a suitable resolution or finding a previous example of a solution online. So I thought I would post this message in case someone either has an existing solution that can be applied to the problem or can work out an appropriate mechanism.

If there are of any use to you, the LDraw files for three build attempts are available;

I'd welcome it if you have a solution and are able to post it here.

I know that the second and third build attempt diverge from Akiyuki's solution e.g. there is no room for the half bush visible in the video. At this stage I'm simply looking for an applicable solution and will consider how to align it with Akiyuki's build subsequently.

Regards,

David

Edited by djm

Share this post


Link to post
Share on other sites

Djm after looking at this I think you should go with the black and white version shown in the 2012 loop. It's design is a lot easier.

Thanks

9v system

Share this post


Link to post
Share on other sites

Djm after looking at this I think you should go with the black and white version shown in the 2012 loop. It's design is a lot easier.

Thanks

9v system

One version at a time :classic:

Share this post


Link to post
Share on other sites

After a bit more thinking, this:

zigzag_puzzle_4.png

has a more promising feel to it. Rather than trying to have 4 connecting rods plus a rod with a 90 degree offset, it has just 2 connecting rows plus a rod with a 90 degree offset. The raising and lowering of a pair of stairs can probably be achieved using something along the assembly on the left hand side.

I still need to explore this a bit further but it feels more closely aligned with the build in the video.

Regards,

David

Share this post


Link to post
Share on other sites

Here's a video of the first draft build of the underlying step mechanism. It feels like I'm heading in the right direction. Excuse the shaky cam - I was holding my phone in one hand while turning the crank with the other.

Regards,

David

Edited by djm

Share this post


Link to post
Share on other sites

After a few adjustments from trial and error, here is a compound image showing the current state of the prototype of the lifting mechanism for version 1 of the Zigzag Stairs.

zigzag_prototype.png

As with the earlier images, the green parts are simply being used to denote the parts used to achieve the 90 degree offset. I'll continue to build this physically before modelling it digitally. In the meanwhile, the LDraw file for the prototype is available for downloading.

I'm aware that the original model does not have the tiles on top of the Technic plates. I'm presently modelling in this fashion as the space available underneath the Technic plates is very tight. Using tiles on top rather than plates underneath the Technic plates seems to alleviate the space constraint a little. This may be amended if the build above the prototype frame won't work with the tiles in place.

Regards,

David

Edited by djm

Share this post


Link to post
Share on other sites

Glad you found a solution to the issue you identified earlier. People in Robotics would tell you that the initial mechanism was not valid because it has a singularity in its Jacobian Matrix ...

If I recall correctly, I think Akiyuki used gears to link the motions of the various axles, though I can't locate the video now.

Share this post


Link to post
Share on other sites

If I recall correctly, I think Akiyuki used gears to link the motions of the various axles, though I can't locate the video now.

DrJB, was

you had in mind? If so, I think that one might use a mechanism similar to the Basket Shooter to achieve the vertical motion. This module build is for the GBC in this video:

which does not appear to be using gears from what I can make out.

Regards,

David

Edited by djm

Share this post


Link to post
Share on other sites

DrJB, was

you had in mind? If so, I think that one might use a mechanism similar to the Basket Shooter to achieve the vertical motion. This module build is for the GBC in this video:

The former seems to be a clearly updated and improved version of the latter. Are you working with the older design just because better views are available?

Share this post


Link to post
Share on other sites

The former seems to be a clearly updated and improved version of the latter. Are you working with the older design just because better views are available?

I started with the older version simply as I liked the challenge to see if I could build it. Primarily an intellectual exercise. With the slightly better view of the older one being available, it did strike me as being an easier build task to try and accomplish. It may stimulate someone else to try to reverse engineer version 2. Version 1 also has a compact footprint and I can see potential for it being used as a gap filler for when our LUG has has GBC displays.

Regards,

David

Share this post


Link to post
Share on other sites

Here's the video of the (rainbow) build Akiyuki's Zigzag Stairs GBC (version 1).

Draft Ldraw file/PDF will be posted once I build it digitally.

Regards,

David

Edited by djm

Share this post


Link to post
Share on other sites

zigzag%20stairs_draft.png

For those who are interested, a draft of the build instructions of the Zigzag Stairs is available as a PDF on bricksafe .The LDraw file is also available for download. A parts list for the draft build is available on Rebrickable (where you are able to "like" the MOC, should you wish to).

My observations about the module;

  1. it is a straight forward build, requiring no timing adjustments. It simply works
  2. when building, manually test the smoothness of the vertical axle movement as soon as you have placed the Technic plates sub-assembly on the frame. Ensure the thin liftarms hit neither the axles beneath it or the Technic Plates above. Adjust the vertical position of the thin liftarms as required.
  3. the motor connection is a total unknown. I'm not aware of any images of the left hand side of the original module. I built this module with no gears and the motor connected directly to one of the axles. I powered it with a LEGO rechargeable battery box and on a low speed setting. You should be able to achieve something similar if using the train controller. Don't run the motor too fast as the balls jump all over the place.

Regards,

David

[Edit: repointed image and files to draft version)

Edited by djm

Share this post


Link to post
Share on other sites

Why does my dog hate my spiral lift?

How did you dog manage to do that?! Do you have a Great Dane or something of that size?

Share this post


Link to post
Share on other sites

Why does my dog hate my spiral lift?

I think your dog added their MegaBlocks extension to the module! :cry_sad:

Share this post


Link to post
Share on other sites

I know GBC balls are expensive, but I think you are trying a little too hard to get creative on an alternative...

Edited by TheRebricker

Share this post


Link to post
Share on other sites

That is why, if you buy old sets from eBay, the first thing you need to do is sanitize/disinfect/wash thoroughly .... :)

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

  • Recently Browsing   1 member