Thierry-GearsManiac

Do you mean his (Akiyuki's) mods and improvements or my own ones ? Akiyuki's mods and improvements (except the turntable optimization = removal of the base, already taken into account in the build instructions) : I would like to build a larger input bin so that it's more convenient for chaining modules (when doing turns). My own mods and improvements : I wanted the robots to spin symmetrically (and, by mistake, I mirror-built one robot with respect to the other one when anticipating the build of the other doubled subsets --which were actually symmetrical) I didn't succeed in making the robots work reliably : none of the friction adjustments did work for me It was a good self-challenge The pleasure of doing mechanical design, even on somone else's original creation

Interesting : although I already saw this yellow version, I didn't even notice the extra-large (and much more practical) input bin, which is a good idea to reproduce... Having built the standard version a few days ago, but failing to tune the module, I modified a little portion of the mechanism (described in a new topic in order to avoid disrupting discussions here). See

A last thing about 2:1 gear ratios : have people already used the perpendicular solution between a Z24 contrate gear and a Z12 ? One can see it my the above mechanism, if we only consider the meshing between the Z24 contrate gear and the Z12 idler gear. Or in the sketches below, where several rigid stop solutions can be used to counter the axial force applied to the contrate gear : a pin with towball (black) two plates (red) an old Id=424 spacer (yellow), as discovered here :

A 1:4 ratio can also be done the planetary way, using either Z8 pinions and the Z24 inner gear from big turntables v.1 and 2, or Z16 wheels and the Z48 inner gear on the 64712 part (see https://www.bricklink.com/v2/catalog/catalogitem.page?P=64712#T=S&O={"iconly":0}), as shown on Yoshihito Isogawa's Power Functions Ideas book, idea #194. When the planetary mechanism is built with an inner gear as the stationery one, the sun gear / planet carrier ratio equals the (sun gear / ring gear) ratio plus 1. However, if the stationery gear is an outer one, then the sun gear / planet carrier ratio equals the (sun gear / ring gear) ratio minus 1. For example, sun gear = Z24 ; planet = combo Z24/Z8 ; stationery gear = Z40 (but in this case, the planet carrier will be hard to interface with) At last, some other possible 2:1 gear ratio solutions : Z20 to Z40 (http://www.planet-gbc.com/modulegbc-torso-cardanlift/) Z28 to Z56 (big turntable v.1 or 2) hidden in a longer gear train, by inserting an intermediate 2:1 combo gear (i.e. two gears on the same axle with LEGO) between two of the gears (see attached example, where the intermediate combo gear is the axle supporting the contrate Z24 and the single-beveled Z12, and the ends of the gear train are the two Z20 gears --the blue one rotates freely on the axle : https://www.bricklink.com/v2/catalog/catalogitem.page?P=35185)

Goals for this topic : Describing the inner workings of the module : highlighting the tricks used by Akiyuki for automating the robots' movements (without compromising his copyrights on the model and building instructions --sold at a very reasonable price, around 8EUR) Showing my mods and improvements on this module, and gather other people's ones and comments too, of course Like everybody, I was impressed when discovering this module for the first time. And although I partially guessed its working, I only discovered some mechanical tricks much later when watching the video again... and even a last one a few weeks ago when I built it ! That's why I want to share all of this. Official Akiyuki's video, as a reference : Now, (1) how it works (hidden for people who don't want to get spoiled, but prefer guessing) (2) My mods and improvements Here is a bottom view of my build, where we can see the most : (a very very simple mod) Bringing the mechanical power from outside : replacing the M motor with an axle which comes out at the right (sorry, hidden by the white beam supporting it) (mod) Spinning the robots in opposite directions : replacing part of the central transmission unit (see Akiyuki's build manual p79-83) : central drive of both robot units made through a right angle gearing, see the 7L axle with Z12T gears at each end and a Z20T gear inside the fork. I also replaced the two Z16 input gears (below the 'L' beams) by Z20 ones. I forgot to say that the robots must be mirror-built too (otherwise one robot will have its crankshaft spinning in the wrong direction, and its front being pushed downwards instead of upwards when it grabs a ball). (for information, the original transmission has the Z20T gear parallel to the robot units' bases, driving them through a parallel gear train ==> odd number of gears from one robot unit to the other ==> same direction) (mod & improvement) Robot's position auto-reset if it loses its ball : When the robot begins to spin with a ball in its hands (stage 3), it holds it too weakly because it is too light for the roller to roll down the tilted platform's slope and maintain the hands firmly enough. Although Akiyuki saw and tried to fix this problem by adding adjustable friction to the spinning of the robots' bodies, this makes the problem even worse when a robot drops a ball while in stage 3 (or sometimes when the robot exits its fast spinning phase = stage 4, without its roller falling into the well) because it stays in its position forever without human intervention. That's why I replaced the adjustable friction generators with a detent mechanism which forces the robot to return to its home position (or to hold the ball more firmly), see the Z12 pinions at the rear, the partially hidden Z24 gears (with rollers mounted on opposite round holes) and the blue angled 4/4 liftarms acting as pawls. My initial try was a simple crankshaft with a spring-loaded piston, but the shaft's torque around the home position was too weak, whereas it was too high around 90°(applying too much force on the robot's hands mechanism which has a high flexibility and backlash, so that it skips a turn, making the robot throw the ball away and returning home). The alternate solution is a cam with the correct shape... which doesn't exist or can't be built in an enough small size to fit in the remaining room. And the solution came by dividing the robot's body rotation by two, hence a bi-foil cam looking like a technic beam (here two rollers in order to decrease wear) ! For more clarity, I can make 3D mockups if needed.

Remark of less importance : in the train's drivetrain, I'm surprised about the use of a single-bevelled Z20 gear on the motor's output for the meshing with the black Z12 double-bevel gear (by the means of the bevels) : isn't it less efficient (and a cause of teeth wear) than using a double-bevelled Z20 for a perfect spur meshing ? Or is it intended to force the Z12 gear closer against the single-bevelled Z20 on the wheels' axle (i.e. angular meshing made tighter) ?

About to build it too (missing parts not received yet), after quickly looking at the mechanism on the 3D model, I noticed this too and intended to implement this optimization too. For people who haven't guessed yet how it works, this part of the mechanism spins the robot faster than its central axle on the last stage of its spin move (thanks to the Z24 eccentric gear, acting like a partial gear, which engages with a fast spinning gear), so that the hands mechanism gets driven backwards from the robot's point of view, making it open its arms, thus releasing the ball. This mechanism couldn't be built directly on the robot's body, due to its complexity. That's why it's remotely placed (with its 1:1 gear ratio to the robot's body built around these Z28 gears and turntables). Also, glad to know that it works, and totally convinced to do it this way too. We could then now submit this slight optimization proposal to Akiyuki, couldn't we ? (I'm aware he designed the "Catch & Spin Robots" module before the availability of the new Z28 gear).

... and it grinds even more and blocks slightly when the axles are held on both sides (i.e. non-cantilevered mounting), after a quick try. (note : sorry for interleaving this answer with the latest sub-discussion ; a good idea would be to transform this discussion into a sub-forum with one discussion per part/assembly)

... whis is, of course, a personal remake of a famous design from Akiyuki ! (because I admired how he turned/staged industrial speed reducers into ball transportation mechanisms) Another motivation was that some famous GBC designers (Josh DaVid, Sawyer) re-create well-known designs in a smaller form factor. And of course, the "self-challenge". So here is my currently final work (despite not yet well-polished regarding input and output's aesthetics) : UPDATE 20200410 : link to a higher quality video of a test : https://diode.zone/videos/watch/fcfe0564-1035-4bb4-b24a-7f0dd04d943d As introduced before, I haven't changed the core mechanism, which relies on the two following main tricks : using 180-53° liftarms for building the heptagonal structure of the "rotor" (only approximately 2° of mechanical stress per side) central pivot achieved by a pseudo gear bearing (pivot = sun gear ; rotor = planet carrier ; no ring gear) because no 7-beam hub does exist in LEGO and less than 7 beams is impossible because 7 is a prime number (whereas Akiyuki's rotor has 9 sides ==> 3 --double-- beams on a 6-hole pulley) The only changes I made (since the above older pictures) were : on the stator, new support structures for the big "teeth" and a brick-based rear wall (for preventing the rotor from wobbling and jamming) the ball catch mechanism : instead of the previous 3-finger layout (with one moving finger), I switched to Akiyuki's 2-(hollow)finger solution, using old crank parts and the material's flexibility : it takes less room, making the ball catching easier the ball can be pushed in/out in perfectly opposite directions, making it possible to load it at the very bottom and to unload it at the very top Then I managed to build the output ramp and input mechanism (input bin with basic steering + 1-by-1 ball distribution) and to adjust them, after several trial-and-error steps too. (I initially attached all the elements on a baseplate in order to quickly adjust them with respect to each other, before linking them together in the form of a hollow but stiff chassis). The mechanical inputs are a crank (behind the stator) and a motor input (behind the input bin).

After no post since more than one month, I'm back now. I managed to pick up a significant amount of my childhood's LEGO stock, and therefore I've been able to continue to work on this first project and turning it into a complete module and make it enough reliable : I'll rather switch to a dedicated topic for showing it .

A problem remains on the clock design for the hand in question : the Z16 gear with crenels seems unreachable for meshing because it is located right in the turntable's hole. Behind it, there seems to be another Z16 one (with a round center hole too) driving it, but how ? a clutch ring would allow too much backlash (almost 2*90°) hot glue ? Not destructive but not pure LEGO rubber (some small LEGO tire or non-LEGO stuff) ? A link to the clock design would be interesting in order to investigate this. (and something else : just notice the clever use of the outdated Z14 bevel gear, meshing with the Z56 V2 turntable : exact 1:4 gear ratio / no more 7 prime factor in it !)

I made a try mating two 654227 together, but the total thickness will be less than two studs, more precisely two studs minus one crenel height ==> a bushing (or anything else, which will be placed "off-grid") will be necessary for holding these gears together. One can also mate this gear to an old technic plate with crenelated holes (https://www.bricklink.com/v2/catalog/catalogitem.page?P=4263#T=S&O={"iconly":0}) in order to freely/independently rotate/spin something on the axis. Another subtle difference between these two parts : you can't mount the old version (654227) on a pin because there is no widening on the center hole for locking the pin's "tabs".

OK : the measurements you'll obtain (voltages on different LED strips with power turned on, resistance of one resistor with power turned off and preferably for an unused output port) will confirm the assumptions. I wasn't aware that the kit was bundled with the car (a fan's model ?) : I believed it came from marketplaces like EBay.

A few posts before, about the question if the output ports do generate different voltages : they are identical, then they don't do it by themselves : on ports where no load (one of our LED strips) is connected, the expected voltage is the supply voltage = 9V, because the circuit is open <=> no current flows through the resistor <=> the voltage at its terminals is 0V on ports where a LED strip is connected, the expected voltage will be the LED's forward voltage because the LEDs clamp the supply voltage down to their forward voltage, forcing current to flow through the resistor, which "sees" the difference between the supply voltage and the LEDs' forward voltage. Therefore, with white LEDs, you can expect 3V approximately on the port, and with red and yellow LEDs, you can expect 1.8-2.0V approximately ==> the resistor will then see either 6V or 7.2-7.0V, which results (according to Ohm's law) in a slightly higher current for the red and yellow LEDs than in the white ones. For measuring the voltage on the LED strips, if you don't want to do it on the surface-mount leads of the connectors (they are very close to each other), you may be able to do it on the strips themselves, either on the cables' '+' and '-' solders, or on the '+' and '-' pads between the LEDs. (and for the blown TV's LED, it's the thermal aspect that has not been studied/tested enough by the engineers : when dealing with higher power LEDs, even when operating them below their maximum current, one can destroy them by overheating if not enough heatsinking is provided for draining their heat away)

@phaenius: After a look at your pictures, I see that the power distribution board has a resistor (SMD, probably 0603 size) in front of each LED strip connector ==> my assumption of three LED dies in each LED package is probably wrong. And since you have red and yellow LEDs, their forward voltages are lower (around 1.8V for red and 2.0V for yellow, versus around 3.0V for pure green and blue/white LEDs). ==> the resistors do matter and are dimensioned for supplying 4 times the nominal current of a single LED (4 LEDs / strip, assuming equal current sharing between paralleled LEDs although it's wrong) ==> don't cut your strips, otherwise you'll overpower = fry your LEDs ! (unless you hack the strips in order to insert extra resistors or reuse only the cable with the female connector in order to build a custom LED circuit) For further hacking purposes, it would be interesting to make the following measurements : the supply voltage (on the black cables' solders)... I guess around 9V since it is today's LEGO standard the LED strip voltages for each colour (on the surface-soldered terminals of the PCB connectors where the strips are connected) the resistor's value (on one of the resistors' terminals, with power turned off and everything disconnected)... unless its value code (3-4 digits or 2 digits + 1 letter) can be read by the help of a magnifier

On my side, after thinking, according to some clues, I may make this assumption about your LED strips : if they are : powered on a 9V supply (typ. batteries) directly connected to this supply without any extra resistor then : each LED package probably embeds 3 LED dies in series, each one having a forward voltage drop of approximately 3V (typical for blue/white LEDs) the current limitation may rely only on the LEDs' intrinsic resistance (and the batteries' one too), but with the drawback that slight voltage variations lead to large current variations (this resistance is low : a few tens of Ohms) : when the battery charge drops only a bit, the light intensity will decrease significantly one can't power the LEDs on a 12V power supply without instantly overloading = frying them

@MAB : I remember about fitting a 5mm LED into a technic brick's hole, but it was so tight that the LED's body could not be inserted on its full length without risking breaking the LEGO brick or rendering this insertion permanent. @phaenius : - if you intend to use other power sources (perhaps LEGO ones through the use of a cut LEGO power cable), never connect LEDs directly to them or they'll burn out instantly. Limiting resistors or constant current source modules will be required - I'm indeed curious about the whole lighting kit (either its specs, or reverse-engineering it). And for the concern of it being considered as advertising, perhaps the problem may arise if its brand is a LEGO clone (some are legal and can be discussed here, some others are not and we must not talk about them, for not advertising them), or not : I guess that universal non-LEGO-related lighting kits can be talked about here (after all, some other people do add custom LED lighting to their designs). Instead of pictures, the name of the lighting kit or a link to its supplier's and/or manufacturer's page is possible too.

Cutting the strip can be more or less "reversed" by re-soldering the cut copper pads together (through a short wire). I noticed on the pictures that this LED strip is very unusual : power LEDs, probably between 20mA and 100mA, slightly smaller than the 5630 type. Technical specs for the LED strip (circuit), the LEDs and even the power supply (constant voltage or constant current type ? For any length of this LED strip ?...) would help to solve the numerous problems. Moreover they seem to be wired directly in parallel, without individual resistors (unless the LED case embeds it, but it would be much more unusual). And having LEDs wired in parallel is usually bad because the current sharing isn't well controlled (and when one LED gets disconnected, then the other ones receive more current, then may fail in cascade because they would get overloaded more and more). Personally, because of the above remark, I don't trust very much this LED strip. In order to gain full control of the current in each LED of it, more advanced hacks would be needed (cutting the strip in single LED bits, powering each of them with individual series resistors, whose values would have to be chosen, depending on the LED forward voltage and the supply voltage... like when designing a LED circuit from scratch). Have you got any idea about the kind of lighting your LEGO designs would need, and if strong miniaturization is a concern, as well than aesthetics (hiding the wiring) ? Some of these informations would help choosing the right lighting solution : - a "long" (several cm) row of LEDs, above/around a stage for example ==> a standard 12V LED strip where the "multiple of 3" constraint won't matter so much (you can center the cut strip), especially if the LEDs are close to each other. - shorter rows of LEDs - LEDs arranged in an array - single LED, either direct lighting or remote lighting (light pipe)

You can post remotely-hosted pictures ([insert other media] - [insert image from URL]). No restriction to some specific picture hosting services : if you own some website-hosting server space, you can put your pictures there and link to them too. Being in electronics, I can say that, on 12V strips (the most common ones), for electrical reasons (using less power while staying not much sensitive to slight input voltage variations), LEDs are arranged in series of 3 with one current-limiting resistor. Therefore one can only have multiples of 3 LEDs on a cut strip. Then I guess that, when wanting other numbers of LEDs, one has to either hide the unused ones, or modify the strip itself (changing the resistors values, re-wiring some PCB traces...)... or even design one's own lighting solution, for example, when only one LED is needed (there are plenty of online tutorials, but if you need further help, you can still ask me).

Here is it : here is an overview of the core mechanism mounted on a very basic scaffold with a few unfinished test elements : a basic ball ramp (bottom right) the currently tested ball holder (bottom left) + a static structure (right) for pushing its gray 'L'-shaped releasing lever Below is a partial dismantling of the mechanism : on the left, the "stator" + the rotating eccentric (rotation center = lowest round hole of the 28-teeth gear). The current structure has been built with my currently available parts but will be made slightly simpler as soon as I get the right parts on the right, the "rotor" = the cycloidal disc, revealing two main design tricks (guess them !) Since this mechanism is smaller, I currently encounter difficulties in reliably loading/releasing balls passively (like on Akiyuki's design, no "active" mechanism for pushing the balls into or out of the holders) : the input ramp has to be placed very precisely, so that the ball gets fully pushed into the loader, and the ramp must not interfere with the holder's cycloidal path. I guess that the current eccentric radius (1 stud) does not give a wide enough move, but since this configuration can work for now, I'll increase the eccentric radius later (because it involves readjusting the whole cycloidal gearing in order to keep it smooth as now !). The places where balls get loaded and released are also different, because of how they get held (first brainstormed idea I had).

Hi everybody. This is my first post (except the "members' introduction"). Not being sure if it is the correct way/place to start I at last go into posting about my GBC projects, which are all only at WIP stage now. I already watched and read a lot of GBC resources (events videos, standards, famous modules and builders, balls sourcing hints...), which gave me a few ideas of custom builds, which will be new ideas as well as "remakes". But my work is currently slowed down by a few difficulties when leaving the dark age (getting used to studless building techniques when designing custom only, missing parts despite a huge initial BrickLink investment), so the results would show up bit by bit only. So, for more concrete stuff, my most advanced work is on a compact cycloidal drive, of course inspired by Akiyuki's one, but being a fully custom from-scratch and trial-and-error build. At its current state, only the core mechanism is available and it runs smoothly. However the ball transport (currently built on only one cycloidal "tooth") seems to be the hardest thing to make reliable, and it is not yet the case. A stiffer chassis (with the input bin and agitator + secondary ball transport mechanism will be required too. I'll build it with bricks mainly (when I'll find an opportunity to pick up my childhood stock). The core mechanism features an outer ring of 8 teeth, based on an octagon made up of 6L axles and 135° connectors, and a 7-teeth cycloidal disc (and some associated tricks), resulting in approximately 2/3 the size of Akiyuki's original design. So is my prototype worth a picture right now ? (a few other modules are planned but I built only draft/mock-ups of the core mechanisms now due to the lack of parts.)

mathieulego, Falandrin, Merci pour l'accueil / Thanks for the welcoming... Something I forgot to tell in my introduction about the interest of using LEGO : creativity comes from technical limits, the same way as when coding demos or impressive games on vintage computers.

Hi everybody. I'm Thierry, from France, currently 45 years old. Back into LEGO since a few weeks only, due to a combination of recent events. Back, because I of course played LEGO a lot when I was a child/teen : - especially with the Technic series since I was 6 years old (I always loved using gears because they impressed me when I discovered them when I was 3 years old) : most often for trying to build things by myself, after building the official models. - later, with the early 80's Space series In general, I was always passionate about mechanics first (clocks, electrical toys...) then electronics, and later computer science (which make my current job : electronic board design, sometimes with embedded software). Some stringent limitations with earlier Technic LEGO parts made me stop using them, especially the inability to rotate gears independently on a same axle (this was before the differential bodies, and the Samsonite gears were totally unknown to me !). For me, the so-named dark age didn't mean forgetting about my special interest in mechanics, playing with non-LEGO parts, either scavenged or, in the beginning of the decade, custom made thanks to 3D-printing. Then, in May 2019, on a makers exhibition, I met PG52 https://www.eurobricks.com/forum/index.php?/profile/98352-pg52/ and saw his masterpiece (a detachable chairlift) and also a few GBC modules he assembled or sometimes designed (I already knew about GBC because I already found this by chance several years ago on YouTube and spent a few evenings watching it). After that, watching a lot of GBC-related videos and discovering impressive mechanisms (of course some of them from Akiyuki, especially the Strain Wave gear, but not exclusively), I felt a strong desire to start building LEGO machines again, especially as design limits were reduced, thanks to new parts (for example, geared turntables, gears with round holes) and new LEGO build paradigms (studless, SNOT...). That's why I'm here. Currently putting money in "new generation" spare parts, having ideas for building variations of known GBC modules at a lower scale, a few not yet implemented ideas... perhaps later exploring the building of clocks too... I don't know for how long this delight will last and I have not the will/motivation (and the time and money) to reach the same level as the big fans I admire, but I want to take as much pleasure as I can in playing LEGO again.