ord

An almost perfect fit that is new for me: The width of three stacked beams (red dimension) is 8 x 3 = 24mm. According to The Unofficial Lego Technic Builder's Guide, beams are 7mm wide on the thin side, which would make the sides of the yellow triangle (2.5 x 8) - (7 / 2) = 16.5 and the hypotenuse sqrt(16.5² + 16.5²) = 23.33mm. 0.67mm undersize might seem like a lot, but in reality the three stacked beams are held rather snugly at 45°. This can be tested easily inside a 5x7 frame. Edit: like this...

Again, I learnt from the great Akiyuki: I was looking into this last night. The train wheels don't stand up on their own so I think they need a carrier ring connecting them together. Maybe only four wheels could fit on such a ring, which might compromise performance. Also, the 1x1 tiles do a surprisingly good job at providing radial support, but the train wheels not as much, so the bearing might require extra support if using the train wheels. I will build one and see...

I built a bearing based on Akiyuki's design but with the smaller banana gears and it works well: I even managed to pass two axles through it. I think I'll be incorporating this into my robot arm as it seems very rigid.

It's an interesting problem @sammypants and one I have been scratching my head at too. Having built Akiyuki's design (first video), I think it's possibly the best solution - it runs smoothly and is the most rigid design that I've built. The design in the second video is easier to build and remains in-grid, but doesn't have the upper gear form-locked below the lower gear, so is less rigid. I'm not sure how either design would go with the smaller banana gears, but would be interested to see!

Making a video is too much work so I don't plan to do that. Yes the subject is complicated but all of the information is available if you follow the installation instructions on the vpype github - this is how I learnt how to do it. The mechanics of my plotter are fundamentally pretty simple. It's the coding of complex plots, I would say, that make it a complex system.

Once you have vpype and vpype-gcode installed (you will need Python installed first to install them), copy the custom profile above into ~/.vpype.toml (open this file in a text editor and paste the custom profile at the bottom). From there it's just a matter of saving your svg file as input.svg in the C:\ drive and running the "vpype read..." code above in vpype. Feel free to share photos/dimensions of your plotter. The only thing that might change for your plotter should be the size of the paper (mine is 297x210cm).

Ah, I misunderstood. Please see the first page of this thread for where I got my source images from and how I converted them to machine code. Basically, for a line-drawing, a vector file is usually needed that contains the lines to be drawn (I have been using the .svg file type). That file needs to be converted to code, which I do using software called vpype (and a plugin for it called vpype-gcode). This is all detailed on the first page of this thread, including the custom profile I used in vpype-gcode to output code that is readable by Pybricks :).

@lego aram it might look complicated but it's actually quite simple. If you copy some of the examples from the bottom of this page onto your hub you should have a moving plotter, and from there it's just a matter of changing/adding lines of code. https://docs.pybricks.com/en/latest/pupdevices/motor.html Feel free to direct message me on Eurobricks if you get stuck :).

@lego aram I suggest that you install Pybricks on your technic hub (https://pybricks.com/learn/) and code it from a computer. By writing a sequence of run_target commands for your motors you can create steps for your plotter to move to.

Putting a few of my WIPs here in the hope it will motivate me to finally finish them: 6-axis robot arm Have built the gripper and joints 1, 4, 5, and 6. Pneumatic clocks Have built 2 that work in sync. Need to make them look nice then make a video. Internal gear hub This could be my favourite - it's a working 3-speed bicycle hub gear. It's difficult to show the internal workings on video so I'm thinking of making a 3D animation (which will also be difficult/time consuming).

@nico71 thanks for the explanation :). It's great to see ideas 'outside of the box' like this!

Beautiful. Some interesting functions while still looking nice. Could you explain this a bit more please? I would think that the distance of the wheels to the centre of turning is irrelevant, as long as their axes pass through it? Does this type of steering exist on any real world vehicles?

Yesss. Great news! This should make controlling many motors much easier than with the Powered Up remote. Also the proportional control looks really nice.

Simply plotted some 10*10mm squares and measured them. Backlash is approximately how far undersized the squares are.

88010 remote control connected directly to the hub with the program loaded on it gives no noticeable lag for me.

Nice! I originally planned to build a CoreXY style plotter but figured out other ways to mount the motors remotely, so went with that. Hopefully we can see yours in action soon.

I agree, a pleasure to watch!

It's just made of liftarms connected together. In order for the cracks to not show up on the plot I used thicker than usual paper - 200gsm seemed to work well.

It also appears that YouTube promotes content with catchy thumbnails and that viewers are unlikely to skip through - both of which might be seen unfavourably by the algorithm for Sariel. It's unfortunate because he has some quality content, and I often refer to his reviews for technical details of sets/pieces.

Indeed, it's a pretty nice sensor! I had initially thought it was just a touch sensor. I wonder what other novel uses it may have... 🤔

Apologies for the late reply. I don't think so. I think a spring just creates a constant opposing force that does nothing to change the friction that must be overcome. The internal friction of the cylinder is what I think is the problem. True, I have changed to a large cylinder (levered onto the force sensor) and it seems slightly better. Still, there is a large drop in pressure before a change is detected. This is the best I could come up with: Interesting, thanks for sharing!

Hm, it looks awfully similar to my plotter design: This is basically how I did it @Knight3, except with a Robot Inventor hub rather than EV3. There is a lot of detail in the WIP thread here (including the python program I wrote in pybricks): I highly recommend the command-line tool "vpype" and a plugin for it called "vpype-gcode". They are purpose built for plotter vector graphics, and with them you can convert any .svg file into gcode, which you then just paste into a program and run it (feel free to use/copy mine).

Wow, so good! Love the camera angle looking at the side of the loop.

Sadly, things are not this simple. I believe the internal friction of Lego cylinders prevents accurate readings, at least in both directions. With this setup, while increasing the pressure, I could obtain fairly accurate readings. However, once I then decreased the pressure, it would take a drop of about 0.8 bar before any change was detected by the force sensor. I think this is simply due to the static pressure (EDIT: static friction) that has to be overcome to move the cylinder. I also tried using a battery box to gauge the pressure. It works, in an on-off fashion, but needs a similar drop in pressure before it falls back down (which isn't necessarily a bad thing for a simple compressor). My hopes of building a fine-tunable compressor are dashed. I probably won't revisit the OP concept, as I don't like the idea of constantly having to run a motor to check the pressure. At least, the SPIKE force sensor can be used as an on-off pressure switch (with on and off points ~0.8bar apart), so this is what I'll use going forward.

Thanks for sharing. It's a smart solution you have come up with to operate many cylinders with just 2 motors. Someone has used the battery weight mechanically with the old switch: https://rebrickable.com/mocs/MOC-113088/olivierz/pneumatic-pressure-controller/#details I don't think it's crazy to use the gyro. In fact, I think it could give accurate readings. The downside I think is that it would always have to be on a flat surface (due to gravity), whereas a force sensor could be put anywhere.