Designing The Marble Machine's Programming System

[00:00:00]I'm going to make programming profiles for this giant mockup programming wheel.

[00:00:04]So, something that sticks out from this wheel and I'm 3D printing something right now. Kind of exciting. Okay, so the idea is that this profile can snap into the programming surface. So, this is printed in one direction and thanks to these L holes, you can press it in and then slide it to lock and then slide it forward and you remove it. So you go up to the wheel like this and you click it in and you click it out. The Mar machine has this giant programming wheel and today I want to research the question, is there a nice design concept for how to make music programming on a giant wheel like this? The programming pins that we used previously can only play a note where there is a hole. The profiles can snap into the grid and the length of the profiles can be changed to play any rhythm we want. The pins can only play here or here. The profile can also play here or here or here. The music is not restricted to the grid, but we can play any rhythm we want. And that's a game changer. If we design a clip that goes into here, the profile is mechanically locked and it can't leave the wheel. You remove the clip, then you can remove the profile. The idea is that this long arm will lock the whole profile. So, this gray looks like the original Marine gray. Let's use that.

[00:01:29]Okay.

[00:01:33]I'm not super happy with this. I want to try a different version of this task.

[00:01:37]So, we have version three here.

[00:01:42]It's a better print.

[00:01:44]The principle is super nice. This can never go.

[00:01:49]It won't go anywhere. You can let's change the design of the task.

[00:01:55]Task is short. The surface interacting with the readers is solid. So here was a long task. Here's version four with a short task.

[00:02:04]Yeah, it's pretty good actually.

[00:02:08]So then you go with your finger here and you take it out.

[00:02:16]This cannot go anywhere. It's locked.

[00:02:20]and you take it out. I really like this one over this one, but there is an issue with this. So, currently the profile is 30 mm high, but because of this elephant lock here, we're kind of wasting the first part of the profile. The reader can't work with the first 10 mm. It would be nice to save material and make the profiles as low as possible. On this design, the reader can work with entire height. I am uh 55% satisfied with this.

[00:02:50]Okay, so we started talk about print in place idea. So Ludvig made a battery holder here and this spring is printed in place. And I've been thinking about printing in place, but I forgot that you can put the spring on the back side. So after iterating on our ideas, I think we're going to try a print in place solution. Ludvig, I think we have version five ready. So I wasn't happy with this. And with this print in place idea, I really like this design. So if we turn it around, boom, things are happening secretly here on the back side. This one had to click in front of the profile. Now we just have a clean thing. So no mechanics in front of the profile there. Instead, we have hidden it in the middle here and we're printing over. So the whole profile is still stiff. If we go layer by layer, here we're printing the spring on the print bed and then we're printing over the spring with new layers that is not touching. This might be the one.

[00:03:50]Ooh.

[00:03:52]Okay. Testing the print in place.

[00:03:55]Yeah, the idea is brilliant. My spring, you unlock it here. My spring is too weak. I can shorten the spring. Version six, new idea. Print in place track here. Uh pin moves manually up and down.

[00:04:11]No spring and it's covered on the back here for stiffness like this. Very simple. So, um let's print it.

[00:04:19]Programming profile seems to work great.

[00:04:21]And now I want to make a huge programming section on the big mockup wheel here. To do that, we have to learn about K factor, which is something that I wanted to learn for a long time. We're going to learn it together now. Imagine this is stainless steel 1 mm thick. The curved base is going to be manufactured flat by water cutting or by laser cutting like this. Which means that we need to have a flat pattern. Our flat pattern will then curve when we put it on the wheel. And when the flat pattern is curved, it's going to be distorted.

[00:04:51]The K factor can helps us calculate that distortion. I can show you the little thing that I know. If we think of this as a 1 mm uh sheet when this straight piece is bending there is something called a neutral axis I think and the K factor is the distance from the inside to that axis. K factor is always smaller than 0.5 because 0.5 would be the exact middle of the material. This axis never lies on the outside. It's always a little bit on the inside of the band. So if we know that the K factor is 0 something, we can then calculate how a flat pattern translates to a curved pattern. And lucky for me, on shape does these calculations automatically in the sheet metal sections that I also need to learn. So we're going to do both K factor and sheet metal section. Now this is quarter of the wheel and I picked a section up here that is small enough to print, but we want to manufacture this flat and then curve it. So that's where on shapes sheet metal model should help.

[00:05:58]I tried to use the convert which didn't work but apparently for this kind of shape thicken and then I choose the top face here. So then we have a flat view and voila. So we're going to use the stainless steel K factor. This is plastic but we we we don't care. If I choose an edge here you can see the same edge lighting up here in the flat part.

[00:06:20]And this is a radius of 800 mm. But in here it has calculated the length. This is what we should print somehow. So if I open a sketch here, I have insert the XF. Okay. So we extrude that 1 mm.

[00:06:37]Boom. And we are ready to print. My first sheet metal work in on shape. So I'll import it into the printer and we're ready to print.

[00:06:58]locking them down. Click. Click. I love that design. So now I'm using the same holes, but we are ending the note a tad later. And we can choose to end this wherever we want in the musical timing.

[00:07:12]So, the profiles gives us infinite resolution on musical timing, which is a huge advantage over the programming pins. Here's a little bit of a longer one. I'm using the same feet here. Boom.

[00:07:26]Locking it down. Click. Now, we have three different endings. We are not limited to the grid. We can make any rhythm we want. Let's put them on the machine. Okay. So, I want it to be kind of here. And I need something to hold it to the correct curve here. So, let's make some plywood supports on the CNC machine. So, we have the Watson M1325 here. And we have Ludvig who has helped me set this up. We're going to cut some plywood.

[00:08:01]All right. Plywood holders. They're going to go on the inside like this.

[00:08:05]This programming profiles hugs the underside and pulls the sheet metal into this correct curve.

[00:08:14]This is only a mockup. We're not going to 3D print this. It's going to be stainless steel. Let's just remove this.

[00:08:33]This is not for touring. This is for composing. I click in, I have a note. I can click somewhere else. Boom, I have another note. We wouldn't do this on tour at all. The music has to be pre-programmed. But when I'm in the studio to be able to manually program like this, okay, I think this note should be over here. Or no, it should be a shorter note that plays faster. Let's switch this one out to a shorter one.

[00:09:01]Yeah, that that'll sound good. So now when when I have this mockup, I feel it on the machine. I can feel how it feels standing on the machine making music here, which informs the upcoming design decisions a lot. There's absolutely no use in me furthering this design until I've written all the design requirements for the programming wheel. If you don't understand the problem you're solving for, you have no actual way of measuring how successful a solution is. So this can never be anything else but a design concept. To get going with this, I need to properly write all the design requirements for the entire wheel. I also realized that these holes will be even better if we just make them square. There's no reason for them to be L-shaped. The machining time will go down, I think, and it will look a little bit cleaner and the profiles will be more sturdier. I printed this to explain something really important. There's two kinds of independence that you really need to understand. First, we talked already about length independence. So, this profile can end here. It can end here. It can end there. It can end wherever. The length of this profile is independent of the programming feet. It's not a programming pin that sticks up right where the hole is. No, it's it has lengthwise independence. But check this out. The fish skeleton, it has heightwise independence as well. If you want to play two notes really quickly, you can. If you want to have the highest opening a little bit, you can. The top of the profile is a canvas. We can paint whatever painting we want on. We can also have different height. Here's one height. Here's another one. But it's extremely important to understand that these are not pins that are tied to the whole locations. We can do whatever music information we want on top of these. So boom, we have height information. We can do whatever we want.

[00:10:55]And boom. And so if I was a reader, the system is really versatile. It's true CAN discs. People have been worried about if we can repeat the same channel fast enough with these profiles, and it's an important thing to keep track on, and the answer is that we will be able to do that. The programming profile is inspired by all the 13 designers who sent me design suggestions in on shapes and I took inspirations from Jackim's snap-in hole shaped. Julian sent fantastic designs and this one the middle part can be 2D manufactured from any kind of material. Sumona had the most creative print direction to print the print standing up. Sam made vintan shaped snap fits. Here's a design from Fabricatis strobe waffle, the Dutch delicacy. This nice material saving design. Really nice design with a clip in the back and these perpendicular feet. This design from Ben Taylor showcases something really important. We can use a 2D manufactured plate. So, we could use really great plastics like delin and stuff and water or laser cut them. And then we can use a clip from another material. I think these two material solutions are really interesting for the touring version.

[00:12:15]Axel sent this nice hexagon hole shapes and as we know hexagons are the best.

[00:12:21]Ludvig suggested to buy this offthe-shelf sheet metal that is produced with these holes as a curved surface. Seeing all these great designs is so inspiring and thank you to all the designers for helping me. Without you, I wouldn't have been able to do this and the same goes for all the backers supporting the channel. Thank you so much. Do we have an answer to the concept question? Yes, I think this actually works surprisingly well. I can see this monstrous programming wheel working super well actually with some concept like this. And for me personally answering this question if I can have a large programming wheel like this is the last research question of the Mar machine M machine. So the Marm machine 3 has found its identity from here. It's discipline, hard work, and elbow grease that is needed to make progress. And those are actually simpler than artistic inspiration. So even though it's super fun tinkering and researching a little bit here and there, I need to take a more systematic approach going forward and starting with planning and design requirements and then actually designing this beast to make it happen because I do not think the project is progressing at a great speed.

[00:13:28]>> Yeah, it's a it's a great concept.

[00:13:30]Actually, I'm very happy with the concept. I'm not so happy with the progress I've been doing on the M machine this spring. I need to work faster. If I ever going to get to the music and to the world tour, I need to speed up. I'm a fast biker. I need to become a faster builder. So, I need to go much faster towards the world tour.

[00:13:49]I'm going to speed up like 100 times. I need to speed up 100 times. And in the upcoming four weeks, I'm on parental leave. So, I'm going to spend every day with my daughter going swimming and stuff. And those four weeks will give me time to reflect on how I can make progress much faster because honestly I'm not super happy with the rate of progress. Also, before having kids, I worked 16 hours per day for 7 days per week. Seriously, my whole life. So, I went from working 250% to now only working 100%. With less time available, I have to make sure that I use it properly to make progress faster. We can, we should, and we're going to go faster. I hope. Anyway, there will be no content during my parental leave. I'm going to be really mindful spending time with my daughter and I wish you all a great continuation of the summer and we'll see you after my parental leave and I'm super stoked to get back to this fantastic monster music box. Take care.

[00:14:41]Ciao.

[00:14:52]Heat. Heat.

[00:15:02]Heat. Heat.

[00:15:54]Heat.