 Boom. What's up everyone? Welcome to Simulation. I'm your host, Stalin Saki. I'm very excited to continue being doing interviews at COFES, the Congress on the Future of Engineering software, for our second annual partnership with them. We are now sitting down with Tom Dobroth. Hello. Hi there. My name is Tom Dobroth. I'm founder of 21geo. Thanks for coming on to the show and talking to us. Really appreciate it. Thank you. Very excited to be learning more about 21geo and what you're building. You're actually doing one of the presentations right now. That's correct. Yeah, and so you're explaining, you were explaining to me earlier about your interest in engineering and building out what is a bearing that is 21geo's technology enables bearings to do more than go around in a circle. They now support complex motion. So the way this works is it's a computer system, some high-end math, computer C++. You give two arrays of two sets of arrays, x, y, and rotation, one set for each component. So the components are going to move relative to each other and you define that. And then what my software does is it provides a bearing that will support that motion. Interesting. Great deal of flexibility. This is a brand new mechanical paradigm on par with gears, cams, and linkages. It has more flexibility than anything else in mechanical engineering today. Interesting. It has more flexibility than anything else in mechanical engineering today. So then your 21geo's software would enable someone that puts in their, like you said, the variables that they would need a bearing to do more than just have something spin around it. And you then go and make and model out how to make the bearing fit their needs. Yes. And then so what would a bearing like this one be used for? Okay, that particular one is half of a speed reducer. A speed reducer. Okay, so if you go on my website, you can see a number of, you'll see some speed reducers on the front page. And the link is 21geo. 21geometry.com. 21geometry.com. The link's in the bio everyone, uh-huh. Okay. Bye. And it would have, I haven't built one yet. This is, we've only been public since November, public with the information. Yeah. I'm seeking to get it produced. I'm seeking for a partner to actually finish the production, get it into and license it. Yeah. But it would have best-in-class features for robotics and motion control applications. Yeah. We need speed reducers in those cases because servo motors want to go really fast and robotic arms don't go anywhere near that fast. Okay, servo motors go fast like 2,000 rpm and they need to be slowed down at times because this is probably 20 rpm or yeah 60 rpm. Sure, 20 or 60 rpm, yeah. And when it's going like that, it needs to be slowed down. Right. And this how does this help it slow down? Well, you put two of them with a different combination on there and it gives a difference. It's a little complicated. You need to kind of go to the website and take a look at that. Take a look at it and how it slows it down. Interesting, because what's the normal process for slowing it down? Use a gearbox. A gearbox is the normal process, yeah. Okay, yeah, yeah. Okay, and that uses a lot of energy. It also has lash, meaning that there's some sloppiness in it and so the things move back and forth and it makes the controllers bad. Now there's some specialty gearboxes with very low lash and on mine I can actually, because my system has no design friction in it, I get, you know, these pins roll freely throughout the motion here. Yeah, yeah. So there's three carry forwards from a regular bearing that I give it that are necessary. First of all is that there's always contact between this pin here and the two surfaces. Yeah. The second carry forward is that it rolls freely. So this pin is rolling just like it would in a ball bearing. And then third is, and this is a little less obvious, is that these contact points are always opposing each other. Huh. They're 180 degrees apart always throughout the range of motion and that keeps it from jamming. Oh, interesting. Okay. Their contact points are 180 degrees apart. Apart. Aha, they're on opposite ends of the pin. Oh, yeah, yeah, okay, okay. Interesting. It's also cool how this is more like a triangle shape and this is spherical shape and then that's how these two things interplay. Yeah. Yeah, that's very interesting. And it's a question of, the combination is plus one. So you have three, three sides, four pins, that's plus one and then this one is five. So that's a plus one again. Five. Oh, it is plus one. I see now. Yeah, yeah, yeah, okay. Well, that's interesting, yeah. But this is one of a thousand ways you can use this technology. Yeah, correct. Okay, so I'm in discussion with an automotive company that they're redesigning their cockpits for lay flat. For lay flat. Excuse me. They're redesigning their cockpits for autonomous driving. Yeah. And they want a lay flat seat. Lay flat seats, yeah. And anyone who's been in business class will tell you it's pretty common that when the seat goes down, they're not flat. Yeah, yeah. Okay, and this is even with the expensive seats you have in business classes, it's hard to do. This is easy for me. So what I would do is just program it, index it with the rotation. So as it rotates down, at some point it starts to move out and then at another point it starts to move down so you get a lay flat out of it. Yeah, yeah, interesting. It's a problem I'm working on right now. That's a great problem to solve. Yeah. That's interesting. That's the big thing is that the seats normally like this and that when it lays flat that there's not, you haven't moved the seat back as well as flattened it out. Right, yeah. And so the types of technology like yours can enable processes like lay flat seats and how would, interesting, how would your tech enable it to move back as well as lay flat? Well, you program into it how you want the components to move. And so once you get that program then you have to just, you know, then I have to sort of move the pin starting position around so that I can find a set of motions that will work for it. Okay. You have to make sure that there's interference problems you have to work with and there's, you also want to make sure that it's fully supported. Yeah. Yeah. Okay. And the other industry applications? Yes. Yeah. So I have a grant application out for the National Science Foundation. Cool. And the nature of the grant is to, I have on my website you can see what's called precise motion technology. And from that you will see crankcase replacements. And the crankcase, the idea is if you have a piston system you want to precisely determine how the piston is going to go through the cycle. And with a crankcase you don't get to decide that. The crankcase does it for you. And that's not maximum efficiency. So for instance we know that 132 years ago there was a patent for the Atkinson cycle engine. Vastly more efficient engine. The Toyota Corporation did an experimental engine in 2014 and got the 38% thermal efficiency, which is fantastic considering most cars operate around 20% thermal efficiency. And it really comes down to just letting the cylinder go down a little further than it does on a compression and harvesting that energy. Yeah. Okay. So what you'll see on the website is because of this very new technology I can develop an engine crankcase replacement so it'll be vastly more efficient. It'll be cheaper and I get the cost down because I can do the power stroke. I do the whole thing at 360 degrees where it currently takes 720 degrees, one revolution. And I reserve 180 degrees for the power stroke. And since it's 180 degrees I only need two cylinders whereas the minimum internal combustion engine right now is four. So the cost is going down. Yeah. Then finally I can program into it the mechanical advantage so that I can have a constant torque output. And that will be pretty revolutionary if I can pull this off. Yeah. And there's some shortcomings to it. And there's no doubt about it. And that's what the grant application from the National Science Foundation is about, is to take that to the next level. This is so interesting that how did these ideas come to mind? This is so cool. You know, I wound up out of work and I was trying to figure out, you know, what can I do on my own? And so I looked at these. This fascinated me. I saw animations like this in college. I have two engineering degrees from MIT. And so I said, okay, let me look at this. And so I went online and said, you know, how come these things don't exist in the wild? Why aren't they being used? And what I came up with is there was a number of reasons that they just wouldn't work. Number one is that they would jam. Number two is that the friction was, no friction was just an assumption. Just because you have a rolling system between two surfaces doesn't mean that it's going to roll price precisely. Yeah. There's a patent that goes back to 1920 that looks a lot like that. Interesting. Okay, but the rolling on that is an assertion. And if you look at their own diagrams, you'll see that the pin is being held like this. And if the pin is being held like that between two surfaces, any motion out of the way, it's going to jam. Yeah, yeah, yeah. And then this new kind of what looks like a little like tank treads almost. Yeah, I call them guides because what it's... Guides, guides, that's cool. It's a good way to put it. They're guides because you're fitting the guide from the center of the bearing to these pins. Yeah, and if it gets out of place, the thing doesn't work. So the final intent of the guide is to keep it very near where it needs to be. Yeah, correct, correct. And it won't move. That's one of the keys is that it won't move unless the guide is perfectly in place. And the four pins hold it in place quite well. This is very, very, very smooth. And so then what would be an application of one like this? That device in particular? That one, like I said before, is half a speed reducer. Yeah, half a speed reducer. And then the other one was the robotic arm needed to move faster, is that the case? Well, that's the reason we use a speed reducer is because the servo motor wants to go really, really fast and we need to slow down the rotation. So we're talking with a robotics company and they want a hundred to one reduction in speed. And that's huge to be able to handle. How does one make a hundred to one reduction in speed in such a short amount of time? Well, you take something like this and you put, instead of four pins, you put in 15. Yeah. And then you make another one that has 14. And then what happens is you join the two rotors together, these two centerpieces. Interesting. And then the output is one of them is stationary and the other outer ring moves just a little teeny. So kind of like gears, but... It's kind of like gears, but it's a little different. Yeah. This is a much more complex center than a gear. Yeah. And again, that idea was tried a hundred years ago. Very interesting that it was tried a hundred years ago. But the surfaces, there just wasn't the math back then. You know, this is high-end matrix math putting this all together and optimizing this shape so that it rolls properly and we're keeping the contact points under 80 degrees apart. Yeah. This is very interesting. I think this can be used in a lot of applications in robotics. I have a lot to, obviously, learn about engineering and its applications, but I know that lots of kids, as they build robots for first robotics and whatnot, are always interested in how to apply the new parts to maximize what they're doing. And if they can, like you said, decrease the RPM of a servo, that's huge. Yeah. Yeah. So the other applications are just in general, if you were going to go out and you wanted to build something with a mechanism. So if you were to use a four-bar linkage for something, you know, that works good. If there's only two points you're trying to make, that works just fine. But if you start to say, hey, I've got to do multiple parts, you know, multiple movements in this thing, and now you're getting into a six, seven, eight-bar linkage, this is going to do a much better job for you. Yeah. Yeah, yeah. Okay, so the more complexity this does a better job. Yeah. Yeah. You know, if you're doing a landing gear and all you care about is up and down. Yeah, yeah. Those are two points. Yeah. The four-bar linkage works great. Yeah, yeah. There's no reason to drag my stuff in there. Correct. But if you're doing something more complicated, then there's a lot of value to that. And so another example would be drug injection. So you can, if you have like an epi-pen or something like that, when the needle pierces the skin, it has to be going perfectly straight. Yes. And it can't be rotating. Yeah. Okay, and it can't be moving sideways. Yeah. Okay. So this is a problem that would work. If you wanted a four-bar linkage, you could replace a couple of the normal bearings in a four-bar linkage with some of my bearings. And you could have a system where the bar comes down and then it comes completely straight while it's doing the injection. Interesting. Okay, and there's some ways to do that using a linkage with the symmetry. But then you've used up all the degrees of freedom with just the one part. Interesting. So there's applications of this in medicine and biotech health care. Oh, yeah. There's interesting. Anything that moves, if it's complicated, that's it, yeah, if it's complicated. The other one is my website. I have an example of a corkscrew. Yeah. Okay, so there's two things, two takeaways from that corkscrew examples. First of all is that I can vary the mechanical advantage any way I want. So in that case, I have the mechanical advantage starting out high and going low because the forces start high and go low. So it's a concept that the mechanical advantage should follow the force. Secondly, you know, at home I have the expensive Williamson Minoma cork remover. And it's a rack and pinion, and that's a constant force system so that it starts out really hard to move. But it also has driven the gearing there. First of all, it has to be well lubricated to work at all. And it's got a very clunky feel to it. So with yet to prove it out, but I'm pretty sure with some decent material science and some interesting work, we could have this have a silky feel to it. That's very interesting. Very different feel. Yeah, instead of that really hard pull-up at first, for a system to potentially move with a greater amount of smoothness. Yes. Yeah, that's another one of those... There's a certain really beautiful feeling to when something moves more smoothly, like when it can slow down more smoothly or when you can pull it up more smoothly. It will be a revolution in the way any mechanism works. It's going to be a revolution because it's going to have a very different touch than what people are used to. That's it. Yeah, that's really interesting. And it's beautifully. It sounds so beautiful. And this feels really interesting. It's as though I haven't... The applications of this type of technology have so many different applications. Yeah. I think there's thousands of ways this is going to work. I think somewhere there's going to be a design lab that's going to put a lot of energy into how do you make instruments work more... Anything that moves. So like the automobile company, we're talking about, hey, maybe I can get the look and feel of how the glove box opens and closes to change. That's right. So when you open the glove box, you get this beautiful feel to it. Yeah. The way that the vehicle seat moves as well. The seat moves, how it moves. Maybe now there's a table that pops out from somewhere and because of the flexibility of my system, that table will just jump up. Yeah, yeah. And it'll have a very interesting mechanism in it and it'll feel unique and then it'll pop out of the way. It'll be perfectly gone. Yeah, yeah. So those are the kind of things that... Interesting. Yeah. Yeah, this is very cool, Tom. We really wish that this gets implemented into the use cases that it needs to be in as soon as possible. And it's great to see you coming up with this after a long period of figuring out what are the most robust applications of this and coming up with this design and moving forward. So I'm looking for my first customer. So my goals this year are to establish a CAD partner, get a first customer, hopefully two, one for the speed reducer and one for some general mechanism work. And then finally, I've got this grant application out there and so I'll find out in a few days whether or not I get the grant. That's great, yeah. The grant application and then the applications, getting customers and CAD partners. So that's awesome. And hopefully people that watch can potentially come on board and help out as needed as well. So thank you very much for coming on and teaching us about this. This has been very interesting, Tom. Thank you. Thank you. Thank you. 21geometry.com. Check out the link in the bio below. Also, continue supporting the artists and entrepreneurs you believe in. Check out Co-Fest's links below. And build the future, everyone. Manifest your dreams into the world. Huge thank you for tuning in. Give us your comments and your thoughts below. Thank you and we'll see you soon. Bye.