University Place

Designing Electric Motorcycles

– Welcome everyone, to Wednesday Nite at the Lab. I’m Tom Zinnen and I work here at UW-Madison Biotechnology Center. I also work for UW-Extension Cooperative Extension. And on behalf of those folks on our other core organizers, Wisconsin Public Television, the Wisconsin Alumni Association, and the UW-Madison Science Alliance. Thanks again for coming to Wednesday Nite at the Lab. We do this every Wednesday night, 50 times a year. Tonight it’s my pleasure to introduce to you Lennon Rogers. He’s with the Grainger Makerspace here in the College of Engineering. He was born in Sterling, Illinois, which is 12 miles from Dixon, Illinois. And, we are rivals. (laughing) He went to school at Sterling, Illinois High School, which means he was a Golden Warrier. Then he went to University of Illinois and studied mechanical engineering for his undergrad. He went to MIT, two different bouts first to get his master’s degree in mechanical engineering. And then he went back to get his PhD in mechanical engineering. Spent two years in Switzerland. And then came back to MIT to work two years there. Then about a year and a half ago, he came to UW-Madison to help open the Grainger Makerspace, where he’s the director. Tonight He gets to talk with us. He’s got a shorter title, “Designing electric motorcycles for racing,” but I like the longer one, which is rare. “Designing electric motorcycles for the Isle of Man Tourist Trophy and the Pikes Peak International Hill Races the world’s most dangerous motorcycle races.” This should be cool. Please welcome please join me in welcoming Lennon Rodgers to Wednesday Nite at the Lab. (audience applauding)

– All right, so here’s a brief outline and I’ll start with how it all started, which is actually something that is kind of complicated to explain. And then I’ll get into the easier stuff which is the racing which follows more of an engineering process. So the Isle of Man, the Pikes Peak and then it evolved into some research that I did as part of the PhD related to Distance to Empty and then I’ll give you a brief overview what I’ve been working on a UW last year and a half. So the easiest way to explain I guess where it all started was that I was working as an engineer at the Jet Propulsion Lab in Pasadena, California in like 2006, 2007 and I had moved out there and I just had a motorcycle and I pretty much bicycled everywhere and so after biking around LA you can imagine for a couple of years I got smog-induced asthma. And so it kind of started making me think I had a little bit of a conflict, I would say, between my passion and motorcycles, and this issue that I had in this awareness of environmental impact. And so I went to the Cal Tech Shop in my evenings. And in 2008, I built this first electric motorcycle.

So I called it eMoto and the whole idea at the time, I think that everyone was talking about electric vehicles being expensive. And so I wanted to show that they don’t have to be expensive. And so I use all components that I sourced online, mainly from China. They were all new. And, it cost about $3,000. And so that was that was really fun. And, it was in Popular Mechanics in August of 2008. They did a little article on it, so it’s pretty neat. But that’s not really totally where it started. It probably started back even going back to Tom was saying in Sterling where I grew up, and I don’t know why but I just always love motorcycles And so that’s certainly where it started. And then when I turned 18, I bought my first motorcycle and then I’m a miracle to be alive. The amount of motorcycling that I’ve done. You know, all four corners of the earth, many different mountain ranges, the Himalayas, and the Andes, and a bunch of places and so it just, it’s something that I enjoy doing. So that’s kind of where it really started and then probably the merging of my passion for motorcycling engineering came when I started out actually as an art student and I was in my art class in my freshman year. And I was talking to my classmate and I was explaining how I’m wrestling with this motorcycle issue that I was having rebuilding a carburetor or something. And he’s like, he was basically surprised. And that was the first time I realized like, maybe I should be an engineer. And so that was that was probably the second phase of where it all started, but I guess I’m connecting back to the electric motorcycles really came in 2008.

And then when I built this motorcycle I decided that I was very interested in electric vehicles in general. And so then I went– I decided I want to go back for the PhD. ‘Cause at this time I already had lived out in Boston, got the masters and actually drove my motorcycle from Boston out to Los Angeles and was working as engineer, planned on working aerospace. This was really the change for me and so I decided I want to go back and work on the PhD. And it was a unique time with electric vehicles in general. This is the, the group of people that I got involved with, in a good sense, so it was a what was called the MIT electric vehicle team. It just started like a year or two before I came back in 2008, 2009. So we were a bunch of mainly guys, but some girls, in a shop, in the MIT– There’s a garage named, the MIT Museum, and we were just building electric cars, electric motorcycles. and one of the things that I was super interested in at the time was fast charging and so we started doing some research on that. We are we’re working with a company called A123 Systems, which is kind of gone now for the most part. But that was a start up from MIT. And we were using their selves, and we’re doing some battery modeling. And we’re also doing some technology demonstration. In some ways, feasibility and it was related to that Isle of Man, which I can get into later on. But this is one of the first projects that we worked on the fast charging and there was a whole paper that I ended up writing on it, so if you’re interested, you can check that out. But that kind of evolved in what we saw in that garage actually, one weekend we were watching videos and we saw the videos of Isle of Man.

It was the first time I’d ever heard of it. And if you– I usually tell people if you haven’t heard the Isle of Man, just do a YouTube search, and you’ll see the type of videos and things you’ll see. So we saw those videos if you’ve done it: the wheelies, the getting off the ground, and all that kind of stuff. And of course, our jaw dropped. And of course, now you hear a little bit of how it all started. Of course, that was a perfect next step for me. So, the Isle of Man actually– I’ll tell you a little bit about the history as we go. But that was that 2009 was that first year that they had the electric class. It was a really unique alignment there. And so I’ll tell you a little bit about the race. So the Isle of Man as a race has really been pushing engineering for over 100 years. It’s a super interesting history of motorcycle development. Here, just give you an example, this was the motorcycle in 1911, and then 100 years later, that’s the motorcycle that we built. And you can just see that evolution and really, you can track all motorcycle development or a lot of it to that race. So just to give you some data to kind of explain it and put some numbers behind it. So this is a plot that I put together: on the Y axis is the fastest average speed of the Isle of Man lap, one lap. And then the the X axis are the different years, so started back in 1907. And you can see again, there’s just a slow incremental increase in fastest average speed. And I have the books on it, super fascinating to read it. You can see that each of those data points.

Well, for the most part, not necessarily each, but as it in general, these trends are related to different technologies, pneumatic tires, different overhead cams and things like that. So it’s, it’s super interesting to see that and like I said, my initial statement was that really the Isle of Man captures a lot of motorcycle development over the last hundred years. So I’ll give you a quick video here of the Isle of Man course. It is a fly through that I put together using Google Earth. It’s a little island between Ireland and England. So it’s 37 miles and the rider starts in one direction and it goes out there’s a little mountain pass that it passes through. You go up over the mountain there. So that’s just to give you a brief geographical orientation of the race. And here’s a brief documentary. So this documentary came out actually right before we were going in 2011. It followed the first two years, 2009, 2010. Unfortunately we didn’t see this before we build our motorcycle. We saw right before we were going there. I’ll tell you a little bit more about it with time, but it’s a really good documentary. This is just the the preview from the two-hour documentary. (whining motor) I highly recommend it.

[Video] The Isle of Man TT, the most famous motorcycle racers in the world. For over 100 years, these hills and valleys have echoed to the howls of motorcycles, pushed to the limit in search of speed and glory. (unintelligible) Until June the 12th, 2009 the day the petrol engines fell silent, but the racing continued on electric bikes.

– Very rare that you have like a Hollywood-ish kind of engineering documentary. But this is one of them and it is actually very accurate in terms of following along with these teams and the struggles that they go through. I was mentioning before is again we hadn’t seen this, so in some ways you guys have more information in that than we had before going. Pretty much at GPS coordinates. And I’ve done a bunch of simulations based on those coordinates and some basic information. But even when we arrived on the Isle of Man and I’d never been there, our rider actually was from the Isle of Man and he picked me up and he took me around in his little European car and I was just like, “Oh, what did we get ourselves into,” kind of feeling when you start going around the course because it is is very curvy and they’re very narrow roads with the walls being very close. Okay, so that, like I said, that gives you a brief overview. And if you want to watch that documentary, it’s on YouTube. And I think you can get it various other places. It costs a couple dollars to rent it or to buy it.

But I also mentioned before that when we saw that documentary, it, really the first two years there was aReally, I guess, the first year there was a very particular type of motor that pretty much all the teams used. And it was really the main one that was available at the time. It was the one that we use, but and you’ll see in the documentary and we saw this as we were going. Every team’s motor blew up and so we’re going to this race and we’re like, oh, no, what did we do because we had the same motor and so part a lot of the stuff even that I talked about, that was really the thing that causes some issues. The only team that didn’t have the motor blow up was the team that had the guy that designed the motor, his team. So we’re proud to say that we, we got it to work and we were one of the few if you watch the documentary, you’ll see why. Okay, so I’m skipping over a lot of details, but like I mentioned before, it was a very unique time. It was a time where we started to try to– Oh, I didn’tI started finding a team. I mentioned the MIT team. And then I started trying to get sponsors and this kind of stuff was really a grassroots effort.

And through various connections, I was connected with BMW in Munich. And again, this unique time where BMW, Munich was willing to talk to me who’s built one electric motorcycle, but it was so new to them that it was that new. And so I went to Munich and gave a talk and I met all the people, all the people in racing, and it was super interesting, especially now that you know my background. I had to sign NDA and I saw all of their projects they’re working on. It was awesome. And pretty much at the end, they’re like, “Yeah, will commit to give you a rolling chassis.” It was there S100RR which was just coming out that year. So the timing was very good for that. And then similarly with A123, there was the professor Yet-Ming Chiang, who started that company and he just so happened to be a motorcycle enthusiast. And so that was another really good alignment. So they sponsored and donated batteries. We worked with them to make them custom to what we needed in terms of the size. So those are the two main industry partners that we had. I’ll just give you a brief overview, just fly through it in terms of the design process. It seemed very clean here and for the most part it was, but I’ll try to interject and explain some of the complexities mainly related to those motors that I talked about.

So early on, it was systems engineering, then I’ll go through a little bit of the subsystem designs and the prototyping, the incremental testing and then the racing, and model validation. So that the system engineering, those are the GPS coordinates pretty much that I mentioned at the beginning. That’s pretty much all we had and so we were trying to estimate how much energy we would need to get around the course. It seems, again, very easy now, but at the time we had no idea, was it five kilowatt hours was it six kilowatt hours, no clue. No, there was no information available online. There was no published papers. So really came down to our own calculations using those coordinates and some of the assumptions. And the second question was, “Will the batteries fit?” And so we knew teams have gone so we knew it was somewhat feasible, but those are things that we practically had to figure out ourselves. And then the second more systems engineering type effort was related to performance prediction, like okay, maybe you could have enough energy but how fast would you go? Would it be competitive and things like that? This is just some of the results from some of the simulations.

So the basic hypothesis that we had was that the battery energy we needed was in between a best and worst case scenario. So this plot just shows battery energy versus distance. And so these are some simulations that that we were in. And I guess the details really don’t matter, but the bottom line was that we had a basic idea of how much energy we thought we might need. So that’s what we built our battery pack around. I’ll get back to this plot at the end. And then we just had various simulations we made. So this is just one plot, power versus speed, and trying to understand the different trade-offs we could make in the design and how we would perform. So just kind of cruising on a prototyping and fabrication. We used a lot of prototyping tools available at MIT and it’s kind of neat. Now jumping, I talked about the end at UW. This is really what we want to enable students to be able to do, is just a prototype to build some of their designs. This is just some examples. This is the structure that house the batteries and the motor and so the on the lower right is that assembly on a jig and we were welding it. And then there’s some cardboard mockups, in the upper right. But there’s a lot of prototyping, using water jets, laser cutters, things like that. And the other thing is that we were taking a gas engine out of a motorcycle, so we tried to preserve the structural loops. So here’s the stock motorcycle S1000. And then you move the engine, that’s kind of the next picture. And the lower one is the one with all of our batteries and our motor on it.

And so we tried to preserve that that structural mechanical loop that’s kind of drawn in green there. And we had a couple things that made us lose some sleep. One part was that there is a part of the course where the motorcycles typically get airborne. And so there’s this thing called chain whip where the wheels start to slow down, and then it hits the pavement and then there’s a huge shock wave through the system and a lot of motorcycles have little thing features to help with that. We are building our own, so we we had to figure that out ourselves. And so that was what– This is a finite element analysis there showing the chain whip. I’ll show you in the video. We did get air, not much air, but we did get air at that spot and the guy that on our team that had done all these simulations, it was kind of, he wantedHe wanted to get a picture at that spot, so he’s very proud of this picture, which I’ll show you, because he designed this thing and he said, “It’s not going to fail at that point” and he was right. We had a lot of sensors and microcontrollers for real time monitoring. That was one of the unique things in our approach, in general. The racing of the Isle of Man or racing in general, is it’s not a very open culture. A lot of academics don’t really go into that space. And in part of that is that they don’t share information and so a lot of the rules that you can’t have GPS, you can’t have sensors, you can’t be logging things and there were a lot of rules. We pretty much just ignored a lot of those until right before scrutineering, which is that process right before you’re going to go out on the course, and we just let them know like moments before.

Then we’re like, just so you know, we have a lot of that stuff they told us not to have. (laughing) We wanted to be honest and they’re like, “Oh, it’s fine, go ahead”. So for that, it’s kind of nice because we were the first people that I know have ever publish any information about the race in terms of this level of engineering detail. So, it was neat and I think it has to do with that the electric class was new. Again, that’s probably going to change in 10 years. Just to give you an idea of for some of the sensors that we had. We had a battery, the current the voltage state of charge, temperature, a lot of fault detection. The motors had current sensors and temperature sensors. The vehicle, we had speed and acceleration and GPS, and we had a display for the rider and it actually was wireless, but we could only read it within about a mile. And an interesting story about the display with the rider is that we initially planned on having some kind of fancy display that had all this kind of information, and when we when we showed the rider and was talking to him beforehand, he pretty much says he doesn’t want any information. He’s not going to look at the monitor, the the display at all, and he just said he’s he pretty much just wants to go full throttle the entire time. So that was kind of his riding strategy.

And even if you know electric vehicles, you can tune a lot of the stuff in the software. And we pretty much made it, so he could just go full throttle the entire time, because he didn’t want us to think about anything besides looking straight ahead. So we did have a display and it had some engineering features to a blinking LED for us to diagnose things, but it was very simple. The next thing that I want to talk about, just some of the incremental testing. This is something that I think all engineers really know about is that the cost of errors, which are showing the Y axis, really increases as you approach whatever, whatever you’re building this thing for, so in our case, it was the race. We tried to have a very iterative design process where we stayed down in this very low cost, spreadsheet sketches, low-cost prototypes phase, but at a certain point you find yourself in this scenario here, which is where your way up the curve and you’re not going to be able to make design changes at this point. It’s very difficult to re-weld things and make any changes. So this is the motorcycle on a dyno, chassis Dynamometer out in Western Mass. Or I guess it was just south of Boston. And this is just a show.

Usually when I talked about this with students, it’s kind of this as an example of why you want to stay down in this low cost, easy, everything’s fun, and simple phase of life. When you get to this phase, you want to have done all your homework and you don’t want to build make changes. And we didn’t have to make any changes again, besides this, this motor issue. So I’ll show you a brief video of some of the incremental testing and it really kind of goes through all the the races to the actual final race. One thing I didn’t know about racing is that they’re actually, there’s much more than just the race, the final race, there’s a lot of qualifying races. So it’s a bit of survival to actually make it to the race. Which is a lot different than like 5K runs that I’ve done, which is you show up and you run the race. And it’s like, yeah, I’m done. Actually, most of the effort is before the race and you’re exhausted. If you survive, you actually then get to race at the very end. The Pikes Peak was actually potentially worse in some ways. But that said, it’s something that we’re really proud of because we may have been the only team that year, or may have been one other one, that actually we finished everything. We finished the two qualifying races, and then then the third, actual, race.

And so that’s what’s shown in this video here towards the end. So these are batteries, just doing a lot of bench top testing. We spent a lot of time at the chassis Dynamometer. We had to make… (whining motor) alot of friends… with a chassis Dynamometer, with a Harley dealer in Boston. And this is in a hallway at MIT and everything was justNothing was welded that at that point. It was very low speed. And we put it in a wind tunnel to try and get some of that drap coefficient. (whining motor) And we took it to afor testing up at a track and New Hampshire. That was probably one of the funnest things about it was just all the people you meet along the way. We were spending time with… (whining motor and roaring engine) motorcyclists didn’t have track days and things like that. And they were super excited to see the technology, and super friendly. And I’d never raised on a track before. So the motorcycle was put in a crate and it was shippedactually in an airplane– to the Isle of Man. And because of various regulations (whining motor) we had to totally rebuilb it. We have to go back, test it on the Dynamometer again, do slow speed test, do high speed tests. (whining motor) And this is the point where we told him about the data and that we’ve broken the rules. (laughing) (crowd cheering) These are the qualifying races.

– [Man] All right good job.
– [Rider] Two in a row!
– Yeah.

– [Lennon] He’s saying two in a row because he was amazed that we, we had two qualifying races. So we had to rebuild the motors in between the second qualifying race and the final race. And it was a super intense discussion about risk mitigation among many other things of why we decided to do that. We didn’t have to, but in the end, I’m really glad we did. (whining motor) This is the starting line. There’s grandstands. But for the most part, you just don’t see the rider, he goes out in one direction. And then you’ve got an app that you can track him on. But we didn’t see this. This is from a camera crew, a TV crew, that was taking this. We just pretty much sit and wait. So there are various speed traps throughout the course. And it will then update the app and you can see. So obviously, everyone’s huddled in a tent, seeing the four or five different spots where they areOkay, this is the spot where they got air few millimeters. (laughing) (crowd cheering) Allen was the name of our rider and that’s him there. And this is our team.

All right, so our average speed was 128 kilometers an hour. When I’ve given a talk in Boston, it’s much easier, in Madison is much harder, because I say, Well, that doesn’t sound like that fast, but imagine going through downtown Boston around all the curves at that speed and then people are like, “Whoa.” So that again, Madison, I don’t know the equivalent because every road’s pretty– You could go this fast anywhere. So we’re the fastest rookie team and then we’re the third place team, we’re the fourth place motorcycle. And I published a paper on it in 2012, if you want to read about a lot of what I’ve been talking about today. So the final thing in what is also talked in the paper is to go back and look at some of our assumptions and try to use some of that data to understand if our models were correct. And so this is the same plot that I showed at the beginning, which is battery energy versus distance. This is when we were trying to figure out how much battery energy we needed. The long story short the racing data confirmed the hypothesis that the energy was in between those best and worst case scenario. So that was good and we did not run out of energy. This is just an example of some of the other data, we had a lot of a lot of data. But as I mentioned, it’s one of the nice parts about it, I think, is that we have all this data about this very historic race.

Okay, so now on to the Pikes Peak race. There was a few years in between. I had lived in Switzerland, as Tom mentioned for a while, and then I had gone back to MIT as a research scientist. But at the back of my head, I always thought of the Pikes Peak would be a fun race to do, mainly because there are just a few hundred-year-old races so. Isle of Man is one, Pikes Peak is another. Indy 500 is another one. And then the Mount Washington Hill Climb is another one that sometimes is given credit. Maybe there’s some other ones, but there aren’t that many. And so the Pikes Peak was another one of interest. The other reason is that it’s such a different type of race, so it’s a mountainous course. And so it’s much more about having torque and power than necessarily energy. So it’s considered the fourth oldest Motorsport event. And here are just some of the pictures that I got online about the more the historic perspective of the race. And here’s a video showing the course. So it kind of snakes along the road there. And there are also these little marking milestones with really awful names like Devil’s playground, which, you know, can’t be a good thing and the finish is at the summit.

So somewhat similar to Isle of Man. I didn’t really know much about the race. I had, like a lot of us, in 1980 something or family got in a minivan and drove from Illinois out to Pikes Peak and we put drove up with the bumper sticker on. That was pretty much last time I had been there. But we did a lot of the same type of things, which I’ll go into. But just to compare the two races. so they were started around the same time. The Pikes Peak is much shorter but it’s it’s much steeper, as you can imagine so much larger elevation gain, but it has about the same number of turns. And I think the interesting distinction here is that the Isle of Man is really largely about energy, you could run out of energy, that’s a real issue. Pikes Peak, you’re probably not going to run out of energy, but your motors might overheat, and it’s really about power and torque, and that kind of stuff, and if those of you that have followed the race, that’s why EVS are kind of nominating this race. They’re not quite dominating the Isle of Man yet. And it kind of has to do with that mreality. So I’m going to just go through very briefly, we followed a very similar process, maybe I was a little bit older and wiser, but was the same idea.

So we wanted to go fast, we didn’t want to run out of energy, and we wanted to be able to understand and improve what we were building. So very similar to before, we generated some models, we tried to simulate various aspects of the course. These are your similar types of plots. In terms of simulations and trade-offs, one of the things that we wanted to do was figure out where we should focus our energies, in terms of should we have a huge motor, should we try to have a light motorcycle. So we did some simulations, and it really showed early on, which is somewhat intuitive, that you just want a very light motorcycle, that’s the because you have such a huge elevation gain. So what this plot shows is the reduction in energy, how much energy you could save, basically, by reducing various parameters. In this case, it’s drag coefficient, mass and rolling resistance, and mass is red. So it’s like, if you can reduce your mass, you’re really going to save a lot of energy going up the mountain. Times have changed even in the few years between Isle of Man and Pikes Peak.

And the big difference is now we could actually go to a local store and buy an electric motorcycle. That wasn’t the case. And so we worked with a company called Zero Motorcycles. It was really fun to work with them similar to BMW and A123 in the past. So what we did is we pretty much took their lightest motorcycle frame, and then took their biggest electric motor, and then combine the two, so kind of the biggest of the motors, smallest of the bikes, and merge them together. And that was our race bike and this was this was a lot different because we actually had a prettyWe had like four international partners. And so the Isle of Man was much more a bunch of guys in a garage at MIT. This one was a lot more about collaborations and working with people, which is really a good experience to have. And so there was a team from Japan and the UK and others. So we all kind of contributed to the motorcycle. Here’s a summary of upgrades, we reduce the mass by adding magnesium wheels, we removed obviously, everything that we could, there was an essential. There was a motor and a controller upgrade that I mentioned. We kind of took the largest of those from their product line. And there was a custom battery.

There was race front and rear suspension and we changed from a belt to a chain drive. And there was a new fairing and we lowered the suspension. So the result was 16% reduction in mass and a 51% increase in torque in power, so is really good. And here is just to show various specs on the motorcycle. So the stock FXS, which they actually have a 2018 version. You can go buy one now if you’d like. This is about what you’ll still get, which is shown in the stock column. And then the next column is the one that we modified. And so that that shows the change of 16% reduction in weight, and then a 51% increase in torque. So the final thing is just dissimilar to the Isle of Man. We’re engineers so we really want to understand this vehicle. And we have a lot of sensors and that was the case here. So we knew the ambient temperature, the pressure, the humidity, motor, temperature, voltage, current, all the vehicles stuff that I mentioned before. Heart rate was new, so we had a heart rate sensor, which is interesting. We had a lot of other sensors, some of them failed. But those were kind of the main sensors that we had on the on the motorcycle. So that’s a brief overview of the Pikes Peak. This is a video showing a lot of the process similar to the Isle of Man video that I showed, all the way to the finish line again. This is at a testing track out in Western Massachusetts. (motor whines) So we all converged at a motel in in Colorado.

You’ll see a lot of Japanese because this was a Japanese filming crew there was following our rider who is also– He’s not an engineer, but he’s extremely technical. So these are the practices that I talked about, extremely early. One hour, two hours sleep best case. Five, six days. You’ll recognize some of the same faces. So that was my friend Mark was also a grad student. But that student there was from Sweden, the other one. This is up, I think like 10 or 11,000 feet. (alarm sirens, motor whining) This is the survival part. A lot of things you’re trying to fix and you’ll notice here he has a backpack on, I think in this one. Our data or data collection systems not working. So okay, throw this laptop on, put in the backpack and go. So that’s kind of the phase we’re in right now (laughs). Working out kinks, you know, we had chargers blow up and I’m driving up. Actually, I don’t know if you guys know, KillaCycle, the drag racing bike that he’s– I know that guy. And so he lives in Denver so I’m like, “Can I borrow your charger?” “Yes, okay.” I had to drive this house and pick it up.

So this is the actual race. This just shows the sensor. So here’s his heart rate and it’s just an overload of information here. But here’s RPMs. This is kilometers per hour and there’s a lot of temperature data on the bottom. (alarm sirens) And here’s a little map of the course, actually in the upper right-hand corner, there’s, like I said, there’s so much information packed in there. (whining motor) So that’s the finish line of the top. And so this is the team for Pikes Peak. And these were all the partners. So a funny story, I talked about at the beginning that I like to motorcycle, I’m a motorcyclist. I’ve done a lot of motorcycling. Well, in all honesty, when I first heard the Isle of Man, I thought maybe I could read in the Isle of Man race. Well, a good story to just totally realize how silly that is. When we’re at the Isle of Man, we were doing some testing over the north side of the island. I showed you some of those videos. That was me kind of flying by. But you could hear– Well, Mark and the other guy and I were the test drivers and we had our actual rider that would ride for the race. So Mark and I were testing up the north side of the island, you can hear us off in the distance. Testings are like, “Eeeeeeh,” going around all the turns. So that’s us, that’s a non- professional rider, “Eeeeeeh.” The rider himself, he gets on the bike. It’s “Eeeee.” You can hear it. (audience laughing)

And it was the same thing with Pikes Peak. I mean, these guys are amazing, the way that they can– You can see that in the video. I mean, it’s a whole different level to be able to ride a machine like that, so I’m nowhere near that (laughs). So these were the results of the Pikes Peak. So we’re the fastest University team. We’re the second- place motorcycle. We were beat by what was originally called Brammo. But then they were bought out by Polaris. So they just had an amazing machine there, that was super impressive. The max speed was 92 miles an hour, the average speed was 67. And then we used about 4.2 kilowatt hours. So to put that in perspective, that’s about a coke can worth of gasoline or a washer and dryer load-ish kind of thing, roughly gives you an idea, so not a lot, they’ll do that. I’m actually in the midst of writing a paper right now on this effort, The Pikes Peak, for the SEC conference in Detroit, coming up here.

But I so I don’t have a lot of digested results to show but just as an example, this just an experiment as an example of some of the data that we have. And what really what we wanted to understand is how could we have gone faster. Was it the motors limiting us. Was it the rider; was it his heart rate; was it other things. And so this looks at temperature and this is a distance. And actually so you can see that we were quite below our temperature limits, so anyway, that’s foreshadowing what I’m writing about in the paper. But it’s something that I’m interested in digging into. So I think I’ve actually never explained kind of the chronology from that original motorcycle and testing, all the way up till now, but you can kind of see that the first one was pretty much all I could do by myself. I mean, I was in a garage at Caltech building this electric motorcycle. It was fun and cool, but the real fun gets into the team, the team efforts. That was the thing that I really learned. I was working at JPL as an engineer, I was building big, complex systems, so I knew the team was essential. But all the stories I described, I mean, it’s much grittier and when you’re with teams, with a team, and working in an engineering team, it’s really rewarding.

It’s really fun to do things that there’s no way on the planet I could have ever done myself and that was definitely the case with these two races. So here, like I mentioned, I showed you pictures before. This is the team. I don’t know why but it ends up always starting very large, and kind of by the end, there’s about this number of people that are in on it, for the last six months or so leading up to the race, and very similar with Pikes Peak (laughs). Yeah, it’s really neat to see it and I think those that have worked in complex things, even in the in the sciences, and engineering, we’re almost all orthogonal axes with our own strengths and it just works together and it’s really magical to have that experience, those kind of experiences. Okay, well, the next part is related to really what happened afterwards. It kind of gets back to the individual work again, because a PhD, you can’t get a PhD on motorcycle racing. And it’s actually has to be very individually focused research work. So that’s pretty much what I did for a couple years after that. Now, I’m going back to the Isle of Man. And in reality, I just say, Well, okay, I spent a lot of time on that, how can I leverage some of that, and then work towards some interesting research question. So I started looking at this the topic of distance empty and this is like, 2011, it was right when we got back.

This is just to give you an idea, and like, I have a whole talk on this. I also had a whole thesis defense on on this topic. But to introduce you to the concept, this is the Nissan LEAF electric car, and that number in the lower right-hand corner, that’s distance empty, so you’re trying to predict how far you can go on a charge. And I don’t think I have to go into much explanation that you can see that there’s a lot of connection between the work in terms of trying to predict how much energy we needed on the Isle of Man does something like distance empty. So the premise of the research is really this issue that a lot of algorithms, especially at this time, they’re trying to estimate how much farther you can go, but they’re only using past information. So that’s what shown here, it’s a Tesla Model S. You’re in the present. You’ve got your past driving information, but how do you predict the future. And so the this is work again, that was kind of finished in 2013. There’s been a lot of work in this area since then, but this is the idea at the time and this was the one of the algorithms I developed. So conventional methods, as I mentioned, only use past driving information and what I worked on was an algorithm that uses past information, plus estimates of the future driving conditions to try to improve the estimate.

So that was pretty much that research. And there’s a paper that I wrote on it, if you’re interested in it. As part of this effort, I went back to BMW and got more involved with the Automotive Group. So it was a unique opportunity that they wereIt was pre-production of the I3, and they had released this car. Has anyone ever seen this car. It’s called the ActiveE. Yeah, so that was it was released 2011, 2012 some sometime around that. When I came along, what BMW had was terabytes of data and they were interested in working with me to try to analyze some of the data. And so I looked at the data, 700 vehicles across the world, in seven cities. Only in the US, are the ones I focused on. And it was over 100,000 trips to date. So it was actually a two of us at MIT. And then there was, they had a data person at BMW in Germany, and we, the three of us worked on it. And we wrote this paper in 2014. Just to give you an example of some of the results. In general, I was using the results to improve this algorithm that I talked about before of trying to understand range estimation. But as an example, this is a distribution of change in energy between subsequent drive charge events. So this is pretty much like, you charge your car, you go drive it, however long you do that you come back, you plug it in charge again. What I was trying to understand is how much change and energy is there between one drive that you did that charge cycle, and then another time. Because if they’re very steady, it’s actually very easy to predict range. You just use the past like what most algorithms do.

But the results here using that BMW data show that there was a 15% probability that the energy use would change by 30% or more. And so that’s when the algorithm start getting screwed up is when you have those kind of significant changes. And it has to do with auxiliary energy use and a bunch of things that I dove into with the paper. So that’s kind of it for distance empty. I’m happy to answer any questions but that’s how I ended up working on EVS for some of the PhD research. Finally, I just wanted to give you an idea for what’s going on here at UW, because that’s what really I do every day now. And I don’t work on motorcycles much anymore. So some of the information about UW: I was at MIT as a research scientist, but kind of during that process you saw, and it really stems way back as an art student, it really was this desire for hands on learning, making, building, I mean, that was the way that I’ve always been. And when I was at MIT, I did these projects and through that process, I met all the shops, I got to know all of them. I became trained with all the shops, and there was a unique opportunity to create a new Makerspace at MIT at that time, so I helped create that one.

And then my advisor was saying, “Hey, we have a collaboration with Singapore. Will you go over there and create a Makerspace there?” So I went and did that. And then when I graduated the PhD, very similar, another advisor, of mine said, “Hey, we’re trying to start this university in Russia. Would you go over and help them create a Makerspace there, too.” I mean, they call it different things, Fab Lab. In Russia, they call it Masterskya. But so that’s kind of the precursor, but so when I saw the job posting a UW super appealing, partly that I love that topic, and also partly married now, two kids and one to come back to the Midwest. And we were super interested in Madison. That’s how I ended up back here about a year and a half ago. And the main thing that I was asked to do initially was to create this Makerspace, which was all started in 2015 with a donation by the Grainger Engineering Foundation. That was kind of the the right part, which I’ll talk a little bit about. But I also helped oversee some of the machine shop, just more from the academic side and I could get more into that in a little bit. But just to give you a little bit of layout of the shops at UW in terms of in the College of Engineering.

So Team Lab is more of the traditional machine shop. It used to be called student shop. There’s a full-time staff. Makerspace has three full-time staff. There’s a lot of student staff that run the shops, and so that’s also part of the effort that I’ve been working on. So we have about 63 student staff total, and there’s about 24,000 square feet. So it’s a huge opportunity for students to work on projects. Here’s a little bit of chronology. I alluded to part of it, In the early 90sI should mention, this is all told to me orally by the people that have been around, so you may have more information. But I was kind of curious about what’s the history of shops, because they’re all quite different. At MIT, there’s a whole history, and each place kind of has a history of how they run their shops on campuses. So in the early 90s, I was told that the engineering shops started to consolidate. In 2008, they established a formalized training program, which is extremely impressive. It was one of the aspects of the job. When I came and toured, I was like, “Wow, these this is extremely impressive,” what they have, in terms of formal training that’s available for students, 2015 was the gift by David Grainger. At 2016, they started renovating. That’s when I came and was interviewing and I saw the library with all the chairs flipped over and they were starting to get it out on the first floor. And then I joined in the summer and then we opened in September of 2017.

And so as I mentioned, Team Lab, that’s the more traditional shop. They also have fee for service, but it’s mainly around metal and wood work. So that’s the mills, the lays the saws and this kind of welding and that’s the website. And in the Makerspace is the thing that’s the newest on campus in terms of capabilities and culture. It’s more about Do It Yourself projects for students. I think the motorcycle is probably an extreme example, but it is an example. It’s more about 3d printers, 3d scanners, laser cutters and all those pieces of equipment. And then there’s a website there if you want to check it out. But they’re just a few points, I just want to make briefly about the Makerspace and I’d be curious about your thoughts, or even afterwards, any of the feedback you have. But one of the main themes and it’s something that I greatly benefited from is having a space that is for students. You know, it’s really students are running this facility.

So the Makerspace is largely student run and so we have about 30 student staff, and here’s just pictures of some of them. And it’s just a really good opportunity for them to have a very long runway of real great opportunity for development and growth. And as part of that they run workshops, so pretty much every other night, we have some kind of workshop in the Makerspace. And it’s really about pure learning, for the most part these are students teaching other students, and it’s also about interdisciplinary activity. So here’s just some screenshots from various workshops that we’ve had from virtual reality to 3d printing, microcontrollers, define element analysis. So there’s a lot of different topics. We had, I think, in the last spring semester, we had about 75 of them. And I’ll show you some of the data from that. But here are some just pictures from events in the space. This is just the Makerspace. So like I said, the first part is kind of the student focused aspect to it and the really the student empowerment and peer learning and all the things that come along with that.

The second part is that it’s really more of a traditional Makerspace, so I would say traditional Makerspace is up in the large physical macro physical space. So these are the routers the drill presses, function generators, and other things. We’re also super interested in the virtual space, so we have a lot of virtual reality. And then we’re also interested in the micro space in terms of some of the tools we have, like a microfluidic printer. The third point is that really the interdisciplinary aspect of it and so we’ve been working a lot on creating various interdisciplinary programs and supporting programs. So it really starts from high school, and we support some of the high school programs that exist already, but it goes all the way up through capstone courses in into a master’s degree. So we’ve spent a lot of time creating this master’s degree, which is going to start in 2020, and it’s a collaboration with the College of Engineering, high school, School of Business School of Human Ecology. And I don’t know if I said the Art Department. The Art Department is in there, and the College of Engineering. So it’s these five and it’s really exciting.

And it’s really around this idea of design and innovation, which is kind of an overused term, but I think how how we mean it, is just really generating new products and services to solve true human needs. And this may seem like a little bit of a like, “Wow, how did you get from that to motorcycles?” I kind of skipped over my whole involvement at MIT and product design and there’s a lot of the courses that I taught. And so that’s very similar of what we’re doing here. And this is a course that we prototype with the School of Human Ecology this past summer. And we worked on the theme of farmers market. So the students went out through the farmers market, look for various product opportunities. Another point, it’s really my final one is just some of the data that we’ve collected. And it really has to relate to the mission statement and there’s a whole nother presentation that I’m not going to go into. I just wanted to highlight it for you guys to show some more relevant, more recent projects that I’m working on. But with the Makerspace, we had this opportunity that the College of Engineering had to shop for many years, and I showed you some of the chronology. And then all of a sudden, there’s a step input of this new Makerspace, which has some of the facets that I talked about. But we were curious, how did that impact thing.

So a couple of the research questions we looked at, and we presented a paper on it out at Stanford, just this past summer. We looked at these two things, so we said, How did this Makerspace impact the number of females and targeted minorities using engineering shop facilities? And then the second part was interdisciplinary activity. So I’m not going to go into the details but the bottom line is that we looked at the data. So we looked at user data from when the Makerspace was added and we can compare that to before when there was just a shop, just a machine shop. So it’s kind of those two together. And the bottom line is that in terms of females, minorities, we saw a 2.9X increase of total people going into the shop facilities, including the Makerspace. We saw a 3.9X increase in females, and a 3.7 x increase in minorities, that was over a nine-month period. And so similar with formal training, that was people coming into the door. This is actually the some of the formal training. So the bottom line is, we’re seeing some increases, which is good, that’s what we wanted to see.

The the second and final point is really about interdisciplinary activity. So the point here is that if with the Makerspace, what we didn’t want to happen is that mechanical engineers come in and just learn about mechanical engineering. We really wanted a space where mechanical engineers learn about other disciplines, where schools, other schools and colleges can come in and learn about other disciplines. So here’s just, again, some of the data. So after the Makerspace was added, there was an increase in departmental diversity within the College of Engineering. So before the Makerspace was added, we saw most dominantly mechanical engineers going into the shop, which is that’s what I am and that’s most traditionally the case. And then after the Makerspace was added, we saw an increase, which is shown in green, of people from Computer Engineering, Computer Science and other engineering departments. And then the final thing was this matrix that we developed. We talked about it all in the paper that I wrote, If you’re interested, I actually brought a couple of copies. But it goes to that thing that I was mentioning before. Is it our mechanical engineers coming in and learning about mechanical engineering topics. So that’s kind of the two row columns, the matrix. Are mechanic engineers coming in and learning about mechanical engineering. And if they are, we would see one down the diagnosis here. I mean, if you see a lot of off-diagonal values, that indicates that there’s a lot of interdisciplinary activity, which is what we want. And so the bottom line is, is that on average, 70% of those that attended the workshops that we had learned skills outside their major, so it was good.

So that’s pretty much all I have. And thank you very much for coming and I’d be happy to answer any questions you might have. (audience applauding)