Embedded - 29: Ducking the Quadcopter

Episode Date: December 4, 2013

Kathleen Vaeth of MicroGen Systems (@MicroGenSystems) spoke with Elecia (@LogicalElegance) about energy harvesting using MEMS devices. Some introductory videos: BOLT™ Micro Power Generator An energy harvester enables TI eZ430 with Linear LTC3588 While we missed it on the show, Kathleen also wanted to mention MicroGen Systems' finite element modeling partners: SoftMEMS and Open Engineering.

Welcome to Making Embedded Systems, the show for people who love gadgets. My co-host today is Dr. Kathleen Fay of MicroGen Systems. We're going to be talking about piezoelectric vibrational energy harvesting. I'm excited about energy harvesting. It seems magical we can get energy from our surroundings. Too much like a perpetual motion machine though. Kathy, I hope you can explain this. I'd be delighted to. Thank you. And you're going to help me verify this is physics, not magic. But before we dig in, what is your background? My background is chemical engineering. I have a bachelor's of chemical engineering from Cornell and a master's and PhD in chemical engineering from MIT.

When I was at MIT, my specialization was actually the intersection of materials and devices, and I've carried that throughout my entire career. Originally I worked on materials and devices for organic light emitting diodes technology for displays and the second half of my career I worked on MEMS devices micro electromechanical devices designing fabricating them and characterizing them the devices that I made were for inkjet printing systems. And I did a lot of that work at Kodak before coming to MicroGen Systems. And how long have you been at MicroGen? I've been at MicroGen a year.

And you were the vice president of engineering? Correct. Cool. So let's get this over with. What does MicroGen do? MicroGen is designing and commercializing MEMS-based vibrational energy harvesters. And they're based on piezoelectric conversion of vibration or movement into energy or electricity. As I mentioned, we use MEMS devices, microelectromechanical devices, and those are basically devices that

are made using the same kinds of steps that you would use to make silicon chips for computers. Right, and they have wonderful economies of scale because it is mostly the same processes. Absolutely, and that's one of the reasons why we're pursuing it. They can be made very cheaply in volume, and then also they can be made to be very small. But I have this idea that, I mean, I have this mental picture of you shake it and electrons fall out. How does this work? It's not quite that. Most of the device is made out of silicon. And as I mentioned, we use silicon processing to make it. But on top of the silicon, we coat a piezoelectric material. A piezoelectric material is something that when you strain it, when you bend

it, it actually develops a voltage across that. And that voltage can be harvested for energy. A lot of quartz crystals are piezoelectric, right? Absolutely. Yeah. Yeah. Quartz crystals are piezoelectric. They're not very good for energy harvesting, but they're very good for other types of applications. You may also be familiar with PZT. That's another type of piezoelectric material. That's actually very good for if you put a voltage into it, you can get things to move. We're doing the opposite.

We want things to move and to get a voltage out of it. So what what do you use? Oh we actually use aluminum nitride as our piezoelectric material. There's two advantages of it. One, it actually has the right kinds of properties to be very efficient for energy harvesting. You can get a large voltage with a little bit of movement from aluminum nitride. And the second thing is that just about any fab you go to in the world is going to have aluminum nitride, so our processes are CMOS compatible.

Okay, why would anybody have aluminum nitride? Because of the CMOS? Yeah, it's used in other CMOS steps. So it's often very present in a fab. Okay, so we're back to MEMS use the same stuff that chips use for fabrication, and that makes everything cheap and easy. Right, exactly, yeah.

If you want to try to introduce something like PZT into a fab, you're going to run into some serious contamination concerns. All right, good. We don't want to do that. No. No. Definitely not. So how much power are we talking about? Well, it is small, on the order of 50 to 200 microwatts DC. But the really cool thing about our devices is that

we, as MicroGen, have been increasing the power of the devices over the past few years and in the meantime, chipset makers have been making RF transmitters and digital sensors come down in their power requirements. So I almost feel like we're going up in power requirements and we're starting to meet the power requirements of some of the components that are necessary for sensing in the Internet of Things. And so I think that that's really where we're going to have our play. Let me think if I have this right.

50 to 100 milliwatts or 200? Microwatts. Microwatts. It's small. It's small. But there are LEDs in that range range aren't there little tiny absolutely yet we have shown and we can get into this later too but we have shown that that's plenty of power to power an led that's plenty of power to power an rf transmitter um a temperature sensor um we've

shown that it's plenty of power to power an accelerometer. But not all those at once. We have done an LED accelerometer and RF transmitter and temperature sensor all at once. Wow. And a processor too? I mean, I know the Cortex M0s are like 11 microwatts per megahertz or something like that. Yeah, it depends. We have used different chip sets. We tend to gravitate towards ones that have been specifically designed for low power, such as the Texas Instruments MSP430s. Yeah. Well, the M0s are low power too, but they're a bigger chip, so the 430s are much smaller and more likely to be in your range.

Okay, so that's a lot of stuff that you're putting on there and you're going to have a RF transmitter. What are you going to, that makes a great demo. What are you going to do with it? Well, our target market for our initial products is in basically in wireless sensors and particularly wireless sensors that could be used in industrial monitoring and in building monitoring. And so I'll give you an example. If you have a building and you want to monitor the temperature in the building, and the building's already been built, to actually put in a bunch of temperature monitors that can relay information back to a central computer to tell certain blowers to turn on or off. It's very expensive to actually put those temperature sensors in and wire them in.

Instead, you'd like to just install them exactly where you want them. Well, you can install them with a battery. It's perfectly acceptable. The problem is, of course, is that the battery is going to die over time and it would be much more useful and cost-effective if you could instead install that sensor and power it either directly with a vibrational energy harvesting or trickle charge a battery using the vibrational energy harvesting to make the battery last longer. Because very often in a situation where you put sensors in and they're powered by batteries, the largest expense ends up being either, if it's wired by batteries, the largest expense is actually paying someone to go and change the battery when the battery's dead.

Yeah, especially in the U.S. Yeah, yeah. And so the idea is to place, create the ability to place many of these sensors over an industrial site or a building and allow them to form a network or a mesh that can then cover that building and extract many data points off of that building or site that can then make that site smarter or allow monitoring of that site to be easier. And instead of having it powered by a central location, by wiring things, or by a battery, instead we actually enable that distributed network to have distributed power across it. And that makes a lot of sense. And you're in an industrial area, and you mentioned blowers as a good example. And that's partially because you're doing vibration stuff. Absolutely, yeah.

A lot of industrial equipment has suitable vibration to power our devices and generate enough power to power some of these wireless transmitters and sensors. What are you using for a wireless transmitter that can survive on that little power? Well, we've actually done several demonstrations of wireless transmitters that have been powered by our vibrational energy harvesters. One of our first demonstrations was actually with the TI-EZ430 wireless transmitter. We've also shown that we can power one of the dust smart mesh modes, and those are from Dust Networks, which was recently purchased by Linear Technologies. technologies. And our latest demo was actually with an RF transmitter from Anarin, which actually uses TI technology.

So plenty of options. Oh yeah, yeah, absolutely. And we've only kind of scratched the surface. Those are just the ones that we've, you know, had time to delve into because with each one of them they have their own unique software and there's a learning curve to go up that software, you know, to set up a network. Well, yeah. And you are probably more concerned with the actual making power instead of making the mesh network run, which is always the fun part. Absolutely. You know, we are interested in understanding how some of these networks work

because that's what our customers are going to have to do. But yeah, we don't necessarily need to become wireless network installers. And so I want to go back to the vibration. I know you have a couple of different products, but one of them, you measure the output of the unit based on what the resonance frequency is. How does that work? Right.

So I'll just describe our device a little more. I mentioned it's a microelectromechanical system-based device. It's essentially a tiny silicon machine. Our vibrational energy harvester is essentially a cantilever with a proof mass on the end of it. A diving board. I use this analogy with accelerometers all the time. Okay. Yeah, exactly.

The movement that we have is much larger than an accelerometer. And with an accelerometer, you're maybe moving, you know, a half, much less than what we're talking about. For our movements, we're moving a half a millimeter or greater. That's huge for a MEMS system. Yes, it's very large. Yes, we call our devices big MEMS. But anyways, so it's cantilever. And so, much like an accelerometer, it does have a resonant frequency and a certain response. And ours have a single resonant frequency.

We actually can tune the resonant frequency of the device to match a particular piece of equipment. The interesting thing is that a lot of equipment, especially equipment that's plugged into the wall, actually has in its vibrational spectra resonant frequencies at multiples of 60 hertz for the U.S. or 50 hertz in Europe. That sounds like the power frequency of those places.

You better believe it. That's exactly what it is. And so we find, we actually did a demonstration where we took a device that we had tuned to 120 hertz and we put it on a fish pump, we put it on the fan outside my office in the hallway of our office, a heating fan, we put it on the microwave that we use to heat our lunch, we can't think, we put it on a few, oh and then one of our engineers actually took it home and put it on his drill press and a couple other pieces of equipment in his basement. And we were able to power an LED off of the vibration, all from the

same device. And it was mainly because we were harvesting off of that electrical frequency. Resonance frequencies in more than one axis, really. Yep. Okay. So I have, what you said makes it sound like it's an accelerometer in reverse. Usually I put energy into the accelerometer and it tells me it shifts based on what the world is and I get a reading off. Right. Are these as similar as I think, or is it just that you've have a proof mass on the end of a cantilever and that's as far as I get with physics? I mean our devices can be used as accelerometers but they're nowhere near as complex as you know the kinds of accelerometers that you're going to find in today's smartphones but it can be used as an accelerometer it would be single axis

but it will with any kind of motion above a certain amount of acceleration, we are actually very sensitive to low G. A few tenths of a G acceleration is enough for us to start harvesting energy. Our cantilever will start to move and we'll detect a voltage. And that voltage, of course, can be used to harvest energy, but it could also be used as a sensor. Right now, we are commercializing it as an energy harvester, but down the road, we feel it could also be used as a sensor. Right now, we are commercializing it as an energy harvester, but down the road, we feel it could also be used as a sensor.

A few tenths of a G. Well, you did mention a fan and a microwave, so a few tenths is not very much. It's not. No, if you put your hand on a digital projector that you'd find in a lot of moon rooms, yeah, or a microwave, that's generally around a tenth lot of room rooms. Yeah. Or a microwave. That's generally around a tenth of a G to two tenths of a G. Okay. Let's see. I saw Robert Andoska, your CEO at the MEMS Executive Conference in Napa, and he put one of these units on a quadcopter

and then just flew it around the room to show the LED was blinking due to the vibration of the quadcopter. Yes, that quadcopter, even though it's not plugged into the wall, it actually does have resonant frequencies in multiples of 60 hertz. It was a little frightening. I wasn't quite sure. I was glad that I wasn't in the front row. Those things, I call those things the man toy at work. They're quite funny, aren't they? But it's a great demonstration. Yeah, it's a great demonstration because actually energy harvesting has, you know, I was talking about building an industrial, but energy harvesting is also of great interest for military applications. Of course.

Something like a drone, having sensors on a drone is quite interesting. I worked on sensors on a drone. Don't put a lot of money on that market. But I'm thinking about soldiers. They're moving all the time. They're in the trucks all the time they're in the the trucks all the time and i you know that when i worked on the drone when i worked at shot spotter with gunshot location i we one time went to go talk to military organization about gunshot location and the

army or air force or whoever we were talking to uh and they pulled out a bag that was the typical bag that they carried and the thing was like 10 pounds of batteries oh sure and they were like this is why we will not let you add more batteries to our system yeah are you looking at at human wearable sorts of stuff we've just started to um started to and it's important to make a distinction in there's two types of ways that you can harvest energy. One I was describing was the resonant mode where you actually have a particular vibrational signal that's specific to a particular frequency and that's our first set of products. But when you go into something like wearables or even tire pressure sensor monitoring, you're looking more at vibrations that are low on the order of 1 to 10 hertz

and can almost be thought of more as an impulse as opposed to a regular steady vibration. And in that case, it's difficult to get a device into a resonant mode with that kind of stimulation. However, we have shown, because our devices are sensitive in G, and they're also rather small, you can actually drive them in an impulse mode. You don't get as much power when you do that, but you can harvest enough energy. For example, we have a video on YouTube where you can power an LED just by taking one of our harvesters' package and hitting it on a table four or five times. You can get the LED to start blinking. And so wearables is something that we're considering down the road. We think that we have a place to play in there, but it will be a couple of years of technology development before we are ready for prime time for that. I did see this demo, and I think I hit it a couple of times, and then it vibrates.

Is that? Yes. That was the demo? You can think of it like a tuning fork. Okay. Basically. Yeah. You, you hit the, you hit the device. It's a cantilever. It's got a resonant frequency. It's going to resonate at that resonant frequency. If, if the cantilever moves, there's going to be a voltage there to harvest. I, I have worked with wearables and you know, it's not that I want to replace the battery entirely. That doesn't seem likely to happen.

I just want the batteries to last an extra couple of days. Instead of six, I want 10. Do you think we'll get there sooner or do you think that's further off? No, I think so. You know, people like to talk about the idea of going completely off battery. And that's a nice idea. But, you know, when you think about it, it's probably more realistic from a risk mitigation perspective to trickle charge a battery, be it in the resonant mode or in the impulse mode. And so I think people would find it perfectly acceptable to be

just increasing battery lifetime rather than getting rid of batteries altogether. Because, you know, especially in an impulse regime, there may be times when you're not moving around too much, so there's not as much energy to harvest. And so you'd want that battery there to allow your devices to continue to work. But when the impulse is there, you'd like to be able to feed that back into the battery if you could. Okay, that makes sense. When can I have my microgen battery recharger?

Well, if you'd like something in resonant mode, you can have it in June. We're actually, the devices that we're coming out with in resonant mode, they actually will be the harvester and also conditioning electronics to present to the outside world DC power. So, you know, the harvester creates AC power, and you have to convert that to DC power.

And we will have the option of trickle charging either a battery or a super cap or a bank of super caps or one of these thin film batteries. For something in impulse mode, that's probably going to take us a year or two, but we'll get there. Okay, so resonance and June. How much is that going to give me if I'm resonating optimally? To start, 50 to 200 microwatts DC, like we mentioned before. That's a big range is that a big range

because it's not done yet or a big range because optimal resonance oh you know no great question actually it's a range because it depends on the frequency that you actually operate at higher frequencies will give you higher powers oh i guess that makes i mean that makes sense. Of course, you're wiggling something up and down. That's actual potential and electrical. Right, and if you do it more times per second, you're going to get more. Yeah, of course. Yeah, and that's the range. So if you pick one of our lower resonant frequency devices,

you're more in the 50 range. Now I should say, what I'm speaking of is for one device. Don't forget these devices are small. They're a centimeter to a centimeter and a half on a side. So you can gang them up if you want to, to get more power as well. And that's more realistic for like the industrial and building market. For the wearables, that's something different. You may not want to be ganging up a ton of them, you know, to build up power. But we also do have on our roadmap to reduce the size of the harvester as we're looking forward towards wearables and TPMS. And so I wouldn't say that we would never

gang them up, but it's just that space would be much more of a concern than it would be in something like an industrial or building market. How high are they? How thick? Just a few millimeters. So couldn't you just stack a couple? Absolutely. Do you do that in dye or packaged, post-packaged? I guess you're just making the packages now. You probably don't want me to use the dye already. Yeah, yeah, yeah.

There's no reason you couldn't do it by stacking dye. But right now, yeah, we're working through our processes and we're doing our first commercialization commercialization our first product so right now I would probably stack them packaged I should mention though we're actually making them wafer level packaged and what that means is instead of you know cutting the dye out and placing caps on individual dyes which are kind of machined somewhere else, we're actually making caps using micromachining as well. And we're making a whole wafer, and we work in 8-inch wafers of caps, and then placing

that directly through wafer bonding on a wafer, an 8-inch wafer of devices. So we have some scale there by doing the wafer-level packaging of the devices as well. Great. Okay, so we're talking a lot about wafers and chips. Is there going to be some sort of Moore's Law for energy harvesting? I don't... We talk about that sometimes, you know, internally. I don't think so, because one of the things you have to remember

is that energy harvesting really depends on the amount of piezoelectric material that's present. And so there are ways to reduce the area that's there and still maintain the same amount of power, but you're going to run into a zero-sum game at some point. You have to have some of it there, basically, in order to get power out. So if you think of it in terms of getting the same amount of power, I don't think it's going to follow Moore's law. But I meant more from my perspective, or the end-user perspective., so this year we're going to be able to power an LED and an RF transmitter and

temperature sensor and a tiny processor. Next year, are we going to get, you know, maybe a wifi transmitter and a tiny processor? Oh, I see. You're thinking more, not so much in terms of shrinking things. You're thinking more in terms of increasing the power. And making it cheap. Yes, well cheap comes from volume. And the power, yes, the power is going to continue to increase over the next few years. Robert Andoska, you know, we didn't really, you mentioned he's our chief CEO, but he's also our founder, along with Jen Rue from the University of Vermont.

And Rob really strongly feels that conservatively we can increase the power up to eight times the values I'm talking about right now. At eight times, you're getting a watt almost. A milliwatt. A milliwatt. Thank you. I knew there was a milliwatt a milliwatt thank you i knew there was a yes a watt would have been phenomenal yeah yeah milliwatt still very very cool especially with the game you can do yeah you can do a lot you know especially if you start to get into the fight you know if you gang it up and you can get into the five milliwatt region, now you're talking smartphones, but that is farther off for us.

So there is a plan to like be able to power smartphones based on ambient vibration and normal human motion. That's, we keep our eyes on that. Like I said, you know, we're just starting to look at the wearables, not for powering the phone itself, but for powering these wearables that communicate with the phone. And it's probably one of the toughest environments for energy harvesting for the one reason I mentioned, which is that you've got more of an impulse input for vibration as opposed to a regular vibration. And then the other thing is that you do have low acceleration levels, low G levels. And so we're analyzing it. we definitely have our eye on it.

Another area that we're working on that's related to it, but actually has the high accelerations that we feel we can enter first, is the tire pressure sensing market. And so if you're not familiar with those, these are the sensors that are in the wheels of cars that sense the pressure in a tire and communicates that back to the main computer of the car. Yes. The first day I got my new car, it immediately, they'd over-pumped the tires. And as soon as it got hot, it said the tires were over-pressured. Yes. We had to take our electric car to a gas station.

It was horrifying. Yes. But it was really cool to have the car say, you know, my tires, could you take a look? Right. Because it is a way, and low pressure tire is much worse because it's inefficient.

Right, exactly. Yeah, I could see how that would be a good and pretty easy market for you because you're getting energy from this and resonant or impact, there's a fair amount of energy going on with the tire going round and round and round. Exactly, yeah.

But when you go to wearables or cell phones, you can't do one axis because you don't know which way it's facing. Yeah, that's true. And we've been talking about that. We actually have a lot of ideas on how to incorporate our devices to try to compensate for that. I can see how you need this for,

you need the energy harvesting for areas where you don't have batteries, like inside your wheel tires or tires. Well, right now, inside the wheel tires, there actually are batteries, and that can be a problem, especially from a landfill perspective. But battery technology in general is getting better. They get smaller they they get more i don't know mostly they get smaller and cheaper which is what i tend to care about yeah i'm not

so sure about the power density but smaller and cheaper is probably true well the yeah and the processors get less so then i i think that the power density must be increasing because I can do more with what I had. But really, is it the power? Or maybe the power. Yeah, it's the management of the battery that's probably getting a lot better because, you know, that same technology that we're taking advantage of can also be applied to the management of the battery as well.

Yeah, and we see them as, you know, in some cases working together, and in other cases you really don't want the battery around. The battery can be a safety concern. A good example is children's toys. There's always a lot of concern about children swallowing the button batteries, and manufacturers go to great lengths to package those in so that, you know, children can't get at them. Yes, yes.

It was the one area where we all had to have screwdrivers because you have to put the battery compartment behind a screw because apparently four-year-olds can't use screwdrivers. Yeah. Who knew? I don't know. My four-year-old can, though. Exactly. It only defeats the ones who have already been eating the batteries.

No, but it is a serious safety concern. That's for sure. But children's toys are, I mean, that's a neat application. That's probably more impulse than resonance. But I can imagine a kid having to bang on it in order to get it to light and work. And that would be kind of fun. Yeah. So I think what you would do with the energy is much more simple and probably less power hungry

than, say, a wearable. And then, yes, you know, children are naturally energetic. I think you can, based on that feedback, get them to maybe put more acceleration in than just you or I running around, you know, on a normal day. And so, yes, yeah, toys are definitely on our horizon. Oh, yes.

That needs a demo. let me know if you want help with that demo i'll feed that back to the team how big is your team um we are five people right now oh so you're really a startup yep we are a startup robert makes it look all you know we're a company and and we're big and we're going to do all of this. And I didn't ask him how big. We're five people, but we're five very experienced and talented people. And you're located in Rochester, New York, right? Yes. In upstate New York, Rochester, New York. Are you hiring? Not that I want to be there. It's winter. Winter is coming. Yeah. And it's, it's snowing today, actually, if you can imagine. Yes. Yes, we have been hiring.

We've been growing the team. Like I said, I came a year ago and we've hired several people since then. Because I keep, I mean, a year ago, I didn't hear very much about this. And then I started hearing about MicroGen and how cool everything was. Is the whole industry growing? Are you guys driving the industry, talking about it more? Well, you know, Rob's a great evangelist for the company and also for energy harvesting in general. I mean, you know, he runs an excellent LinkedIn users group where he will post, you know, lots of different things about energy harvesting, not only about vibrational energy harvesting, but the other forms of it as well, and the other companies working in the arena.

He has a great passion around energy harvesting and bringing that entire market to fruition. A little bit of history about MicroGen. Rob started the company with his Ph.D. advisor back, I believe it was in 2007. But the company really didn't get going until 2011, late 2011, early 2012, when we had our first major strategic investor. And so that's when Rob really started growing the company. But you're not shipping dev kits yet. You mentioned June.

Is that the first time that I, as a developer, might be able to play with it? Right, yeah. So most of the technology development was done at the Cornell Nanofab in Ithaca, New York. And that was a great place and a great partner to develop the work, but we couldn't sell commercially out of Cornell. And so over the past year, what we've been doing is transferring that process and improving on that process to our current manufacturing partner,

which is XFAB in Germany. And so that's what we've been concentrating on over the past year is the technology transfer, working out the yield curves, and fleshing out the entire development kit. What they had worked on at Cornell was just the harvester, but the dev kit is more than the harvester. It's the electronics to condition the power and a few extra tools to help people understand their

environment and what the vibrations around them are. Yes, I can see you getting a lot of people saying it doesn't work and you saying, well, are you vibrating it? Right, right. And you know, interestingly enough, there aren't a ton of tools. There are some, but there aren't a ton of tools out there to help you know. And we actually have built several internal tools that we use when we're evaluating a site. And we're going to be taking a fraction of actually installed a wireless sensor network at a local university, Rochester Institute of Technology, who's been a great partner to us. We installed a wireless sensor network in one of their buildings, harvesting energy off the air handling units in the top floor of the building. And we learned a lot doing that. We learned a lot about what our customers are probably going to need and the kinds of questions they're going to run into and the kinds of challenges that they might face in trying to evaluate where the proper sites are for installing the harvesters. that magnetometer manufacturers have, which is people say it's broken,

but then you go to their environment and you explain that, that maybe you shouldn't be right next to a metal wall or you are going to be saying maybe you shouldn't be putting it on a springy table. Yeah. Yeah. Yeah. Yeah. There's going to be some edge.

I mean, this is new. This is a new field to a large extent. Um, and, um, at least using it for the, yeah, yeah. Yeah, there's going to be some education. I mean, this is new. This is a new field to a large extent, and at least using it for the wireless sensors and forming them into a network. So we recognize that there is some education that's going to happen, but we're happy to do that. Who are your competitors? Well, from a MEMS perspective, I don't know of anyone else that's actually commercializing MEMS vibrational energy harvesters. There are other companies doing more macro vibrational energy harvesters that are piezoelectric. My understanding is that their devices are more made by hand that don't have the scale that the micro machining has.

Well, they're going to be big too. Yeah, and they're bigger. They're bigger. They're not as lightweight. So they're not really a direct competitor. I mean, they serve a different market and they have excellent products. They just serve a different market and they're going to have a different cost structure than what we're going to be able to offer. Then there are other types of energy harvesting as well, solar and thermoelectric. Of course there's a lot of solar out there and there are a few thermoelectrics out there. The way I look at it is that there's a lot of room for a lot of different energy harvesting and which one you use is going to be very customer specific.

We think the industrial monitoring and the building monitoring is a very good one for vibrational energy harvesting, particularly when you're like us and you're going to be offering such a small unit because then it really isn't going to interfere with the rest of what's going on in the system. I have heard some neat things about thermoelectric, but I haven't ever seen it work. Oh, really? So, I mean, just, that may be another one of those, the user is doing stupid things.

So many things are like that. But, yeah, I, you know, they say you can get a little bit back from your hot part of your device, but, yeah, none of my parts are that hot i'm not a good candidate for that um and then we use i've used solar but never on a device uh not for for shipping it to people so this i i can completely you know, putting it in and it charges the battery when nobody's looking. And that would be so cool. Are there inherent limits to the piezoelectric part of it?

I mean, you mentioned the chemicals. Could you, when you rule the world, can you move away from the aluminum nitride or and actually start maybe contaminating somebody's fab with something more efficient well aluminum nitride is one of the best materials out there in terms of simple materials binary materials materials made out of two components two elements and out there to use. Zinc oxide is probably a little better, but it's harder to process. So we don't see a large need to actually get away from the aluminum nitride at this point. It's RoHS compliant, as I mentioned before, it can go on any fab without any kind of pushback.

So no, I think from a material, of course there's always opportunities to improve on materials. You can even improve on the aluminum nitrate family of materials. But that's probably not going to be one of our primary interests going forward. And you are small and you don't have a product yet. Does that mean you're looking for funding? We're always looking for funding, yes. We have a $5 million round open, which I'm sure Rob mentioned at the MEMS exec congress.

If he did, I was too busy ducking the quadcopter. I'm sure he must have. That was probably a good choice. Yes, we do have a $5 million round open. Well, I wish you the best of luck with that. Thank you. It's a tough market to be in MEMS and getting funding, but boy, I just am so excited about this. It's green and small. Yeah, well, We are too. I mean, that was one of the reasons that, you know,

I decided to join the company. The technology is very compelling. I think we're at the beginning, just at the very beginning stages of what could really be done with this, and we've talked about a lot of different applications. In some cases, you know, we're ready for it. In other cases, there's more technology development

that needs to occur, and I think there's just tons of opportunity. And I'm very excited about the future. So is there someplace I can sign up to find out when your dev kit's ready? Well, you can keep checking our website. And I'm sure there'll be a press release. And there's always, you can contact us at gotenergyatmicrogen.com. Well, if you think about it, do let me know and I'll make sure that I mention it on the show because I want one. Sounds good. You personally have a deep background in MEMS.

You mentioned a lot with the microfluidics and the organic electronics with the OLEDs. How did you get into that? I mean, chemical engineering, that goes in lots of different directions. Yes, it does. Well, when I was in graduate school, like I mentioned, I was doing OLED materials, you know, deposition of thin films, processing of them, but all organics. But my lab, I worked for Klaus Hansen, and his lab is a wonderful lab. He always keeps a diverse set of projects going on, which ends up being a free education for the person who's studying with him. And about a third to the half of the lab was actually working in MEMS.

So while I wasn't working in it. I was exposed to their talks and saw some of their devices, saw their excitement when things worked or didn't work. And some of my closest friends were working in the area, so I always had an interest in it. And then I went to work for Kodak, which was a wonderful experience, working in display. And one of the reasons I went to Kodak is some of the seminal discoveries in Ola technology were actually done by Kodak researchers and so I was able to work there

for a while once my projects had finished up I was actually given an opportunity given my background in thin film technology and to move over to the inkjet area. So it was being in the right place at the right time, but also the opportunity of working at a company, a big company that had a lot of wonderful technology development going on, and getting a chance to basically change fields without having to move. And so I spent the next six years learning how to design and fabricate and characterize these etched- characterize these essential devices. And towards the end of my career, I actually worked on piezoelectric actuators, designing them and characterizing them. And that was the link into doing the reverse, harvesting energy instead of trying to actuate something with piezoelectrics to lead me into micro gym. Neat. You were at the IDX, ID Tech X?

Yes. Energy harvesting and storage event in Santa Clara last week. Yes, I was. Did you hear anything particularly interesting at the conference? Well, you know, I didn't get to attend the talks that much. I gave one, but that was about the extent of it. And I was at the booth a lot. One thing I did see, though, I did attend, you know, the booth, it's actually very interesting

because the people come to share the enthusiasm that you have around the product that you're working on. I did see one interesting talk that was one of the first talks of the session. A gentleman from Disney was actually talking about a form of energy harvesting, really. He was using tribology, where you would have to actively rub a patch, but that was enough to generate energy that was enough to drive an LED or actually make a needle move on a voltmeter. I have this idea that we put these on all of the people who have chicken pox. That's interesting.

What he was showing, though, and I have a six and a half year old and a four year old, so I mean, this is going to be on my radar screen. But what he was showing was, and this was Disney, was it in the context of a children's book. And so I could see a child being willing to put that work in to get that needle to move. And I thought that was interesting. I mean I don't know if it would be a widespread use of energy harvesting. I don't know enough about it and the jury may even be out on it, but his devices were very simple to make. He showed that he was basically making them by cutting pieces of material and taping them together. And so at least for the children's

book application, which would be right up Disney's alley, I thought that was kind of cool. Oh, yes. I mean, okay, so you have to scratch it or press it or whatever. And then you can fight the evil witch. And then the arrow can go over and point to the thing. Yeah, exactly. It's so cute. Yeah, it really was kind of neat. So that was something that had not been on my radar screen at all, but I thought it was a really cool idea.

So having spent so much time in the booth, what question did you get asked over and over again that I haven't already asked you no i think you hit on it i think that it's it's people people understanding the difference between resonant frequency and impulse um what each can do for you because that each can do something for you and just uh what kind of situations now we're ready to uh to uh provide a solution for energy harvesting and and what we'll be able to do in the future. And I suppose everybody wants to know how much power they can have.

Yeah. Yeah. And if it's an engineer, they always want as much as they can. Oh, all of it. I want all of the power. That's right. Well, that's the show for this week. Thank you so much for joining me.

Oh, thank you. It was very nice talking to you. I've been speaking with Dr. Kathleen Vaith of Microgen Systems. Links will be in the show notes and on Embedded.fm. If you have comments or questions for me or Kathleen, hit the contact link on Embedded.fm. And thank you for listening. One final thought this week from Ray Bradbury,

the science fiction author. Yeah, I'm tying in last week's show too. His quote, It's perpetual motion, that thing man wanted to invent but never did.