Batteryless IoT Sensors: A Demo

Everactive’s batteryless sensors provide cost and time savings by requiring no maintenance, pose a solution to the ecological catastrophe of batteries and allow the opportunity for the expansion of the industrial IoT through groundbreaking low-powered technology.

Alright, so the topic of today’s presentation and live demo section is battery-free IoT sensors. So in about a half hour, I’m going to take you through a little bit about what our company does. We’re actually going to see some real sensors here powering themselves or powering themselves with harvested energy live on the desktop here in my little lab. And then we’ll talk a little bit about why our customers love us and kind of how we brought this technology into the world. So that’s our agenda here for today. We want to be respectful of your time so we’re going to try to exit right at 30 minutes. But I will include contact information at the end, so if you have follow-up questions we can get connected. 

So a little bit about me, my name is Peter Woodman, I’m the Principal Sales Engineer here at Everactive. I’ve actually been with the company since before we put our first product out on the market. It’s been really fun to see this new concept of batteryless self-powered sensors go from an idea to something that you can hold in your hands. And we’ve delivered out to customer sites and reproduced thousands of times. So it’s been a really fun, exciting journey. Thanks for joining us today to talk a little bit about it 

The Battery Problem 

First, I want to talk a little bit about batteries though and why batteries are not a great idea for the Internet of Things (IoT) in general. There’s a big human cost. We’ve been all over industrial environments around the world in our company. We’ve met thousands of people who work in these environments and I’ve never met somebody with the title battery changer. So if you add a battery to a piece of equipment at your facility you’ve now made a commitment that someone’s going to have to go back and visit that sensor to replace the battery. Usually that’s gonna require a trip you made all on its own. You could go and pull a battery out while you’re already out there, but then you’re gonna sacrifice some of the energy that’s in there, so it’s a wasteful thing to do.

Particularly in industrial environments, with our customers the people who are doing this are maintenance professionals. They’re really highly trained and could be doing better things than changing batteries. You’re stealing cycles from their job to get them to go out and do this. So it’s a wasteful thing from a time standpoint. Maintenance planners have to make sure you have these parts in stock, schedule the actual replacements and do some things with the batteries that come back out. So across the board, there’s a big human cost here in adopting battery-powered sensors around your environments.

 But there are other costs too and trade-offs I should say, in terms of what you get back. So because of all the things we just mentioned we see in many environments people are instrumenting fewer things than they’d like to. Maybe the top five or ten percent of their facility is getting a battery-powered sensor because they know they’re going to have to go back and do maintenance and visit those sensors again. But the sensors themselves have trade-offs : they ration how much data they transmit in order to milk that battery and make it last as long as possible. So you’re getting fewer insights than you could. We’ve seen that with many of the pieces of equipment around industrial and manufacturing facilities, if you’re not getting continuous data you can miss important things. So you’re being asked to get by with the minimum amount of data to run your facility instead of the maximum amount of data.

Some people may say that some data is better than no data. They haven’t had a choice until now. They’ve had to deal with what they can get from a battery-powered sensor that transmits a couple of times an hour to a couple of times a day. But we have designed a better way here at Everactive, a different way to think about this and that’s sensors that are powered completely by harvested energy.

What It Means to Be Batteryless

 So let’s take a look at what it means to be batteryless. There are five sources of energy that we can harvest from with our sensors. They’re outlined here on the right side of this slide. The two most plentiful in the applications we have out today are temperature differentials: so something that’s warmer than something else or the presence of light. So these are the two I’ll show you, the harvesters we have for today. In the case of our first product, it was for steam systems and steam trap monitoring. You know, in a steam system you’re gonna have a hot pipe so we take that heat that’s radiating off the pipe already. We scavenge that waste heat and turn it into electricity and there’s enough energy there for us to power our sensor and have it run perpetually. 

In the case of the machine health monitor, our second product, which I also have one to show you here today. There are some times where a motor or a piece of driven equipment is warm and sometimes where it may not be. So we’ve added a second harvester there to harvest light and that could be LEDs or CFLs like the room you’re sitting in now or our traditional incandescent bulb. But it could also be outdoor light like the sun. Outdoor light, that energy from the sun, IR light in particular is a very plentiful source of energy for us. So that’s another great way to run our sensors. If you think about something like a motor that’s in a dark room and it runs very close to ambient temperature, so we don’t have temperature differential.

The next harvester we’re working on is for electromagnetic fields. So if you have an electric powered motor, it’s going to throw off a lot of EMF and harvesting. Scavenging it and using that energy to power our sensor is next on our roadmap. There are other ways you can harvest energy too like RF or vibration; these are less plentiful. So, those are things we’ve been working on, we’re not shipping a product that uses either of those harvesting methodologies yet, but it’s something we see down the road as another way to help keep these sensors going without having to be reliant upon something like a battery

Our Technology

 All this technology was spun out at university research. Our co-founders are university professors at the University of Michigan and the University of Virginia. They met at MIT when they were getting their doctorates and then went back to their home universities to continue this research. It’s really that the super low power electronics and specifically the low power radios that they designed, allow us to get away with these humble sources of energy that are already there powering our sensors. So that’s a little bit of an overview. We’ll talk more about these individual components here in our live demo. I’m going to stop the slides for a minute and just switch over to my camera here in the room. The first thing I’m going to show you is a steam trap. 

Steam Trap Monitor 

So our first product is the steam trap monitor. We won’t dwell too long on these but just know this is what it looks like when it’s installed inside a customer environment. I’ve actually got a hot plate here today since I don’t have a steam system. Looking here at the table top you’ll see we have a hot plate set to C so this is warmer than I need it to be. We can actually power off, more like 20 degrees C temperature differential or even lower than that depending on the application. But that’s what we’re using here and I’ll track these temperatures. So keep that in mind we’ll see it in our dashboard in a minute

 So sitting on top of this hot plate are two of our energy harvesters. These are for thermoelectric generation, so they take heat and turn it into electricity. I’m actually going to take a cold one that hasn’t been sitting on the hot plate and walk you through the individual components. So this aluminum is for heat transfer and this rounded one is made into a pipe for steam systems. You can see we have a flat one too for motors. And then these fins are for heat dissipation; it’s actually a difference in temperature that allows this device to produce electricity. So if you look at the core here it looks kind of like a scrabble tile. That’s called a peltier device, it’s two pieces of dissimilar metal. If one is warmer than the other then it starts to emit a trace amount of electrical current down this wire. Not very much electrical current, I’m going to give you a point of reference .I’m wearing an Apple watch here. This is considered to be one of the most power efficient consumer electronics available in the world. And three of the big four consumers of energy on this device are radios, LTE is the big one, Wi-Fi, there’s a screen that’s the third biggest consumer and then Bluetooth Low Energy is number four for the device, number three among the radios. Bluetooth Low Energy is considered to be an ultra efficient radio it runs at about 50 microwatts. and The way it gets away with that is it turns itself off most of the time and then fires on in little blips for fractions of a second to transmit data back and forth. So by shutting itself off it can hit 50 microwatts.

The Evernet radio we have in this steam trap monitor, that’s always on, always listening runs with a power budget of 200 nanowatts. So it’s a thousand times lower than BLE and that’s what allows us to use a small energy harvester like this. If I plug this into my watch it wouldn’t even boot. You couldn’t charge a phone or anything like that. An energy harvester to do that would be many many many times larger and need much much hotter inputs in order to generate that kind of electricity. So that’s kind of the core IP that allows us to do this almost magic radio that is so low in its energy consumption. So this thermoelectric generator is one of several different styles that we make. There are also some thermistors that are here on this harness, and you can see they’re color-coded red and blue for the inlet and outlet side of the assets that they mate to, and then a plug that goes into our sensor.

Machine Health Monitor

So this is the sensor pack itself. This is our first gen product. The little cube you see up there in the upper right hand corner is our second gen product, our machine health monitor. So the thermoelectric generator for that looks a lot like the steam one. Same peltier device in the middle, but flat here with some stabilizing arms to mate to things like motors. And I should note that these are magnetic, so you can stick them on metal and they mount that way you don’t have to necessarily epoxy them down. Our second gen platform also uses USB C connectors. So there are rubber gaskets in there and screws to keep them from being disconnected accidentally. But this is a more standards-based cable and it’s modular, which means we can use a bunch of different lengths. So I’m using a longish cable here but we also make these in a bunch of different sizes depending on your application, so here’s a shorter one.

You may see some little blips coming off this steam trap monitor and that machine health monitor up there. They’re reporting back to our gateway anytime you see one of those flashes. So we’ll take a look at those data points here in a minute. One other thing I wanted to show you while we’re in this view though is our harvester. So this is a solar harvester, it daisy chains down from the TEG. So you can see there’s a port there, an open USB C port and one here. So we can use those modular cables to connect these and then run this to somewhere where we can get daylight or indoor light. So it’s magnetic, so I can stick it right there and you’ll see it connect up. Okay so that’s what’s going on on the tabletop demo. Just the heat of this hot plate powering these thermoelectric generators and that’s what’s keeping our sensors running.

Now when we take a look at the data you’ll notice that there are times where I’ve turned this temperature down. Like last night, when I went to sleep I turned the hotplate off so I didn’t have something running so hot last night. It is important to note that the sensors didn’t go off; you can actually turn the heat all the way off and these sensors will continue to run for hours at a time. We don’t use batteries to do that because even rechargeable batteries have a limited number of cycles before they wear out. But we do have banks of supercapacitors inside our sensors so we can do some energy storage. So if you have a process or a harvesting method that’s only available part of the time while heat is there or while light is there, we’ll fill those super capacitor banks. And then we can coast for long periods of time. So if the thing that you’re monitoring shuts off, we can tell you we would say hey wait a minute it’s not generating heat anymore please go check on it or the thing that you’re measuring.

So in the case of the steam trap monitor we’re measuring temperature. So I have my probes here and we’ll see the output of that. The machine health monitor is driven primarily in an accelerometer. So not a lot of accelerometer data to be had here just stuck to my pipe. So we’ll take a look at this temp data and I’ll show you some trends here.

 Alright, back to our screen share, so I’m going to move out of our slide deck for a moment where we are. Thanks for bearing with me and here’s a view of that same sensor we were looking at in our cloud platform. I’m gonna hit update here, and you’ll see this data draw and these are measurements that I’ve taken since we started our webinar.  Alright, so each measurement we’ve taken here is in our cloud platform streaming as we go. I’m rounding these off with a five-minute down sample right now because we’re going to look at a couple of days worth of data, but we could actually show every measurement down to the minute. So here in our software, we have graphs and you can roll over each individual measurement to see what it was at that time. We also have a suite of analytics algorithms that run in the cloud. So for something like a steam trap where the behavior has to be learned we train on it, and then when there’s a change like a failure we send you a notification. But if you want to inspect it yourself we make all that data available to you. So if there’s something like this curve last night you can see when I went to bed I shut off the hot plate and you can see these temperatures dipping down. 

So this green line is the steam supply side, it’s that red tag that’s on the harness we looked at. This navy blue line is the outlet side, that’s the blue tagged line on a harness we just looked at. And then this ambient actually comes from the sensor pack itself. So we can give an ambient temperature measurement in the room. Typically in a steam application that sensor is on an insulated stretch of pipe or strut nearby, so it’s not exposed directly to the heat. So that’s how we can determine what the ambient temperature of the room is. When there are failures we can call that attention to those with our notifications they are here in the dashboard but we also can send emails and text messages to get you that data. So take a look at our peaks and valleys here since today. I was playing around with our hot plate a little so we’d have some data to look at up and down. 

Software

We keep all these temperature measurements available online so you can go and see them at any point. For example, if you have a piece of equipment and you want to see how it was behaving a week ago, a month ago or even a year ago we keep every measurement we’ve ever taken available there for you on the web platform. We don’t like to nickel and dime you to go back further in time or delete after a certain amount of time. Because it’s actually really valuable to be able to say hey we know this asset, we know its history, now we can look back and see when it was under different loads or in the same exact behavior it’s exhibiting now, what did it look like a year ago? That can be really valuable to engineers who are troubleshooting the equipment.

We also store a lot of metadata that’s all up here in our cloud platform. Since I’m on a test unit there’s not a lot to report here but if it were a piece of industrial equipment we’d brick all that in. So before you go to do service on a failed asset, exactly what parts you need and what tools to take with you. We put metadata in there about the location and asset tag information so exactly what asset you have to go out and service. All from one digital interface so you can go and find it really easily.

 Alright, so that’s our kind of the software portion there. You can see that these measurements continue to merge as we go. So if I update this to our current time, we’ll see a few additional ones have come in. I’m rounding off to five-minute centers but if we down-sample to a minute you’ll see each individual measurement come and go. And actually, if we go back to the graphs you’ll see individual measurements in there too right. So we can scrub in between and see exactly when I turn the hotplate on this morning at 10: 30. Pretty cold overnight

Our Next Sensor

 So one more thing I wanted to show you here actually in person this is a sneak peek of our next sensor, and I wanted to do this to give you kind of an idea of where we’ve been and where we’re headed. So this steam trap monitor. That’s our first gen product, it’s about the size of a deck of cards and that represented a breakthrough kind of the world’s first batteryless wireless sensor form factor, kind of large. About a year later, we developed our machine health monitor. So you can see kind of the change in size. This is more like a salt shaker in size. The primary control board that runs this that sits in here is much smaller like a two inch by two inch instead of this big deck of cards. This also brought enhanced compute capabilities with it. There’s a full system on chip in here, which means we can run and execute code remotely. So we can send software and firmware updates to this over the year without ever going back and flashing the firmware or something along those lines.

Our Gen 3 Technology 

 So gen 1, and them gen 2 about a year later. For the first time publicly today, I have an example of our gen 3 sensor with me. So I’m going to pull that out now and this is about the size of an Oreo cookie, maybe a little bit larger than that. It fits easily in the palm of your hand. So we’re bringing these to market next year hopefully. We’re working on prototypes of them now, you can see integrated solar cells here on this one. And once again just a really tiny form factor we’re working on a bunch of different sensing modalities that can connect to this device. The idea here is if you shrink the form factor these make them easier to deploy and increase the range. Each time we’ve bumped up a generation of our technology, we made the radios better too. So as these grow and expand you’re able to enable what we call pervasive sensing; placing these all over your facility to unlock new data streams things you’ve never thought about sensing before. We actually had an article that was written up in the Wall Street Journal where they talked about our technology as smart dust right. You sprinkle it out there and it reports back up to the cloud telling you precisely where you need to go. And like I said, unlocking new data streams that maybe you hadn’t thought about sensing before. So pretty neat. For one piece of comparison, I have a can of coke here so you can see exactly what the size of all these sensors are. It’s really really small which allows us to place them in a bunch of different places around facilities that occupying a lot of real estate and not make you run wires everywhere like traditional sensors would.

Alright, I am going to switch back to our slideshow here and talk a little bit about what’s coming next for us right. So we looked at our heat and light application. Once this data starts streaming you can use it as a reference point right at the machine. So you don’t have to bring a specialized tool with you to figure out if that motor you just walked past is behaving correctly or not. But we’ve also digitized it and since we’re connected to the cloud you could be at home in your kitchen working from home and see that same data. Or you could be away from work and home altogether and access it from anywhere in the world. So we think by digitizing these processes and making that data free to access all over you now have the ability to interact with this stuff regardless of where you’re located

That said, most of our customers are not spending a lot of time looking into these interfaces snooping around trying to find a problem. We have a complex analytics suite on our cloud platform that calls your attention to things that are failed. So you can get an email or a text message saying hey there’s a failure here this is what you need to focus on. I think this second order is really an important point of emphasis. So instead of thinking about inspecting these facilities, something where you have to go and get the data and then go and analyze it by taking that first tear off your plate, you never go and check on something that’s good ever again. So self-powered sensors sit out there like little sentries waiting for a failure to occur and then they call your attention to it and then you can respond to that input instead of having to go out and take those measurements to begin with. So it’s a subtle difference there if you think about it but it adds up to huge changes for our customers.

Cost Savings & Environmental Impacts

In fact, if we kind of look at this by the numbers we saved over a million dollars just in wasted steam alone for our customers using the steam trap monitor. So that’s a tremendous amount of savings across our fleet and that assumes that somebody was checking those steam traps before on a manual inspection route of some kind. The difference between finding them exactly when they’re failed and being able to respond much quicker versus waiting until the next time a manual inspection would have occurred. That’s what adds up to those huge savings. 

But there are ecological benefits too and we pride ourselves on those at this company. We’ve actually diverted over 110,000 tons of co2 from overproduction at the boiler level. So by fixing these steam traps sooner and not having all that steam blow into a condensate receiver tank or down a drain, we’re actually helping make industry a little greener. Finally, this may sound familiar from an ecological standpoint from what we talked about at the top. We’ve diverted a heck of a lot of batteries from going into landfills. Industrial sensors usually use batteries with heavy metals in them and they’re heavily packaged which makes them really difficult to recycle. Many of them don’t get recycled at all, so in the time we’ve been out there in the world we’ve headed off a lot of batteries from ever having to be put out there at the edge.

Energy Efficiency 

So kind of an interesting point here about energy efficiency to date. We’ve gathered over 424 million data points from measurements on our sensors. We talked about power efficiency; essentially these things work on like a leak by a current, a very small source of energy. But if you sum up all those data points we’ve measured across all of our sensors that sit out there at the edge, we summed that electricity together came back with a number of about a third of one kilowatt hour of electricity total across the entire fleet with those sensors at the edge. Which is just a tiny amount of energy. In fact, I’m going to give you a comparative here. If you were brewing a pot of coffee you’d use more energy than that just to brew one pot of coffee. So all that sensor data we’ve ever collected, the million plus dollars, and 110 000 tons of diverted co2 is all powered by less than a single pot of coffee on those sensors sitting at the edge. Pretty remarkable. So think about that next time you go and grab a cup of joe, really kind of a radical shift once you have these super high efficiency sensors. 

Future of Everactive

So thinking about the future last year we released that first generation sensor in that deck of cards form factor. It had about 30 meters of range between the sensor and the gateway and it needed about a 20 degree celsius delta for what it was sensing to power on. In a steam system, you’re always going to have that much temperature differential in steam so that’s not a a big deal there. Our second gen product which came out this summer that’s the machine health monitor that has the accelerometer in it. It does vibration analysis that’s a smaller form factor. Like the salt shaker, a huge step up in range on our second gen evernet radio there and a big step down in the amount of energy we consume in order to do that. Not just temperature there, we can also use light. Lux is the measurement of light. The room you’re in now no doubt probably has 200 lux in it, even a dimly lit room is probably enough to keep that sensor running. So that’s our second gen and kind of where we are today. 

That prototype that I snuck out and showed you there that’s on the desk that you can still see, that’s our third gen. It’s got our new radio in it. There’s several different applications we’re looking at for that, but kind of the big breakthrough there is in the range of the radio. We’re looking at about a kilometers worth of range. And the testing we’ve done so far with that radio off of a self-powered energy budget. Really remarkable to think about that we keep taking down the requirements of harvested energy and increasing the range. If you think about radios across industry, typically anytime you want to get better range you’re going to throw power at that problem. So for us to be able to bring down our power requirements and increase the range is truly something no one else in the world is doing. A pretty remarkable team of engineers we have here. So a couple minutes left, I do want to talk about the difference this is made for our customers. 

Customer Quotes 

We talked about the numbers’ impact but we also asked them for quotes about what they liked best and really love about our system one quote here from a reliability engineer at a plant in Indiana. This is a pet food plant he said we didn’t have to bother my IT  guy once and he’s exactly right. 95% of our installations run completely independent from customer IT. Our gateways have LTE cellular modems in them that can roam between all the major carriers in the u.s, the top four. And that means that most of the time the gateway, once you plug it in,  just starts streaming data right away. 

A maintenance leader at a chemical plant in Virginia said you made installation really quick and easy on me. And that’s true too; we think of this as a very light lift. You don’t require any specialized tools, you don’t have to be an IoT expert, we’re there with you to help with the install and you can usually get data flowing within a couple of minutes, from the time you unbox the sensors and put them on assets. We use thumb screws to put on the thermoelectric generator on the steam trap monitor, that’s about the trickiest part. The machine health monitor we can use magnets, so let’s stick right on 

Our Service Model 

Our service model removes all risk. This is from an innovation director for a CPG company and we have really tried to de-risk this in a couple of ways. For starters, we’re right there with you kind of on this journey. So every part of this solution is covered by us. None of it is third party where you go and source something from an external company and then try to integrate it together. We’re there from the questions prior to the sale through the installation process. And then afterwards continuing on your journey if you ever have a question or a feature request or need support that all comes from one company. 

You’ll never get caught kind of pointing fingers between companies. Which we see sometimes with complex IT integrations where one company owns the sensor, someone else owns the backhaul, someone else owns the analytics stack. And so if you need a change that affects all those things you’re waiting, sometimes years. But there’s another way that we’ve removed risk we’ve de-risked the financial aspects of this. 

Since we sell our monitoring as a service, if you see at any time that it’s not providing value for you you can simply discontinue monitoring and at that point we have to kind of pick up our ball and go home as I like to say. So that makes it much simpler for you. Some of these industrial sensors can be in excess of a thousand dollars per instrument, not to mention the fact that the batteries aren’t free. So you pay recurring costs for the right to go out and swap out those batteries and usually software maintenance costs as well. So our solution comes in kind of one flat fee that’s lesser than those combined and you don’t have to make that big upfront investment just to get started. You can start gathering data now, ensure that it’s valuable to you, and if it’s not you haven’t made this huge investment in something that could take years to pay off after you’ve seen that it’s not going to be a good fit for your environment. So those are the ways we’ve kind of surprised and delighted our customers. I’m just going to flip the camera back over with the one minute we have remaining here and kind of show you the whole solution from the top again. So sensors here, including this new form factor, that we’re bringing out in 2021. The gateway here on the side has a many to one relationship between these and the sensors we include as part of the service. So this is the one that’s actually live. You can see our green status light gathering the data from these sensors and shipping it up to our cloud. 

So thanks for geeking out with me here today in non-coveted times we’d love to do demonstrations like this on your equipment in your facility but we we are doing installs now and we are able to do some limited travel so please do reach out to our team we’d love to meet you and hear how we can bring this technology into your environment where we can leave make a change for the better for your environment as well.

Alright, well thanks again everyone coming in right on time here I’m glad I had my pot of coffee this morning. A lot to fit into a half hour, if you do have any additional questions please feel free to reach out to me. Or if you’re interested in pricing and kind of next steps please feel free to reach out to our sales team that’s sales@everactive.com. I’m part of that team as well so if you email that group you’ll get my attention there

 Okay we’re at time, thanks everybody. Appreciate it!