Expeditionary Power, Software-Defined Energy, and BKPK with Sascha Dainat

0:00Welcome to the Startup Defense. My name is Callye Keen. Today I have Sascha Dainat with Backpack. We are going to talk about expeditionary power, software-defined power, enabling new missions. What does it take to get into the battery and power business, which I have huge respect for, very difficult line to break into. Sascha, before we get started and we talk about power and the opportunities in defense, same question I ask everybody, what are you passionate about right know? I'm passionate for a while. I'm passionate about building systems that kind of solve old questions and old problems with newer technology.

0:38I don't like necessarily all the buzzwords, but if you throw in AI and ML and so on in there, they're only good if they help people to do something better, whether it's in a civil environment or in a government environment. Power is kind of the ultimate example of that, right? I think it is often not really thought about until nothing works. And I've worked with a lot of different power or clean energy or clean tech, all these different types of startups and different types of technology. It's interesting that when something clicks, so many other use cases or missions or applications or other pieces of technology are enabled, but it's often the last piece that people think about.

1:22Like if they're thinking about AI, they might be thinking about data centers, but they're not thinking about backup power or alternative power for data centers. You know, they're thinking about expeditionary systems. They're thinking about how cool drones are or sensor systems. They're not really thinking about, well, how do we extend that mission or make sure that it turns on or, you know, power is this fundamental horizontal layer of enablement across. And it's that boring bit that makes things work. What is your approach? How are you solving that? Is that informed by your background? First of all, power is boring until it isn't because a lot of things need power and an increasing amount of devices, whether it's in your private life or in the field, need power, electrical power.

2:14It comes from somewhere. Just to go back to your question, I'm a trained electrical engineer and telecommunications engineer. I liked high frequencies, distributed systems, and so on. That's why I went into telecommunications. The interesting part is that power is on the way to become ultra-distributed systems that are actually connected, and you get all the data and information from it. So that draws me personally into this. Apart from it's a comeback, I'm a third-generation German army kiddo. Did all the moves with my dad when I was young, then went into the German army for a short while and then went out and decided to go into business.

2:53And when I look at where we are in terms of power in the field, oftentimes it's not very systematic approach. It's oftentimes an afterthought. To give you an example, I went to a counter UAS exercise in preparation of the FIFA, the coming soccer world championship. And everyone kind of brought their own power and you ended up having small generators which were running in the midst of a field which is a fire danger and b they were impacting the sensors still the acoustic sensors then others brought batteries but batteries were really really heavy on tracks because they needed to have enough power others again brought battery solutions in pelican cases which are which are nice but they're missing some aspects of ruggedness all in all you didn't you never had a common picture of how good the power is, where do we run out of power or batteries, or where's the gas low, or what do we need to do to shut off generators completely, to have complete silence for watching out for drones.

3:53So once again, many times it's not really systematically looked at. And that's something that we do at Backpack. We started looking at power, not only from how can I take it with me, but also how can I get enough power out? and also how do we make it a systematic approach to power where we know at all times what's going on on the power level and be able to act before power runs out. I really like the approach that you've taken with modularity and then being able to add intelligence to the batteries. It's not just I have power, but I have an actual power system and I can learn what my usage is.

4:32You're really extending what the concept of delivering power at the edge could be. I find that really compelling. We want to work on these ultra critical systems and we want to do these very important things. And to me, again, it would be humorous if it wasn't so serious, is how often there's just an assumption that, yeah, we'll apply a genset or that power will be available. You see these multimillion dollar systems effectively plugged into 20, 30 year old power backup systems. Very little modern thought put into it. So yeah. If you look what's out there, power is tricky because you might have in a medium-sized camp, you might have HVACs.

5:17They pull quite some power. At the moment, we are solving it with generators, but even generators are choking sometimes when the whole team comes back from a mission and then everyone's switching on the HVAC systems, which means there's a huge spike in power all at once, which can bring down your power network pretty easily. On the other side, those generators need to be transported in with heavy equipment. They are so heavy, nobody can carry them. So the big question is how do you strike the balance between having a single unit that you can carry, still putting out enough power, and then putting it together.

5:53And you touched on modularity and scalability. That's one of the fundamental principles next to looking at it from a systematic level is modularity and scalability. We wanted to design a system that works on, let's say, an M-Razor. And you can strap it on the M-Razor, you can take it on a small mission, but you can also come back to your camp and then forward deploy camp. And then you have much many more battery packs, which can be then recharged and prepared for the next mission or run your HVAC system overnight when everyone wants to sleep and wants to have silence. And then even checking it through to the third level is if it's a fixed installation or a fixed base, then how do you secure communication nodes when there's a grid outage?

6:36So we wanted to have a system that's flexible enough to be used in all of those cases. I have this suspicion less startups go into the power market because of safety and compliance, just the cost and difficulty of addressing this problem. Can you speak about overcoming those challenges and bringing a product to market? Because you also have a dual-use product. So you have all of the commercial compliance issues and you have all of the potential compliance and interface issues that you would have with Fence. So you have quite a stack of engineering problems to meet. I think that's impressive. But if you can speak to that and give some advice for making a highly compliant, difficult product, I think that would be really helpful for everybody that's listening.

7:24Yeah. So first of all, as a startup, it's always focus, focus, focus, right? Because we could go in 30 different markets with batteries that type, but we decided to go into two. And the deployed market is really the one with the emergency side and with the defense side that is rolling out right now. And we have success already with Californian wildland firefighters or emergency teams in Poland that face very similar issues to the defense side. And that's why it's a due use product. To give you one example of the compliance, you can only transport so and so much lithium on a common cargo plane.

8:02And we designed the battery case, what we call a pack, which has four kilowatt hours. We designed it in a way that it has enough power to be meaningful, but it can't get too heavy to go above and beyond. It's about 35 kilograms, about 75, 77 pounds that you can transport on a regular cargo machine. So we designed the pack to that limit so that when emergency responders, they go out when there's an earthquake, they fly out. Often they fly out in military and other machines where compliance is less of an issue going out. But on the way back, they have to abandon the equipment because it's hard to get it back and you can't just order a flight on a DHL cargo flight and fly it back.

8:47So with our equipment, now we just recently reached, and insiders will know about it, the UN38.3 certification, which allows us to put our battery packs on cargo planes and just fly them out, which is easier for delivery if something needs to go quickly. and then also easier for the teams in the field that need to kind of fly it back and transport it back and need to comply with all these guidelines for lithium on planes. Just for completeness, I will say we are working with LFP because it's not as violently burning as the siblings, the lithium ion batteries. So that which is another compliance aspect which we designed in the product.

9:24There's many, many others that we looked at. Do you see partnership as a go-to-market strategy? package Backpack as an enabler with a kit? Or how do you go about breaking into defense with such a broadly applicable product? Collaboration partnership is the only way to go because power is not the primary idea and mission that you want to complete. There is another mission, right? Do you want to save lives? Do you want to defend or you want to solve a problem or find people under a thick layer of rubble when there was an earthquake? So to do that, you need other devices, right? whether it's drones that you fly out and survey the area, which have heat cameras, and then you would see whether there's someone below rubble.

10:08And that's the primary idea that you want to accomplish. But then for that, you need the power. And that's where we come in often in discussions. Just to use that drone example, we have a partner that specializes in a drone ground control station or unmanned vehicle ground control stations. And the video processing power has been really going up and is amazing. Now we can get much better sight of what's in front of the unmanned vehicle or around the unmanned vehicle. But it also means that the ground control station needs much more power, let alone being the radios, the computer. So the mill tops or laptops, depending on which environment you are, whether it's civil or the defense environment.

10:50Many more components around that typically that you need for command and control and other use cases that need to be powered in that moment. And what's coming up more and more is where you had in the past rather static setups. Now it's and we see it from the modern warfare and the impact of Ukraine and other places where we are active. We see that it's much smaller setup and much more flexible. So it needs to be packed up again and then moved quickly to the other one, the other side, and then set up again and be available in an instant. So that comes into that as well. I want to pick on the AI or ML portion of this because it is, one, it's the hot topic, right?

11:34So what people are talking about. But two, I think this might be a little abstract for our listeners. Like how do you add software or AI or machine learning to PowerPacks? I think more importantly for me, what does that let you do, right? What does that actually enable you to do is a much better way of putting that question. That's a question that we get very often because it's not immediately visible where AI or other intelligence comes in. It happens on multiple levels. One is if you go very, very granular, each cell, even in your existing equipment in your phone that you have at home and so on, Each battery cell is slightly different and has variations in which it's charging or discharging.

12:19And also it has a specific charging cycle. Now, imagine a pack which has many cells and they are all acting slightly different. And you need to keep having an idea of how they are charging, what's their state of charge, how secure are they? Do we need to switch them off because they get too hot? Or similar, and that leads back to compliance, to similar use cases where you need that ruggedness and the safety first because you don't want to impact people, but also you want to get the maximum output of the batteries. Then on a much higher level, imagine that those batteries would be deployed and distributed in a camp, let's say, five or 10 with one setup here to power an HVAC system or to work in a hybrid scenario with a generator.

13:04Another set of 15 systems there at the medical tent, packs are being swapped through to power that medical setup and keep people alive in the end. And you want to have a complete view on what it looks like, first of all, before you go on the mission. And there we are investing a lot in the software side and developed a tool which we call Orchestrator that allows to plan missions and simulate power and power flow up front. Then during the mission, you want to make sure that the area which needs the most power and most urgently, you give it the priority. So there's a lot of logic and so on to arbitrage power.

13:41Obviously, power requires still mostly a line. So rarely power is transmitted through the air. It can be. but in those setups it's mostly fixed but you can see in the system which one is using the most and then you can juggle it there's some still some parts in development and at a later stage we also be able to push configuration from the orchestrator into the devices that are distributed and then after the fact when you went went back home you're you're safe and sound you want to debrief you want to know what what has happened how many devices connected which one pulled the most power? How many times did the VAC system switch on?

14:19So you want to have all that information. And the other thing that's crucial is the equipment is used in different environments. I'm living in Arizona. It's very sunny, hot, and very arid. So we don't get a lot of rain here. So it needs to work in this environment and the system needs to understand in which environment it is. And then if you go into the Pacific area where it's very humid and you need to deal with the humidity, but it can be also hot. Or if you talk about the northern front and you see that it's very cold, the system needs to intelligently adapt to its environment. And that's where we use artificial intelligence and machine learning and other ways of algorithm and human intelligence that we put into the software.

15:03There's an interesting opportunity for security here, I think as well, because if we have another source of truth of what devices have connected to a system, we can then understand, was there shadow IT or was there other devices that are not permitted or not part of that plan then connected to that power system, which I think is interesting. So borrowing a concept that we all kind of understand from networking, right? Being able to understand what's on your network, to be able to understand what is on your power system is very intriguing from a utilization and from a security standpoint. At least I think so.

15:45That's where my head goes right away. I met a gentleman from the Marines recently, and he had data in Morse code. He will smile now if he ever listens to this, but he had data in Morse code on the slides. So data is almost more valuable than the actual physical stuff in the meantime, because you need to have that information on the system. But it also goes in two ways, right? So you mentioned security and cybersecurity. It starts with the physical aspects. So we've, for example, designed our own plugs, which carry data and power from the packs. So what we call the dock, the dock is an inverter converter system.

16:22And we need to make sure that nobody can easily slip in a chip or something else in between the connectors and sniff out whatever is going on the data pipe. So that's one thing that we are looking at. The other thing, and we are in some of the early stages of the compliance process in that, but we are looking at if we are developing something like this, we A, want to integrate with existing systems like ATAC and similar. Also there, we are slightly early, but coming back to your partner question, that's an area where we definitely partner with others and set on something that is existing so that we don't introduce another thing, another new thing, which wouldn't be used, but rather set on common platforms.

17:04And then in that light, we need to clarify the question is, how do you collect the data? Where does the data sit? Who has access to it before, during and after the mission? And who owns it in the end? Where we are really open to collaborate and we're collaborating already with different people to solve these kind of questions when we go into data and data security. I normally ask what's next for a startup because all the founders are kind of living in the future, right? You're already saying, hey, you know, we're early in this. We're thinking about this. We're thinking about this, which I love.

17:36This is why I love working with entrepreneurs. But there is just this infinite roadmap of where Backpack can go. What are the next big steps that you're excited for? You're like, this is a big milestone. You're prototyping. You're scaling up production. What's the next like big, massive thing for Backpack? What's special about Backpack is that we thought about mass production from the beginning, which I know it's listening to your other episodes is one of your pet peeves is to not only develop it one time, but also have scalability in mind and have it on the rise and how you go from one to 10, validate it, go to 100 and go to thousands.

18:15And that's pretty unique about Backpack that we thought about it from the beginning so that we at some point can mass produce it within this year and spit out hundreds and thousands of packs, which are all the same quality and the same standards and they are rugged and you can dump them in the mud and they are still functioning. And that has been considered and designed into Backpack. What's next is really, and we're in the process of doing that, is the overall validation, one on the civil space. So it's getting into the hands of, as I mentioned earlier, wildland firefighters who prevent fires by controlled burns.

18:50And then they fight fires when they happen. And then after the fire, they make sure that the fire is really off and they're going to test it and really go through it. A similar thing we are searching for and partner opportunities in the defense space now that our products are becoming available, which is always the exciting point in the journey of a startup is when they really come out and people can use them and test them. and so on, is that validation with the scale in mind. So if we are testing it now, what's the purpose behind it? And where do we need to go afterwards when it goes into, oh, we need hundreds of those or thousands of those?

19:27Sascha, I really appreciate you taking the time to be on the Startup Defense. Kali, it's a pleasure working with you overall and enjoy talking to you. My name is Kali Keen, and this has been the Startup Defense.