MachinePix Weekly #5: Sanjay Dastoor, founder of Boosted Boards
An interview on the scourge of counterfeit batteries. Posts about a demolition and a ballpoint pen were neck-and-neck in popularity this week.
This week I sit down with Sanjay Dastoor, the founder of Boosted and Skip—both pioneering companies in light electric vehicles, and talk about challenges in working with batteries in the early days of LEVs.
The most popular post last week was the demolition of a grain elevator. As always, the entire week’s breakdown is below the interview.
I’m always looking for interesting people to interview, have anyone in mind?
- Kane
Interview with Sanjay Dastoor
You started your career in aerospace and robotics, what was the inspiration for starting Boosted?
My technical background was an interest in robotics. I always thought I would go into robotics R&D and spent time at JPL (Jet Propulsion Lab) and SRI. I also did a stint at a machine tool manufacturing company that made honing machines for engine blocks. With Boosted, my two cofounders independently came up with the idea, and I was third to the party. My engineering classmate Matt loved motorcycles and snowboarding, and wanted to combine the two. Meanwhile, in our graduate research lab, we were using brushless motors and high power lipo (lithium polymer) packs for robotics projects. These components had just become available in hobby shops. My labmate John had the idea to put them on his longboard to get around campus more easily. It all started as a side project, something we wanted for ourselves.
Electric vehicles have come a long way since you started Boosted—what were things like in 2011 for people working on EVs, especially light EVs?
In some ways it was really exciting, because some stuff had just become possible because of the hobby industry. In 2005, if you wanted the highest performance RC airplane, you were using a gas engine. By 2011, you were using electricity and brushless motors and lipo packs. I mean, they weren’t very safe then—but they were so much better than nickel metal hydride batteries and brushed motors.
These were all hobby parts though. Building an extremely reliable battery pack for Boosted was a huge challenge. It wasn’t obvious then that electric vehicles would be a thing at all, and even tougher for light electric vehicles. We really had to convince people to care about reliability for small vehicles: operating temperatures, fail safes, etc. We’d call battery companies and ask if we could buy their cells and they wouldn’t even return our calls. They weren’t really thinking of vehicles as a market.
In addition, there were a lot of counterfeit cells - if you were buying off of eBay or Alibaba you had no way of knowing if you were getting the real thing or something dangerous.
Tell more more about the counterfeit batteries
Many of the reputable vendors don’t want to sell to small companies and deal with liability. If we used a Panasonic cell, and we sold 100 units of something, and they caught on fire because of something we did, Panasonic would still be associated with us. They didn’t want to take that risk with small companies in new markets.
There were vendors that would sell to small companies, but they could be selling cells that couldn’t pass testing, or straight up counterfeiting. It was impossible to know until you received them. And even then it could be hard to tell if it was safe. It’s a lot better now.
(Kane’s note: I’ve had similar experiences with batteries)
What are some misconceptions people have about electric vehicles?
I think the big thing for us, and this is true for other vehicles although I only have experience with sub-5kW vehicles, but roughly if you think about a gas engine, you’re trying to control an explosion in a useful way. With an electric vehicle, you’re also trying to control a potentially very violent chemical reaction and apply it to a useful mechanical problem like getting up a hill.
With electric vehicles, in order to go up a hill, or regenerative brake, the possibility that the energy gets released the wrong way is quite high. It’s also tough because the bar is much much higher today. If you see a battery fire, it gets a lot of press and attention. This is a good thing. But think about all the car fires and gas explosions when we were building internal combustion cars. We’ve developed technologies like self-sealing tanks. We’re still early with EVs, and we’re held to a higher standard now.
Building an electric vehicle is about working with a team that respects all the corner cases and effectively builds around those. Any battery that can move you for a hundred miles can also burn your house down, just like a tank of gas is like a bomb. In the same way you think about software, and how much real effort goes into designing good interfaces, the same thing exists with discharging power from a battery. Whether in motor control or the battery management, there’s a lot of engineering challenges in order to make things feel effortless.
The other part is something we felt very acutely early on, and we were lucky to have a team very focused on this—if you think about consumer products, there are some products that you can get hurt seriously on if they’re not built well. EVs are challenging in the sense that if you put an EV out there, if someone gets hurt on it, you have to think about what your responsibility is. Take a Square Reader, if you fail as an engineer, it doesn’t swipe correctly. Maybe you damage the phone. With a vehicle the burden is much higher.
How can people learn more about EVs?
There are great communities online. For light electric vehicles, Endless Sphere is a place where a really helpful group of people are posting about their random projects. They’re really pushing the envelope of performance and providing a lot of write ups. Debugging, modifying, customizing - it’s great.
What are the craziest stories you’re allowed to share from running Boosted or Skip?
Initially, we were going to launch with a pre-order campaign for Boosted. One of our investors told us to just build a few and sell them. We were convinced we needed something professional: tooled plastics, professional assembly, etc. Our first boards ended up using a third party Memorex Nintendo Wii controller paired to an Arduino. The idea that you could sell something with an Arduino with it is wild. We used a remote controlled car controller and motors since RC airplane motors didn’t let you adjust speed. We laser cut the rest. That’s how we built the very first boards.
When we were shipping them, we realized it was completely legal for us to ship a potentially very dangerous battery. The requirements for safety and reliability just didn’t exist. When we shipped we had to create our own safety standards because we cared. Ironically, all the hoverboard fires happened a few years later. You couldn’t track any manufacturing. Amazon and Best Buy were pulling them off their shelves.
There’s a lot of ways electrical systems could go wrong. We’d gotten invited to do a demo at TED while we were still prototyping, so we 3D printed a plastic shell for our prototype battery. At the very last minute, one of my colleagues mentioned we should have a way to cut power to the battery without disassembling it since this wasn’t a fully debugged system. We were giving a demo, and the board started to smoke - it was subtle, most people wouldn’t notice it, but we recognized the smell and casually pulled the key while still talking to all these celebrities. If the key wasn’t there it probably would have burned up. You’d be surprised by how chaotic a lot of tech demos go (Kane’s note: it’s true! There’s a great article on the chaos of the first iPhone keynote).
Even in production, we once had a battery that started smoking. No one was hurt and there were no fires, but the lengths we went to find the root cause was insane. The user had thrown the board away in a dumpster, on the way to work. We went to Google Street View and called every business on his commute to help us find the board in the dumpster.
Any cool projects you’re exploring now?
Side projects for me, which is less around hardware development, but more around governance systems - I think a lot about how I can be more knowledgeable and involved in local politics. I’ve been reading a lot of writing by DJ Patil and Matt Cutts who started in tech and worked for the Obama administration.
I’ve also been tinkering with fun little connectivity projects - I put a Particle cellular module in my Boosted and could remotely track it and turn it on/off.
What’s your favorite simple (or not so simple) tool that you think is under-appreciated?
A butane soldering iron. It’s actually portable, no electricity needed, and heats up very fast. I discovered it doing repairs for Boosted customers. For a bench-top, I still use an electric one but it’s the best for projects in the field.
The Week in Review
August 14, 2020: Time-lapse of assembling a wind turbine at the Cerro Gordo Wind Farm in Iowa.
These are NEG Micon (now Vestas) 750KW turbines according to the farm’s project manager. I love giant turbines—everything about them from testing to transport just looks epic. Also, Cerro Gordo just means “fat hill”.
Manufacturer: Vestas
August 13, 2020: USS Abraham Lincoln Nimitz-class aircraft carrier performs a high speed turn.
Nimitz-class carriers tip over 100,000 tons, so the amount of force generated during these evasive maneuvers is incredible. It’s driven by two A4W nuclear reactors which each provide 104MW of drive power as well as 100MW of electricity. My back-of-envelope math shows ~278,933 horsepower for the maneuver 🤯
Manufacturer: Newport News Shipbuilding
August 12, 2020: Demolishing a grain elevator with a Hitachi ZX450LC excavator in Saskatoon, Saskatchewan.
This was an old Parrish & Heimbecker elevator. In researching this I discovered a technique for felling large structures called the Dibnah technique, which does not require explosives.
Manufacturer: Hitachi Construction
August 11, 2020: Switchable glass public bathrooms in Tokyo by the Nippon Foundation, via @kydeanderic.
These are in Yo Yogi Fukamachi mini park and the Haru-no-ogawa community park, and actually designed by Pritzker Prize winning architect Shigeru Ban! He wanted users to be able to tell the cleanliness and occupancy of a public bathroom at a glance. I can’t help but think there aren’t many countries outside of Japan where this would work 💩
Manufacturer: Shigeru Ban Architects
August 10, 2020: Closeup of a ballpoint pen by @macrofying.
I’ll be honest, I did not expect this to be the second most popular post this week (after the grain elevator demolition). But! Since people liked it so much, I have a few cool links to share. I’ve always been obsessed with ball bearings (long story), and ballpoint pens are an incredible example of cost engineering. They used to be super fancy and hard to make—Smithsonian has a great history of the ballpoint pen and how László Bíró changed the industry. China wasn’t actually able to manufacture ballpoints domestically until 2017! And yes, they actually have a little ball in the tip.
Manufacturer: Unknown
Postscript
I just bought a Brother PE800 CNC embroidery machine. It’s amazing the kinds of tools that are available to hobbyists these days 🧵
If you enjoyed this newsletter, forward it to friends (or interesting enemies). I am always looking to connect with interesting people and learn about interesting machines—reach out!
- Kane