Cracking the Tesla Battery Module BMS

Last episode we talked about Tesla batteries and that seemed to get a good response and we've had a lot of activity here since then. Let's start generally and then try to get more specific. Elon Musk and Tesla are now in operation with the world's largest battery factory and I want to put that in scale.

Last year there was about 35 gigawatt hours of battery storage produced in the world by all the players combined. The Chinese, Samsung, Panasonic, which now I think owns Samsung, LG, Kim and everybody else. The Tesla Gigafactory when it reaches full production will do about 35 gigawatt hours a year, essentially doubling the production of lithium-ion batteries in the world, in the largest plant in the world, in the building with the largest footprint in the world and one of the largest photovoltaic arrays on the roof in the world.

Let's talk about the Tesla battery pack just a little bit. I've got here a 60 kilowatt pack up top and over here our 85 kilowatt pack that we just got this past week. 85 kilowatt pack is made up of 16 modules.

This is one of the modules. Each module weighs 25.4 kilograms or 56 pounds and it's made up of six series cells. Each series cell made up of 74 18650 battery cells, an NCR 18650GAB I think, about 3400 milliamp hours or 3.4 amp hours, giving the module about 255 amp hours at 0.21 volts.

These are nominally 3.6 volts. You charge them to 4.2, 3.4 amp hours per cell, a little bigger than a AA cell and that's what makes up the battery. Each of these cells is 48 grams, giving it about 3.552 kilograms for each of the six cells and that's, oh I've got it here somewhere, 21.32 kilograms.

Of the 24 kilograms making up this module are these batteries. Each battery is about 12.24 watt hours and so the module contains 5,434.5 watt hours in theory. We actually weighed the 85 kilowatt pack and it came in at 1330 pounds.

We have calculated the 60 kilowatt at about 1150. The 60 kilowatt modules, 60 kilowatt battery only has 14 modules. Oddly, we found that each module has 60 less cells, 10 less for each of the six series segments and so that's our 60 kilowatt pack.

Our total active cell weight in the 85 kilowatt thing, our total weight of 1330 pounds is 603 kilograms and of that 406 kilograms is the module and 341 kilograms is the battery cells. And so the actual cells providing the energy only comprise about 55% of the weight of the pack. As I've said a number of times, they put so much engineering into enclosing the cells and modules, BMS, and then this package that they give up a great deal of the advantage of the higher energy density of these cells, about 250 watt hours per liter.

By the time they get down to it, we're looking at 214 watt hours per liter for the module and about 201 watt hours per kilogram, not liter, per kilogram of the pack itself. So they give up a great deal on the energy capacity of these batteries per weight to packaging weight, structural weight of the pack itself. Anything that could be done to reduce that would be a big deal.

Ion has announced that their battery factory is going to make a new cell, a 21700, that is 21 millimeters in diameter and 70 millimeters long. This is a Sanyo 2700, 20 millimeter in diameter and 70 millimeters long. And so you can see kind of the difference here in the size of the two cells, very little actually.

I used to make whiskey. I was amazed what different product I could get by using different size barrels for different time periods in the barrel aging process. And it has to do with sort of some amazing elements of the volume of a cylinder and the unit volume of the cylinder and the surface area of the barrel.

We have a similar situation here. We make it three millimeters larger and five millimeters longer. We increase the volume of the cylinder from 16,540 cubic millimeters to 24,245 cubic millimeters.

That's an increase of 46%. And so we've increased the volume of this battery 46% by making it three millimeters larger in diameter and five millimeters longer. And that's the secret to their selection of 21700 as being a significant gain for the battery.

We also believe that the cell they're going to make will be 75 grams and these are 48 grams. There being a thousand grams in a kilogram. What's amazing is Rumor Control Central would tend to have us believe that they're going to get 5,750 milliamp hours out of these new Tesla cells.

Now part of that is the increase in volume, but sort of simultaneously this is the period where Panasonic is going to an increased level of silicon embedded in the graphite of the anode. Silicon is much better at lithiating or intercalating lithium ions than graphite is. The problem is you can get a three to four hundred percent physical expansion in silicon when you do that.

Silicon being somewhat glass-like, it breaks into a million pieces. What Panasonic has done is go ahead and break it into a million pieces and then mix it in with a binder with the normal graphite. And so the individual little pieces of silicon can swell, but they're kind of buffered by the graphite anode material around them.

And that increases, is one incremental way to increase the energy capacity of the cell. So that's kind of where we're going with the new cells. The bottom line is they'll be able to put 45 of these in place of 74 of these and get the same capacity.

But drastically simplifying their package here, hopefully reducing some of the weight of the package and to get the same energy out of the cell. And those cells will primarily be used in the Generation 3 initially. There was a Reuters report this week that the Generation 3 was going to be going into production in low numbers in February 20th, which is a couple days from now.

It's not true. They're making some mules, test mules this month, but they're not production thing. They're still waiting on a huge press and some press tooling to stamp out the Gen 3 parts.

Elon Musk is swearing they're aimed at the 1st of July. He has never made a date. And this is actually a management technique.

You have to give your people an aggressive schedule for them to miss if you're going to have a moderate schedule for them to hit. And so we're probably looking at September, October to start producing Gen 3s. Again, they're talking 100,000 units by the end of the year.

Historically, they announce something in the summer, actually get started in September, October and have already made 2,000 of them by the end of the year. But the Gen 3 is remarkably imminent, given the announcement and the pace. And so I have reservations for two of them.

I'm hoping for something in 2018, but maybe the end of this year. That would certainly be nice. Along the way, we have past reported on a guy named Jeffrey Dan.

Jeffrey Dan is a battery materials researcher, kind of part of the team that developed the nickel cobalt manganese oxide or NCM cell cathode material. And we had previously reported some of his work at measuring round trip efficiency of charge to discharge as a way of detecting age and batteries. I don't know how that ever came out.

But he's kind of moved on. And a couple years ago, he went out to Tesla and proposed that they fund research into an improved battery. Tesla has no real partnerships or formal partnerships with universities, but they did enter into one with the University of Dalhousie and Jeffrey Dan.

Let's take a look. I've got a little film clip here of Dr. Dan. I thought I'd start with just a few words about myself.

I'm from the Lunenberg area. And I came to Dalhousie to do my bachelor's degree. Then I went to UBC to do my graduate work at University of British Columbia.

And I focused on materials that could have application in advanced batteries. At that time, lithium batteries, and then later on, lithium ion batteries. And I've been involved since 1978 in this whole process.

And I've watched the lithium ion battery go from an idea to a reality. And without the lithium ion battery, you wouldn't have smartphones, you wouldn't have tablets, and you wouldn't have electric vehicles coming along. And if you read the bio of me in the program, you'll learn that I've been a co-inventor on a lot of different patents.

And patents are funny things. I think one out of 100 patents actually has value. The rest of them are just junk.

But one of the patents that we worked on in the year 2000, a postdoc, Zonghua Liu and I, invented a class of materials called NMC. And NMC stands for lithium nickel manganese cobalt oxide. And it's now used as a positive electrode material in lithium ion batteries at a rate of about 6,000 metric tons per year.

So invented at Dalhousie University. And that means about 1 in 10 lithium ion cells has material invented here within it. So I'm very proud of that.

Okay, so on to the main topic of the talk. You know, I'm showing you the picture of the Tesla Model S here. It's an awesome electric car powered by lithium ion batteries.

It was Motor Trend's car of the year in 2013. It has a driving range of 425 kilometers before it needs to be charged. And it really is the first electric vehicle ever produced to show that electric vehicles could be functional and green and yet amazingly cool and awesome.

Of course, we need electrified transportation because we can't keep burning fossil fuels. If we do, global warming is going to get more and more out of control. And that's really, really bad.

So Tesla now is coming out in 2017 with what they call the Model 3. And the Model 3 is meant to be an affordable electric vehicle for the masses. Nobody knows what the Model 3 is going to look like yet or what its specifications really are. But on March 31, Tesla will unveil the Model 3 in Palo Alto, California.

And we'll be able to see what it looks like and how far it will go on a charge. But Tesla's very optimistic about the future of the Model 3 and also its other electric vehicles. And to meet the demand that they see coming, in 2014 they started constructing this factory called the Gigafactory.

The Gigafactory is located outside Reno, Nevada. And this is a photograph taken in the construction phase. It turns out to be the building with the largest square area footprint on the planet.

It's a massive thing. And if you look at the rooftop there, you'll see it's very shiny. And that's because there's solar panels on the roof.

The entire factory is powered by renewables, which really shows Tesla's commitment to renewable energy and electrified transportation. This factory will produce enough lithium-ion batteries for 400,000 electric vehicles per year. And the way that lithium-ion battery production is measured is in gigawatt hours of lithium-ion cells per year.

So what does that mean? If you took all the lithium-ion cells the factory produced in one year and charged them, they would store 35 gigawatt hours of electrical energy. To put that into perspective, in 2015 every lithium-ion cell produced on the planet by all the producers in the entire world represented a production of 35 gigawatt hours of lithium-ion cells. So the Gigafactory will double lithium-ion cell production.

At the moment pretty much all lithium-ion batteries are made in Asia, Korea, Japan, China. So this Gigafactory coming to North America and doubling world production is a huge deal. And for me, involved with the lithium-ion business from the beginning, I just said, you know, listen, somehow I got to be a part of this, you know, because having this manufacturing capability in North America means a lot.

So I traveled to Tesla in the summer of 2014 and I met with their battery group there. And I talked about, you know, how about we try to set up a research partnership where we can bring our talents to your problems. And maybe we can help improve the lithium-ion cells that you use in the vehicles and also in your energy storage products that I'll talk about in a minute or two.

So after some consideration, Tesla decided that they would enter a research partnership with us here at Dalhousie. And in June of 2015, JB Strobel, shown with me looking into the cockpit of a Model S, he traveled here to Dal to sign the research partnership agreements. And JB Strobel is the CTO and the co-founder of Tesla.

So he's like number two in the company. And this is an article taken from the online version of Fortune magazine where I'm called a new weapon. So hopefully I'm not a weapon of mass destruction.

Hopefully I'm a good weapon. Anyhow, what are we going to do? How are we going to try to improve Tesla's projects? Like what's this research partnership all about? So I'll tell you in a second. Tesla also makes the Powerwall.

And this is a photo of two Powerwalls side-by-side. Each one will store about 10 kilowatt hours of electrical energy. And 10 kilowatts of electrical energy would pretty much power your house without a problem overnight.

And if these are charged by solar panels or by small windmills during the day or when the wind is blowing, at night when there's no sun or when there's no wind, they're discharged to power the house. And the Powerwall can be charged and discharged once per day and it should last around 10 years. So this is where the research partnership starts to become clear because when you purchase one of these Powerwalls, it costs money and it's going to last 10 years.

Well, what if it lasted 20 years or 30 years or 40 years before you would need to replace it? That would be much, much better, right? And this is where the problem comes because it's pretty hard to design a lithium-ion battery that lasts a long time. But it's even harder to prove that it will. If you want to show that a lithium-ion cell will last for 10 years under real conditions, you have to test it for 10 years.

And if you want to show it's going to last now 20 with the improved chemistry that you bring to bear, you have to test it for 20 years, right? So our research group in 2008 started developing advanced diagnostic methods to detect the very low levels of unwanted parasitic reactions that occur in lithium-ion cells that ultimately lead to their failure. And it's these methods that we've perfected from 2008 until now that interest companies like Tesla so that we can study new chemistries and learn in a short period of time whether the lifetime of the cells will be dramatically improved or not. So we have to stop burning fossil fuels.

It's as simple as that. And renewables are the way to go. Solar, at the moment, if you buy a solar photovoltaic panel with a lifetime of 30 years, the electricity it produces today will cost you the same as electricity from coal.

If you erect a big wind tower, windmill, and it operates for 30 years, the electricity from that windmill will be roughly equivalent in cost to the electricity from coal. The problem with the wind and the sun, of course, is they're not there all the time, right? It's not sunny at night, and there's no wind on a calm day. So to implement renewables at a large scale on the grid, you've got to have energy storage, however you're going to do it.

You know, if you're heating your home, you can store energy as heat. But for certain things, you need it as electricity. And the idea is you would build gigantic power walls, and Tesla does do this, and so do other companies, where you can store the energy from solar and wind.

At the moment, if the power wall were to last 10 years, storing the energy and re-delivering it would about double the cost, okay? So solar plus the storage would be double than solar alone. So the idea is by extending the lifetime of the lithium ion cells from 10 to 20 to 30 to 40, or however long we can ultimately make them last, the cost will come down proportionately, and that's a good thing, okay? So that's why we're in this. And I want to leave you with a really sobering note here, which is, okay, we have to do this.

And wind and solar are established technologies that costs are quite good at the moment. Let's say that we wanted to power Nova Scotia with wind. We direct enough windmills, and we'd also have to put in enough storage to be able to store electricity for a period of time when the wind is not there.

So let's say we want to store electricity for one day, to power Nova Scotia for one day, okay? How much would we need to store? Well, Nova Scotia power generates 2.3 gigawatts continuously of electrical power. If we want to store 24 hours worth, 2.3 gigawatts times 24 hours equals 55 gigawatt hours we need to store to power Nova Scotia for one day. If you remember back to my previous slide, the Tesla Gigafactory, which will be half the world productions of lithium ion cells, makes 35 gigawatt hours of lithium ion cells in a year.

Not even enough to power Nova Scotia for one day, okay? So the scale of energy storage that we need for renewables is almost beyond comprehension. It's massive. And yet, you know, people like Elon Musk, who's the CEO of Tesla, recognizes the importance and has the guts and the passion to go ahead and build this giant Gigafactory and get people to buy into this idea.

You know, the man is truly a visionary. He's been ranked alongside Bill Gates and Steve Jobs as one of the most visionary guys, you know, in this era. And I must tell you that Tesla has only one university partnership in the world, and it's with us, okay? And I feel incredibly thrilled and incredibly honored to have the opportunity to work with a company like that.

Thank you very much. Now, what's interesting about that, aside from his 15-minute TED Talk commercial about how great Dr. Jeffrey Dan is, and now he's a university, it was almost an information-free presentation, if you noticed. He didn't talk anything about what he's actually doing to improve batteries, but it is actually somewhat consequential.

He is talking about replacing ethylene carbonate as one of the solvents used in the electrolyte for the batteries. Now, why is that important? Lithium, of course, we have to have a non-aqueous solution, and what we want in that is some lithium ions, and we use typically LiPF6 lithium carbonate salts. And that's essentially our electrolyte, but it has to be in an organic solvent to be non-aqueous.

I've described this in the past. There's four or five major ones, which devolve to carbonic acid, ethylene glycol, and the biggest percentage is ethylene carbonate, EC. And that is the one that we know when we damage batteries, that makes everything have this Swedish smell like pears.

It's also the limiting factor in the voltage we can charge the cells to. Dr. Dan, in October, published a paper about how to remove EC from the solvent mix, replace it with a couple of substances that have the same effect on the passivation layer, a necessary effect performed by the ethylene carbonate, and remove the ethylene carbonate. What that ultimately promises to do is to make a battery that we can charge to a higher level.

For example, instead of having a 3.6 volt battery charged to 4.2 volts, we might wind up with a 4.4 volt battery that we charge to 4.7 volts. And that can be huge, even if it was a nominal 4.2 charged to 4.4 or 4.6. That would be huge. Understand that one of the main ways that we do better with these NCM cells is they're 3.6 volts.

Our lithium iron phosphate cells are 3.2. And so if we can get that up to 4, you would get again, the same gain that you get from going from a lithium iron phosphate to an NCM cell, again going to the new cell. So Dr. Dan's work is significant. And it is funded by Tesla.

So we're not to the ultimate battery cell by a lot. Musk is talking about $190 a kilowatt hour, and going in in the first or second year of the gigafactory to $100 a kilowatt hour, making a Tesla battery pack $8,500 instead of $16,000 cost to Tesla to make the car. Of course, the Gen 3 pack will be smaller.

I have heard that they've tested up to 70 kilowatt hour packs as an option in the Gen 3. Musk has said 100 kilowatt pack is probably not happening based on footprint. So let's talk a little bit about our 60 kilowatt pack. I'm going to sit down here at the laptop a little bit.

And we're applying 12 volts here to the input. And we have one of our little EVTV Dewey's hooked up to the CAN lines in this connector to make that connection. Again, I have improved our diagram a little bit, but I'll put that up on the screen so you can see how we're connecting to it.

And I'm running a little program that simply decodes the 6F2 CAN message. A guy named Andrew Heber kind of decoded that CAN message. And I wrote a software program then on the Arduino that displays the data.

And you can get that device and plug it into the console adapter of any Tesla. And it'll work, but we're adapting it for this to read our battery as it's here. Okay, here we go.

We've got our laptop connected to the Tesla monitoring program. And that's to the CAN wires on XO35, I think, one of our battery connectors. And as you can see, we're getting cell voltages through cell 84.

And then the last two rows have zero. Well, that's because those two modules aren't there. And so we, for some reason, do the module voltages.

And again, module 15 and 16 are zero volts. And here we have our temperatures. And this system resets every few seconds, but it is providing us.

And we have checked this. These are accurate temperatures and accurate voltages for the 60 kW pack. And so we get them.

And we have a pack voltage of 317.19 volts. One of the reasons I wouldn't turn on the contactor if I was a BMS is cell number 23. It's always a 0.14 volts.

And so that seems to be a problem. We've actually got a bad cell in module four, which reads 19.23 when all the others are 22.92. And, and so we know we've got a problem with the pack. So that's the results of our reading the battery.

And as we know, we have a module four cell 23 is, oh, it varies between 0.11 and 0.17 volts. Not quite what we're after. So we peeled off the plastic and took out a one Brazilian screws.

And then the, you can see on here, this is a metal cover, but it's very thin metal, thicker than aluminum foil. But we actually rolled it up like a sardine can, because it's attached with a very strong industrial adhesive to the frame of the battery. And so we've exposed our entire thing.

We're lacking the two modules in the hump up front. But other than that, it has all the usual things. Our BMS board is right here, our two contactors, and of course our output here.

In all of that, we have failed to close the contactor. And on this battery pack, for obvious reasons, we have a bad cell. So we hooked out the same thing up to the 85 kilowatt pack, and it has all good cells.

They're not as highly charged. They're not at 3.8, about 3.2 volts, each of them. Remarkably consistent.

And they're all on board. Still couldn't close the contactor. Played back some can traffic that we captured from my car when it first starts up, and you can actually hear the contactors close when we do that on the car.

Played that back into the can, and we're still a little bit stymied. So we're working on getting the entire Tesla battery pack to work as a pack. I like it as a pack, but it's enormous.

It fits in the bed of our Doka, but it overhangs the end about three feet. And so I'm not too sure that this is something you can use in a car, but we could certainly use it here for solar storage and to power our DC-powered CHAdeMO charge station. So we've got an awful lot of batteries in a big brick contractor's box and everything to do 33 kilowatt hours.

Here we have the Tesla 85 kilowatt pack, and I actually got some 980 pound casters and bolted them into the holes on the side. So we can just roll this pack around on the floor and then hook it up. But I need to be able to close the contactors.

Of course, we can open this up and put, there's a little four pin connector under the BMS board, which drives the contactors. So I can close the contactors. I want to do it without opening the battery.

I want to keep the pack intact and be able to open it and charge it or discharge it. And so that's what we're working on and we're struggling. Colin will be here March 13th to the 17th, and he might be able to help a little bit.

He's a pretty smart fella. And speaking of which, after last episode, I put the episode up on a blog and kind of expanded on it a little bit, put a better diagram of what we have going with the batteries, but also examined one of the Tesla modules to some degree. And in that, I traced out and identified a chip, a BLQ76 chip is designed to monitor six battery cells by Texas Instruments.

And it works over the SPI bus, the Serial Peripheral Interface bus. And they actually have the holes here for connector J1, and they have the SPI consists of a clock signal, a MOSI and MISO. MOSI is Master Out, Slave In, and MISO is Master In, Slave Out.

So it is kind of directional and it's synchronous, it uses a clock. And this SPI bus is available on your Arduino board. And it can do serial data up to about two megabits per second.

And in fact, we use the SPI bus for Bluetooth low energy modules. So that bus is exposed here on this thing, but there was this connector that connects to something different. We found a Hackaday project, which was done by a guy named Jared Tumulth.

Jared was with the University of the Royal Melbourne Institute of Technology in Melbourne, Victoria. He's a graduate from there. He actually works for a company called Indy Semiconductors.

They do LED drivers, but they also do some interesting efficiency add-ons for power electronics like three-phase inverters. And he had started a Hackaday project to decode this module. And he had a buddy who was working with him, who I believe is named Tom Debris, and I don't know a lot about him.

In any event, they had a Hackaday project to reverse engineer this BMS module board. Unfortunately, it had gone dormant last April 2016. And so I posted our video, the spec sheet for the BLQ76.

There's also an 8051 multi-controller on here, and I put the spec sheet for that up. And a little RF isolator chip, which is kind of fascinating, and I put that up. And the title of the blog is Help Wanted, Battery Meister.

And asking if anyone knew anything about this and cared to fool with it. Well, Jared discovered our posts and was reenthused on the project. And he and Tom got on, and Colin and I were on there, and we had a big flame war over global warming.

But along the way, we talked about what this might be. And what we decided, this TI chip normally has three buses. The host bus, and then the north bus and the south bus.

And you stack them using the north and south buses. The problem with these little BMS boards is they're powered by the battery itself. And in a stack of 16 of them, you're at a very different potential at the high side than you are at the low side.

And you wind up destroying all the electronics with common mode differentials. For some reason, Tesla didn't use the pins at all for the north side and south side buses. But on the host bus, they put their own host, an 8051 multi-controller on there.

And they interface that to this connector on top using an RF isolator, which provides about five kilovolts of isolation to the thing. But it looks like a serial data stream. So kind of a UART, universal asynchronous receiver transmitter type daisy chain from one module to the next, hooked up to the battery management system.

So Jared had actually downloaded the firmware from the 8051 and one of these boards to a file. And Tom started decoding it, disassembling it, which is kind of hard work to read somebody else's disassembly language with no variable names. I've done it.

It's non-trivial. Colin jumped in and started doing the same thing. I said at the time, wait a minute, guys.

Why are they reinventing the wheel here? So here on the spec sheet is a protocol to talk to the BLQ. And you address it by addresses. It can be from 0 to 0x3e, 3e hex.

And there's a set of registers in there. And there's a command to read or write to each of these registers. So it has its own language.

Why are they reinventing the wheel? If it was me, if they're doing this for isolation with the RF isolator and the MC8051, why would they change anything? They're looking for the magic words to be able to read the voltages and temperatures. That's already done. And the BLQ, well, they said, well, there's 4 or 5k of code in there.

That would be too simple. They've got some sort of trick, double throw now, quick disconnect. Tesla super sauce in here.

And so they're busy disassembling it. But Jared got to thinking about what I was saying and went back and actually did some testing and built a little serial port adapter and a Python script to send the commands right off the data sheet. And lo and behold, you can daisy chain up to 63 of these.

And you can read each of them by address and every voltage, all six voltages and two temperatures right over that bus. And so he has basically cracked the communication protocol to talk to the Tesla battery module board. I think this is huge.

He's saying, well, you've said the third message. And let's go back to the climate change thing. Here's a theory.

Here's me typing myself smart. That doesn't count. The only thing that counts is when you test and demonstrate it.

And that's what you did. That's the breakthrough. And so congratulations to Jared Tuma and to Tom DeBrie for some excellent work.

I think this is extremely important in that it's very difficult to put the Tesla battery pack in a car. But we can take these modules and re-stack them. And if we can find the connector part number on this thing, make our own daisy chain.

And we'll take a little EVTV CANDue and code that up to talk to that. And then we'll be able to transmit that via CAN to any system. And perhaps even BLE to any pocket phone to monitor your pack of Tesla battery modules.

And so I said, I understood you wanted this for cheap. And they were available for cheap. And they're very capable batteries.

But they're not safe without you having knowledge in your system of the individual voltages and temperatures. And the ability to shut off your pack if something goes awry. And we'll be able to code all that into the EVTV CANDue system and do that.

And this is not going to be a big proprietary deal. We're going to publish the software open source. Anybody can build something to do this.

My concern has neither been selling the batteries nor anything else other than I just don't want you burning your cars to the ground and causing a lot of regulatory examination of the concept of building your own electric car. They were dangerous. And thanks to Jared Tuma and Colin and to Tom Debris, I believe we're going to be able to make that safe and make it relatively inexpensive to make it safe and fairly flexible where you'll have CAN messages that are published containing your battery voltages and temperatures.

And perhaps wireless BLE transmission of that. And you won't even have to change the board on the module. We're going to try to find these connectors and you'll have to wire them up in series and to our device or you can make your own device.

Whatever you want. But and we're going to open source the software as to how that's done and make it available. So I think this is a huge move and a big hurdle.

And kind of per John's wisdom of crowds theory, we simply posted this online and within two weeks, it's been worked out as kind of a group EV-TV hack team. So we haven't got the battery pack done yet, but we're getting close. And we don't have this done yet, but we know how it works.

And it's a matter of doing some work to complete that. I don't see any game stoppers there. It is kind of an odd 612.5 kbps bus, which is probably sort of halfway between your normal UART speeds and the SPI bus speeds.

So you don't run out of buffer space in going back and forth between the two buses. Basically the 8051 and the isolator chip act as a translator buffer between those two types of buses. And so I'm just extraordinarily pleased to report how this came out and the breakthrough that Mr. Tuma and Mr. Debris and Colin have worked out.

Colin is busy working on how to get a Due to do 612 kbps. We may have to go to a teenier of the new ESP32 or something to do that. But hopefully we can do it with a Dewey because it's already got CAN on it.

We already know how to use that. I've already got a shield that does BLE for it. And we've already got the code that reads, you know, like the 6F2s and so forth.

CAN message so we can turn that around and make 6F2 CAN messages pretty easily. So I think that'd be the easiest way to do it. We'll have a product, a box, and some connectors and some pins.

And you'll be able to wire that up to a series of modules and have a BMS that you can use with your system there as kind of a building block and safely use this thing. But we're going to open source the software if you want to make your own device. That's, you know, certainly yours to do.

The important part is that we get this out there. So you don't have people building cars with these batteries and blowing themselves up. And I'm not being, you know, superior in this in any way.

We're actually building a Doka right now with a LEAF battery in it. It doesn't have a BMS. It's going to kill me.

So there's some urgency here. The batteries are at an attractive price. They are.

People are going to want to use them, but we have to tame them. And the best way is not to reinvent the wheel with our own BMS, but to enable you to use the BMS that's already in them. And that's where we're headed with this.

And that's what we're going to do. Stay with us. you