Hacking the Tesla Drive Train

Notice everything, and I'm sure you picked up on my Donald J. Trump official coffee cup. I saw this on Facebook and had to have one. It has the official seal of the President of the United States on it.

Actually, it's a official presidential seal with Donald Trump in the middle of it. But that's okay, it's all about Donald. I was fascinated with this cup and fascinated with Donald Trump.

I don't agree with most of his policies, but I love the way he has manipulated the media and about 50% of the country, particularly the alt-left libtards. And it's given me great joy to poke at and get reactions from our many ecologically-minded friends in the space of alternate energy, electric vehicles, and solar power. They jump pretty good.

In any event, this is the official Donald Trump coffee cup. Actually, not from Donald Trump, so I guess it's not very official. And I was amazed to find that it's made in China.

So, I'm very excited this week on a number of fronts, but I'm a pretty excitable old guy. But this one's got me. John B. Goodenough is actually the chair of the engineering department at the Cockrell School of Engineering at the University of Texas at Austin.

And one of my old buddies, Bob Metcalfe from Internet Days, has been down there for some time doing the entrepreneur class school. I'm not sure he's still down there, but he did it for quite a while. But I do digress.

John Goodenough actually is a key man in the development of the lithium battery. A little background. The lithium battery was very exciting, but it didn't work very well.

And that is because they used lithium as the anode, carbon as the cathode. And lithium is an ideal material for that, except for a couple problems. Lithium is our third most common element in the battery. Everybody thinks it's kind of rare. It's not rare, but it's rare to find by itself because it reacts with everything.

It has three electrons. The electron in the outermost valence band of the atom is separated from the conduction band of metals by a very small electron band gap or electrical band gap. Electrons can go up in valence band when they gain energy.

And they normally, when they go down toward the nucleus, they give off energy. I usually like photons. And so that's sort of the basis of our LED lights and the photovoltaic effect, where we add energy to electrons and promote them to the conduction band to make electrical power.

The band gap, it determines how much energy is needed to do that. And lithium has a very narrow one between the orbit of its outermost electron and the conduction band. As a result, that electron is promoted to conduction very easily.

And this causes a couple of things, but mostly that it's extremely reactive. It exchanges electrons or gives up electrons to anybody who wants one. It's kind of the working girl of the periodic table. She'll dance with anybody that's, you know, got a quarter. And so it combines with almost everything. And so we think of it as rare because it can't go dig up lithium metal.

It's combined with something. It actually exists in seawater at 14 parts per billion and can be extracted commercially. Perhaps not inexpensively, but it could be done.

Lithium is our lightest metal, so light that it would float on water. And so reactive that it would burst into flames. In fact, it bursts into flames on water somewhat explosively. It'll actually boil all the water out of the glass before it's done. So you would think of it as somewhat volatile. And there go part of the problems with our batteries.

It was certainly a problem with the original batteries. In about 1991, I'm guessing, Panasonic came up with a solution. They would invert it and use the graphite as the anode material and lock the lithium in a crystal structure of lithium cobalt oxide.

And then take electrons away from the lithium on the cathode side, resulting in a lithium ion, a cation to be precise, a positive ion, which would then migrate to the graphite carbon anode and intercalate into the crystalline structure of the carbon. Held there, not by any covalent bonds, but in suspension due to the electrical charge of the electron, which we took from the cathode and placed on the anode. And so in both cases, the lithium was in a crystalline structure and not really barren.

In addition to that, a fortunate accident with some of the organic solvents in the electrolyte would cause the deposition of a, variously called a barrier layer, but a SEI layer, covering the anode and to some degree the cathode with kind of a rubber-like or vinyl-like covering, which contributes to the aging of the batteries, but also keeps the volumetric change in the carbon crystal from breaking up so easily. So it promotes the life of the battery. And that was the Panasonic battery.

Still had a number of fires. John Goodenough sort of retired from Oxford and took a position at the University of Texas at Austin. And there, with a group of researchers, and as the head of that group, developed the concept of a more intricate crystalline structure on the cathode side, which as many of you know, we are big fans of, called lithium ion phosphate, which is an oxide, but it's sort of a tetrahedral crystalline structure that is somewhat complex.

And it made the batteries much, much safer. And those have been the cells we've used since 2008, the earliest days of EVTV. John Goodenough is a brilliant man and really the leading top researcher in mind in lithium batteries, and has been from the beginning.

For the last couple of years, I wasn't sure he didn't suffer from a bit of dementia because he kept talking about returning to a lithium metallic anode. This week, it was announced that he has a paper published in actually Energy and Environmental Sciences where he announces what I consider an enormous breakthrough in the concept of lithium batteries. He was approached with the work of Mana Helena Braga of the University of Porto in Portugal, who had been playing with, I suppose researching would be a better term, I play, they research, they play too, with a glass electrolyte.

And she came up with a very innovative idea of producing glass using water and managed to produce a form of glass that would readily transmit lithium ions. Now, the problem with the lithium anode is that, again, it reacts with everything except for a few inert materials, like glass. As you charge it and discharge it and so forth, it tends to form dendrites, projections, sharp projections, we may call stalactites and stalagmites, forming in caves from the dripping of water.

This is really quite similar, and they call it dendrites, but it's forming from the dripping of lithium ions. And those would occasionally, or actually rather quickly, penetrate the glass trash bag and use it as a separator in the cell to prevent electron flow. When it does puncture that and bridge to the cathode side, it causes immediate discharge, a huge thermal event, and given the low temperature at which lithium cobalt dioxide released free oxygen, it's essentially an explosive thermal runaway.

So there were a lot of problems with the lithium anode, but you just can't beat it for energy density. As I said, lithium is our lightest metal, and nothing will intercalate the lithium ions, particularly with electrons, unlike lithium metal. And so the ability to use a lithium anode promises at least a three-time, and probably greater than that, increase in energy density of a battery cell if you could somehow hermetically seal it from the world.

Mr. Goodenough, Dr. Goodenough, recognized the work of this woman from the University of Porto and had her added as a fellow at the University of Texas at Austin. And she is part of a team, but he really provided the expertise to get this to a working cell and what they developed was a means of electroplating, really, much as you electroplate a gold onto copper or silver onto nickel, but electroplating or plating lithium metal on the surface of this glass, which acts both as a electrolyte and a dielectric. And so we have no need of a plastic bag separator, liquid solvents, lithium salts, and so forth.

The glass itself, having been chemically altered by the addition of minute quantities of water, was able to carry lithium cations while totally blocking electron flow, but also hermetically sealing the anode from the outside world and mechanically, basically, preventing the growth of dendrites. They may form at the surface, but they immediately take the shape of the glass. And so they're inherently flattened right back to a flat plating surface.

And so they can, again, reverse the whole thing, have lithium metal as the anode and carbon as the cathode material again. And this could be a profound leap in battery technology. They would be a three times, minimum three times greater energy density, meaning that your 1,330 pound, 85 kilowatt Tesla pack would be about 450 pounds if they could reduce the weight of the can, similarly to the batteries.

They're seeing 1,200 cycles in initial experimental cells. And so very long life, virtually completely safe. There's no solvents.

There's nothing to initiate the thermal runaway. They're almost a totally safe battery. Long lived, good, great energy density and relatively inexpensive to manufacture.

Indeed, implying that they could even make batteries from other alkaline metals, such as potassium or sodium. Imagine a cell made of salt, glass, and charcoal. Pretty cheap materials, but that would be a sodium carbon battery with a glass.

Dielectric and electrolyte, that is basically cheap as dirt. And so maybe not quite the energy density of lithium, but certainly at much lower cost, easier to find. If you think you can get lithium out of seawater, you can get a lot of salt.

And so I think this is one of the most profound breakthroughs in battery technology that I'll probably see in one lifetime. And it's amazing that this was brought to us by a 94-year-old guy, but he's a bright guy and a neat guy. Quite religious, and he's doing some great stuff.

Time from laboratory to commercialization. You know, these things go on, but this one probably does not face a lot of hurdles. I think they're into some issues about power density, the ability to deliver large amounts of current with all solid-state batteries, and ultimately understand that this lends itself to the type of manufacture of computer chips where you have the electrodeposition of lithium or sodium on one side of a glass plate and carbon on the other.

And they're stacked, and this is all done photolithographically. And I mean, that would be a fairly expensive process unless you scaled it up, at which point it becomes dirt cheap too. And become subject to Moore's Law like the computer chips.

And so, you know, all around it's a great thing. Solid-state batteries tend to have to operate at high temperatures, 60 or 70 degrees centigrade. This operates down to minus 20 degrees centigrade.

And does not appear initially, the initial take is that it does not appear to degrade in capacity or current capacity. With temperature to nearly the degree of even the existing lithium batteries. So this is all around just fascinating good news.

And it's a function of time and money to get it there. But I think that it's going to overnight cause a huge amount of R&D in a lot of places and a lot of commercial interests through the Patent and Licensing Office of Austin University. And so I think that's going to be quite a thing.

In other news, China has their, you know, we've seen our documentary by Chai Jing on under the dome about the pollution in Beijing. Well, it's just become insufferable. And the Chinese government is taking somewhat dramatic action.

There's probably a buck in there somewhere, has something to do with BYD and their electric taxis. But commencing immediately and henceforth thereafter, it's illegal to introduce a new taxi in Beijing that runs on gasoline. You can run the ones you've got. But when their life is over, they can only be replaced with fully electric taxis. Now that may sound like a limited area, just taxis. There are 71,000 taxis in Beijing.

And so to support them, of course, they're going to have to develop quite a bit of charging infrastructure and so forth. It really jumpstarts their whole electric vehicle thing. A few years ago, China talked a lot about electric cars, but they'd sell like 35 cars a year in all of China. But they are getting on board with the EV thing very quickly at this point. And it's a matter of survival. Their air is simply unfit to breathe.

Let's see, Elon, a couple of things. He got a letter from a little girl in the fifth grade who proposed that he have a video commercial contest..Apparently, a lot of Tesla fanboys have made commercials for the car, unfunded and unaired, but just to be doing it. And some of them are pretty good. And her father is a writer for U.S. News & World Report and Inside EVs, which was recently acquired by a larger automotive publishing group.

In any event, they sent the letter to Elon Musk by Twitter. And within minutes, he replied he loves the idea and they're going to make it so. It is not a dream.

It is a simple feat of scientific electrical engineering. Electric power can drive the world's machinery without the need of coal, oil or gas. Although perhaps humanity is not yet sufficiently advanced to be willingly led by the inventor's keen searching sense.

Perhaps it is better in this present world of ours than a revolutionary idea be hampered in its adolescence. All that was great in the past was ridiculed, condemned, combatted, suppressed, only to emerge all the more triumphantly from the struggle. Our duty is to lay the foundation for those who are to come and to point the way.

Yes, humanity will advance with giant strides. We are whirling through endless space with an inconceivable speed all around. Everything is spinning. Everything is moving everywhere. There is energy. SpaceX had a fascinating launch this past week, CRS-10 Dragon launch to replenish the International Space Station.

And they had a very, very successful first stage landing, not on one of their floating docks, but at a launch pad ashore. The capsule itself had a relative GPS circuit malfunction and they aborted the docking with the International Space Station for a day, got that worked out and successfully docked it with a bunch of experiments and supplies. It's a pretty large capsule and so quite a bit of stuff.

Let's take a look at the first stage. Good day from the International Space Station Flight Control Room at the Johnson Space Center in Houston, where another attempt will be made today to reel in the SpaceX Dragon cargo craft and its priceless cache of scientific experiments and other supplies for the Expedition 50 crew. There is a great view of the Dragon spacecraft, the SpaceX cargo craft that was launched back on Sunday from Launch Pad 39A at the Kennedy Space Center.

So the Dragon craft is now approaching the 350-meter mark, about 1,148 feet below the International Space Station on what is called the R-bar, the radial vector, which is the imaginary line drawn between the station and the Earth. And from one of the truss cameras on the International Space Station, a view of Dragon as it approaches an orbital sunrise about to cross the west coast of South America over northern Chile, less than a minute away from arriving at the 30-meter waypoint. Dragon now holding, putting on the brakes at the 30-meter mark.

Everything in great shape. 23 meters now separating Dragon and the International Space Station. Houston, Station on cue for Dragon beginning approach.

Houston, copy. The robotics officer here in mission control reports that the Canadarm2 is now in motion. You can see it very clearly in the monitor that Thomas Pesquet and Shane Kimbrough are using in the cupola at the robotic workstation.

This view now from the end effector on the Canadarm2. Two and a half meters to go. The station now in free drift. One meter to go. Capture confirmed. Dragon in the grasp of the Canadarm2 robotic arm.

Capture confirmed at 4.44 a.m. Central time, 5.44 a.m. Eastern time, as Dragon and the International Space Station flew 250 miles off the northwest coast of Australia. As you know, a big part of our life today has to do with reverse engineering Tesla CAN protocol for the operation and communication of the various devices in the car, the battery, the drive inverter, the charger, the DC to DC converter. We're having a lot of fun doing it.

It can be frustrating. It's better to be lucky than good. But we stumble across things occasionally and piece it together patiently. And we're getting a larger and larger body of tribal knowledge on the topic. But I came across an interesting video this week by a fascinating guy named Ben Krasnow. And he has a great YouTube channel called Applied Science.

Not Tesla specific. He just does all kinds of little hacks and design things. And he's a professional engineer and really quite talented. But he did come up with an unusual approach to displaying Tesla CAN data. Probably not something we're gonna do here at EVTV, but I thought it was very entertaining and truly creative. Let's take a look at Ben Krasnow's means of hacking.

Today on Applied Science, I thought we'd talk about extracting information from a modern car's data bus. So we are sitting in a new Tesla Model S on this rainy day. And I thought I'd talk about what I've learned and how I plan to build a system that will show the data on the car's existing console.

It's pretty normal, of course, for a car to show you the speed it's going on the dashboard speedometer. But there's actually hundreds of other variables that are commonly sent through a car's data bus. And some of them can be pretty interesting.

So for example, motor torque in this car, since it's electric, you can see the number of amps being pulled out of the battery pack and literally on and on. So if you know something about car buses, you may have heard this term ODB2. And this is a somewhat standardized port that was created mostly for emissions control hardware.

And the Tesla has an ODB2 port in the driver's footwell. But since this car doesn't have an internal combustion engine, there's no emissions equipment. And so there's really not much on that bus. So avoid that. What you really want to do is check out these two connectors that are beneath the center console here. There's a little cubby hole that you can just pull off. There's two clips that just snap down and it pivots in the back. So just pull that off. And then the smaller connector is a four pin connector.

And this is actually ethernet. So a couple of years ago when the Model S first came out, someone connected to this and realized it was ethernet and started poking around in the car's private network. And then a couple of months later, Tesla issued a firmware update and shut this port off.

So what they did was they disabled this at the car's ethernet switch. So if you still want to get on the car's ethernet, what you can do is find the wires, the ethernet wires going to this center console computer and then put your own switch in there, basically tap into it. And you can still snoop the ethernet if you want.

But we're not going to talk about that in today's video. What I'm most interested in is this other connector, which is a CAN bus connector. Tesla calls this a diagnostic connector.

And there's actually several CAN bus buses on here. If you haven't heard the term before, CAN bus is used in the automotive industry and it's more real time. So ethernet might be good at sending something like an audio stream that you're going to play on the car's sound system, whereas the CAN bus is actually carrying the low level signals that are connected to things like the accelerator pedal, the motor controller indicating how much current it's using and that sort of thing.

I spent quite a bit of time piecing together all the information needed to get this going and I'll put all of it in the description of this video. Feel free to ask me any questions. There's actually not that much information out there. Like, for example, this connector is not a very standard connector. In fact, you can't get this at DigiKey or Mouser or anything like that. But after hunting around on the Tesla Motor Club forum for many, many hours, I found a seller who actually has these connectors.

So thank you very much for finding these and making them available to hobbyists. As I mentioned, this car has several different CAN buses and currently we're going to talk about CAN bus 3, which is the one that carries drivetrain information. So motor power, battery power, accelerator pedal position, all that sort of stuff. And as you can see, currently the scope is looking at what's going on on this CAN bus, CAN bus 3. And you can see a very high rate of data going, even though we're just sitting here. The car is basically idling. OK, here's a closer look at the scope. And as you can see, there's a lot of data going by on the bus here. And this scope can decode all the CAN bus information. So if I turn bus 1 on, we've set that up as a CAN bus and the bit rate is 500k per second.

And CAN bus is, in fact, a differential protocol. It's two wires, plus and minus. And normally the wires are always driven differentially. But interestingly, on this scope, you only need to sample one half of the line. I guess it's good enough. And so we can do a single acquisition and you can see there's data going by.

And so to make this a little more interesting, what we can also do is set up the scope so that it triggers only on a specific packet type. So we'll go to the trigger menu. And instead of just triggering on the edge of the waveform, we'll trigger on the bus.

And this is B1, the CAN bus. And we're going to trigger on the identifier. So the identifier in a CAN bus packet describes sort of which part of the car it pertains to. So I've already set it up so that it's going to trigger on identifier hex 106. And this is one of the drivetrain components. I think it might actually be the rear motor controller.

What I'm going to do is step on the gas pedal or the accelerator pedal of the car. And what you can see is that this value changes as I step on the pedal. The car is beeping because it tells me I have to put it in drive if I want to go somewhere.

These other ones are zero because we're not moving. This would be describing motor torque, I think. But as you can see, if I floor it, this goes to FA. So that's a maximum pedal position. And it's zero, of course, when my foot is not on it. Another really useful thing we can do with this scope is capture a large quantity of data.

So if we put the trigger type back to just edge and I'm going to stop the acquisition here and I'm going to change the record length to a really high value. So when the scope was running really quickly like that, we're storing 100,000 data points. Having it just show this little section.

But let's say we wanted to capture like 10 seconds worth of CAN data to do further analysis. We can do that too. So I'll set the scope's memory to be 20 million points. And then I'm going to change the time per division to be a much higher value, like one second. So this should be 10 seconds worth of data. So I'm going to press the single button. This scope doesn't show you in real time, but actually it is computing or it is recording right now. And so it's probably, it's been about five seconds since I pressed it. There's all the data.

And what's happening is it says B1 not yet computed. So in another few seconds, it's going to go through and figure out all of the CAN bus messages for this stream. And there they are. So if we wanted to see them, what we can do is go to B1, this bus that we're decoding, and then go to the event table. And this is a list of all the messages that came in during that 10 second acquisition. And hopefully it even shows what time the message came in. So minus five seconds. I have the trigger point set here and the whole acquisition was 10 seconds. So minus five would be, you know, the first thing that it saw.

And here's that identifier column that I was mentioning. So if we scroll down through this list of data, there's a 106. And you can see it was 00.

If I had my foot on the pedal, this would be a different value here. So what's neat is that I can say, I can save this event table and it's not highlighted right now because I don't have a USB key plugged in. But what I was doing was collecting lots of data while driving with a USB key in here. And I could hit single really quickly and then go collect some data. And then after it had decoded all of the CAN bus stuff, I could say save save table and it would save a CSV file to the USB key. And then I could take the USB key back to my computer and plot it in Excel. And I ended up with something like this. OK, I know it's dark, but that took quite a bit longer than I expected. I'll do another video in the future with better lighting.

But let me tell you what's going on here. I've got a Raspberry Pi 3 and I bought this from Copper Hill Technologies and it comes with a CAN bus shield. So this is connected directly to the car and it even has a 12 volt to 5 volt buck converter.

So it's literally just four wires connected to the car. And originally I was thinking, well, you know, I'll make another screen or something in here kind of like this, an extra LCD to show the data that I've been logging. But really, it's kind of annoying to mount another screen and I've got this huge 17 inch touchscreen built into the car's dash already.

So I started thinking about ways that I could get the data from the CAN bus onto the screen here. Now, Tesla has this pretty well locked down. However, there is a web browser built into it and I started looking at ways that I could get the CAN bus data from the Raspberry Pi into the web browser.

This browser is WebSockets compatible, which means that I have a pretty low latency pipeline between the browser and the server. So what I'm doing is I'm running a Node.js server on the Raspberry Pi. It's pulling data in through the CAN bus shield in real time.

And then this Node.js script is running a server and the browser here is sucking in the data. So you can see it's actually really responsive. It's updating 10 times per second and I'm pretty sure it can go much faster than that.

So what I'm doing is just stepping on the accelerator pedal and you can see it's updating nice and quickly. There was a couple of interesting gotchas along the way that you might find interesting. So the way that I'm connecting the car to the Raspberry Pi is through the wireless network. Like I say, I didn't want to hack into the car. I haven't touched that four pin ethernet connector. And so the first trick is that if the car will not connect to a wireless network that doesn't have internet available, or at least it seems that way. So to fix that problem, I got my Android phone and connected it via USB to the Raspberry Pi and used USB tethering from the phone. So the Raspberry Pi actually has internet now through the wireless network on the phone. Okay, so now with routing and everything the car will eventually get to the Raspberry Pi through wireless and then from here to the phone and then out to the internet.

So that's all fine. The next interesting problem is that the Tesla browser doesn't allow you to connect to local IP addresses. So 192.168, whatever. And they did that purposefully, I guess, to keep you from doing exactly what we're doing now, I guess. But anyway, since we have full control over the router, obviously this Raspberry Pi is routing all of the packets. We can just set up a static route. So I've gone to the address 42.42.42.42 and that just reroutes all the packets to the local host here and we can serve those packets. And the browser in the car has no idea that it's talking to the Raspberry Pi. It thinks it's talking to a public web server.

But this seems to be a pretty good solution. It's very responsive and it just keeps running. It's not like I have to keep reloading the page or anything like that. And so with a little magic, you can imagine what this is going to look like. You know, I'll have a graphing in real time in the client side and I'll be able to do things like select different things. Like, you know, if I wanted to see acceleration over time with motor torque and all that stuff, I could construct a graph in the browser.

And that's what we'll be hopefully looking at next time. So in the next video, we'll be talking about the physics of the car a little bit more. The whole purpose of making this data collection system is so that I can record all the data from doing, you know, kind of extreme things with the car.

And then we can, you know, see what's interesting with that. Okay. I hope you enjoyed that. See you next time. Bye. Okay.

Ben Krasnow, Applied Science. That was a very interesting video and he makes good videos too. The Stephen Otto and Nicholas Kurali have been patiently waiting for weeks and months and in the case of Nicholas for a Tesla drive unit.

I've apologized often enough. I usually fight them off by sending them an updated manual showing me the stuff I added this week. And it's pretty good stuff.

So they kind of like it. But it's gone on long enough. We are going to ship both next week. I do intend, but I thought it worth a little walkthrough of what the hell we've been doing. And let's take a look. We are back with the Tesla drive unit and I've got two of them that have been on order for several months while we've been playing around with it.

I thought I'd explain a little bit of what's been going on. One of the issues is we have upgraded the controller to our new JibQ 6.2 which gives us a nice mod ice cinch enclosure and series of connectors. I've never met a connector I like.

I loathe these two somewhat less than most and they screw in very securely and they have a pretty good numbering system that I can almost read. We have had some software issues and none of them really with the drive unit. We haven't changed the code that runs the drive unit in over a year.

And there's a lesson in there for a bunch of us. There's a bit more to producing a product that other people can use other than there is to hacking a drive train and getting the wheels to turn. In doing that, it's better to be lucky than good but Mark Weissman and Colin Kidder and several people came down and we had it turning in the first morning that we convened to attack this.

Since then, I've been trying to get a system that you could put in a car and it makes some sense. I got a little carried away. There's a lot of CAN going on here. We got CAN to the drive inverter, CAN back from the drive inverter, CAN to our EVIC display. It's a seven inch display. It looks a little bit like a Tesla but a lot smaller.

And even on the Tesla Model S, that big screen, if you're moving down the road, I can't hit anything on there and get it right for the third try. So we got this little KP2400 CAN switch. We have eight switches for park, reverse, neutral drive, creep on and off, zero trip and increase and decrease our regenerative braking.

And they're very positive IP67 switches. They're kind of a normal where even my fat fingers can find them. Have a very positive press. You can tell when you pressed it and anything you do on here shows up on the switches and anything you do on the switches shows up on here. Of course, we have our forward throttle and that is wired directly to the drive unit and that's how you accelerate. We've been adding some things.

First, the software problems are, you know, this is being your own worst enemy. I come from kind of a background where real-time software was sort of a deal. Today, I think the Apple guys call them delegates and callbacks or something.

We call them interrupts and interrupt service routines. And it's where things could happen in the real world, interrupt your processor, go perform a function based on that interrupt and return. And what I found, I talked Colin into putting this in Dewey CAN and he did a pretty good job of it, but the little SAM3X just can't deal with all that.

I've got interrupts coming from our IVT current sensor module, the EVIC, the switches, and then on the other side, the other bus from the inverter itself. And it was interrupting itself to death. The bottom line was it would work, but we found oddities like 27 minutes into it almost every time, the whole thing would just quit.

We could reboot it and work just as well as it did before. Those kind of intermittent, especially long cycle problems are hard to work out. So I bashed my head against the wall for a while. Finally went back and gave up on the whole interrupt scheme and just pulled everything, very old school. And all the problems went away. We've been working a lot on the wiring harness.

We've changed, we've wired it, simplified it quite a bit. We have a much slimmer wiring harness now, a lot easier to deal with, a little bit longer because of vehicles like the DOCA. And for Nicholas Corrali, whose shiny new Tesla drive unit is a full sub clip, note that the sub clip has these electronic parking brakes.

Seemed a shame not to let them work. And so we devised, and I think I mentioned this in a show, a little controller box using a Pualu motor controller that will drive those electronic parking brakes. And it's right here.

The conceptually, it's no big deal to get it to talk to our Tesla controller, but we kind of had revisited the issue and decided that, why we sell them as an individual drive units and as a sub clip. So we wanted to make the controller to be completely independent. It needs about 30 amps at 12 volt DC.

So we put separate power wires to it, but then it basically observes the can between our controller and the motor. And we wrote software where it knows to apply the brakes or release the brakes based on what year we're in, much like the Tesla Model S does. And so it took a bit to get that working correctly. And so we've been delayed. It's all coming together pretty well now. We did add a Tesla diagnostics connector to the wiring harness.

And I'm including for Nicholas, one of our Tesla can monitor. This is a little box. You can just pull the center console shelf down a little bit and hook this up to the Tesla diagnostics connector and read can using Savvy Can and the Jibret and so forth to read and analyze can messages on a Tesla Model S. So we use the same connector that we use on the Model S and that lets us plug in the same box.

And we added that to the wiring harness. And it reads all the can messages that go to the display or come from the switches. As it so happens, almost everything happens with the drive inverter that you would be interested in.

Obviously, we have to send to the display so you can see it. The voltage of the pack, the voltage of the 12 volt system, the current usage, the torque usage, speed of the car, the RPM of the motor, the temperature of the motor, the temperature of the inverter. All those things, including some detailed battery voltages and currents and so forth are sent to the EVIC display for, or we call it the all set display, but it's Andromeda interfaces.

EVIC is what they call electric vehicle interface device. And since all that's there, we brought it out on a Tesla diagnostics connector and defined it in the manual. So it's easy for you to monitor those messages and to decode them and to get any information you want to about the drive unit operation.

In the same vein, Bill found that Speedhut has just announced and are shipping a whole series of customized dash instruments, gauges. And we use Speedhut gauges before we put a little EVTV on it or something. And then that's in our dashboard for GPS speedometer or something.

In this case, they've introduced a line that are J1979 compliant CAN devices. J1979 defines the protocol used to report a basic engine operational data on the OBD2 connector of your Chevy. And so they've made speedometers, tachometers, water gauges, a fuel gauge, the, what else did we use? Oh, just the 12 volt gauge to see what your 12 volt says.

And we're going to put all that in the DOCA. And I have added code to the controller to respond to a very basic set of the J1979 PID or program identifiers is what PIDs are. And it will automatically respond to PID requests from those gauges to provide the water temperature, the RPM, miles per hour, kilometers per hour, the voltage of the 12 volt system and so forth.

And that lets you have any design of gauges. They have sizes from like two and three sixteenths up to four and a half inches, all kinds of background colors, fonts, tick designs, bezel colors. You can put your own logos on them.

You can put text on the dial. We just say Tesla drive train on ours, but in any event, you can order those speed hut gauges and your Tesla controller will derive it. This is necessary.

I wanted to have something that looked like the Tesla display to be familiar with the kilowatts and the meters and regen and so forth, much like you have on a Tesla Model S, but it is a seven inch display. For speed and RPM and so forth, we're going to put motor stator temperature on the water temperature gauge, for example. That's just much easier and more of a professional looking car gauge set.

So I'm kind of excited about that. And it's built into the software now that Nicholas Crowley and Steve Otto will be receiving. And so that's kind of a roundabout introduction to what I've been doing on my summer vacation.

It sounds like we've gotten off on batteries or done some other stuff. Actually, I've spent a lot of time on the drive unit, which I thought was pretty much done three months ago and ready to ship. And it's been kind of, oh, I got a great idea.

There's one more thing we can do. And boy, wouldn't that be cool. And among those being getting it to not quit after 30 minutes of operation. So let's fire up our drive unit. I got to press my brake switch here. I'm going to put it in drive.

And you can hear the parking brakes release and the startup of the drive unit. We can turn that up as high as we want. We have regen. Of course, we can change the regen. I can turn creep off and it will slow down. It still turns a little bit, that's because it's not got any car holding it back. In operation in a car, this would not really be enough power to creep you forward or certainly not very much. But if you put creep on, it applies a little more power. And it would be like an automatic transmission car.

You take your foot off the accelerator and it'll creep forward a little bit. And so that's that. I can go back to park. And it applies the electronic parking brakes. It takes it out of drive, but it's still sort of coasting. And actually that little clunk I feel in my Model S. So we're not doing anything different there.

Let me turn it back on. And there we release the brakes. And if it starts turning, we're creeping. We press the accelerator. And we go. And you can hear it spinning down. And the motor is very quiet. And so this, I think this is going to be a fabulous drive unit. But hopefully a good system that he can then build and expand on to get the car he wants, which as I recall was a 63 Chevy Impala. And I can't wait to see that on the road and working. We're going to look at a couple of other drive units today because we have kind of a problem. And it isn't anything that Tesla has done nefariously.

But it is something you should be aware of in procuring drive units. We're learning some kind of better lessons there. We have a kind of an interesting one here, don't we Bill? We do. We do, Jack. It's got a story, hasn't it? The... It turns. It's a good day.

It regens. It works pretty good. We have acquired a number of drive units.

And this is not going to go very well for me. I know you all think they're pretty expensive. We're charging pretty much $16,000 for this and $25,000 for a full sub clip. I think I'm financing EVs worldwide. You're a hell of a guy, Jack. Out of 10 drive units, we've had like four bad ones. We're getting about 60% good. Bill, show me your half shell there. I bought several drive units for this guy.

And what do we have here? Well, this whole sub unit you bought as a good unit, didn't you? As a complete... A full sub clip. Ready to... It was broken. Ready to go.

Yeah, it had some busted suspension joints. I sent the guy a thing. He said, oh, this couldn't be.

It was a good one that left there. I said, no, this is not new damage. And here's some photos.

One of the half shafts was gone. One of the half shafts was missing. One of the links was broken.

And it was broken. The part where it attaches to the frame. To the subframe, yeah.

The subframe was broken off. It wasn't the link that was broken. It was the subframe that was broken.

Yeah, it was a bad break. So I looked at it and said, okay, I kind of made a mistake here. I'll send you a new subframe and a caliper.

One of the calipers was missing. Break calipers. Yeah, it was.

It was just a mess. But at least it had a grind unit in it. Yeah, it did.

Yeah, it did. So we had a grind unit for Steve Otto. And it shined up like a new dime.

We put a cleat in it, new seals. Sure did. It was sitting here ready to go, wasn't it? All ready to go.

It even turned. But... A little bit. Yeah, a little bit.

It was throwing faults left and right. And it was very weak. It just barely... It was like being in reverse and forward.

So I'm not too sure what's wrong with that one. So we pulled this one. And the whole bottom of the end plate on the inverter was broken off.

Yeah, it's got a big chunk missing out of it. The connector was busted off. The little orange dealie is gone.

And the cast here is broken quite badly. Show me your half case. I don't know if you can hold it up where they can see anything.

But where the roller bearings are in here that hold one of the intermediate gears, that's cracked on the outside. And this wouldn't hold hydraulic fluid. Fortunately, we had another bad one that I had bought from a guy.

And he sent it to me. And this was a different guy, by the way. He had to know this.

This was a flooded one with the inverter completely burnt out of it. So we took half of that case. And the can and the end piece for the inverter.

Which were both good. Put it with this inverter and motor, which worked fine. Figured out that the cable between the two halves, we've been very careful not to... It just sent plugs from the circuit board.

Yeah, it's a piece of cake. So it's just a temperature sensor in the motor stator and on the cooling inlet. And so Bill took all that apart and learned a lot about... Inverters.

Inverters and motors. That's pretty cool, yeah. And it made it a lot easier to put the quafe in, right? Didn't it? Well, it made it a lot easier to clean up our surfaces and everything.

Yeah, and the Quaife, we put that one together. It went together no problem at all. Uh-huh.

No problem at all. So if we take two or three of these drive units, it seems like we get one. We can get one.

And then we can put it back together. This thing is just spinning like a top now and making full power. Woo! Just about got me.

We're supposed to be standing here wrapped in tension to keep that from happening. I don't know that we have another one to fix if we break this one, do we? Yeah, we've always got another one. Yeah, we've always got another one.

Remember your intervention where you were trying to get me to stop? I do remember that. Buying these things. Yeah, it was heartfelt.

So anyway, now I've got fluid on Steve's highly polished. We polish these things up. Actually, we spend quite a bit of time cleaning them up.

You get new seals here. This is all new Bondo or whatever. RTV.

RTV, yeah. The new Quaife differential. Yep.

And the new clips around the devices. Yep, we'll have you. And seals there.

And check the cables out and so on and so forth. So we're learning a lot about drive units, but we're getting a lot of bad ones. And so this one will go to Steve Otto.

And I'm pretty confident it's going to be a good one because we've been able to examine it and test it inside. Yep. Do a little more testing.

Mostly make sure the fluid goes through. Yeah, I'm obstructed. So we got a couple of things to check, but I think it's going to be a great drive unit.

But we're down to swapping out piece parts to make one at home. That's not entirely all bad. We get to look at all of it.

It's fun. Except for the part that I got to buy three of them to ship one to you guys. That's not making it very easy.

Again, I'm not generally selling these. We don't have them on the website. But I've got a lot of people who are interested in these.

And if they sound technically competent and willing to maybe update software and stuff like that and make wiring changes and so forth as we go along, I'll sell them one. It's kind of we're in beta test. We're beta test, yeah.

But so far it's going really well. They're the pioneers. Yep.

So let's take a look at another interesting failure that we're going to put in the docker. Okay. It's oriented differently.

Oh, well. Bill, this is a nice shiny one. Yeah, the power of editing.

What is going in? This is the one going in the docker. Excellent. This drive unit works, seems to make full power.

But it was a little bit noisy. And I was hopeful that with the magic power of your mechanical genius in putting in the quiff that that would get all better. So was I. It didn't really get better.

So fire it up and see what we've got going on here. It does turn. Yes, it does.

Would you hold on to it so it doesn't come at me? I'll try. Runs good. Runs like a champ.

What's that noise? A growl. It's a growl. It is a growl.

It's a growl. I can hear you almost. It's a growl.

We have a growl. There has been quite a bit of reports among Tesla Model S owners. One of them stopped in here and said he had had his drivetrain replaced four times because they went bad.

And I'm like, so it didn't drive? No, it drove. But it made this noise, kind of like a coffee grinder or something in the back. And a number of people reported this.

Tesla jumped on it pretty good. If you had it, they would swap it out pretty quick. They just gave you a new one, didn't they? Yeah.

And then they extended the warranty on the drive units to equal the battery. Yeah. And said they had it now fixed where they were good for one million miles.

Hmm. One million miles. I don't think this one made the trip.

This one didn't get swapped. And so very possibly, this is exactly what they were reporting was this noise. We have eliminated the differential.

Yeah, yeah. Because we've replaced it. Yeah, that's not the problem.

And we replaced the differential bearings. Yes. And of course, Bill has some of our new... Stub shafts.

Stub shafts that hook up to the Doka. The Volkswagen axles. Yeah, the Volkswagen drive shafts, yep.

And so it's ready to go in. Fortunately, he has engineered a slightly skewed mounting system where he can put this in and take it out. And how long would it take you to drop a motor? Oh, 10 minutes by myself.

10 minutes. With a lift. With a lift.

We have lifts, yeah. 10 minutes by myself. 15 minutes to put it in.

It's 10 minutes to take it out. Another 15 or 30 to put it on in. Yeah.

So guess what I think we should do? Run it. I want to hear it come apart. So do I. We got all kinds of stuff that could potentially come apart.

Yes. We got axles could fly apart. We got gears that could fly apart.

And it's kind of a neat noise in a way. A lot of people complain about EVs being too quiet, don't they? At least they think they will be. This might... Got it.

You're supposed to be hanging on to that. You're supposed to warn me. Actually, when it spins up, it doesn't sound too bad.

Nuh-uh, no. No, let's try it. I'm ready.

I'm ready for you this time. Yeah, well, I think so. I don't want this jump there.

I don't either. It kind of quiets up once it gets going. I mean, it sounds like a regular one.

But it has this growl. It's at idle. It's a little idle growl.

So that's interesting. Not that they idle. Let's put it in the doca.

Yeah. And we'll drive it. This is kind of the theory that the butcher gets the five-day-old meat.

The vegetable guy gets the rotten tomatoes. Yeah, that's right. Yeah, yeah.

But I'm curious of what the outcome is. Yeah, yeah. Because they were complaining about the noise.

And some said that it broke down on them. I'd like to break it in front of me and then take it apart and see the pieces. Absolutely.

Yeah, yes. Give us... Just to entertain me. Yeah, sure, sure.

And so let's go ahead and put this in the Doka. Will do. And let's see how that plays out.

Hold my beer. Hold my beer. Watch this.

Watch this. Yep. Awesome.

I like it. Yeah, that was the three most dangerous things in the military. It was the lieutenant who said, in my experience... Yeah, right.

The captain who... What did the captain say? And then the chief warrant officer said, watch this shit. Right, right. It's an old joke.

And I remembered part of it. So... 66%. Alzheimer's, yeah.

Nope, 66% there. So this one has a problem too. And so I guess the mission is or the message is we get a lot of people who just are desperate to buy this from a salvage car themselves.

Yeah. And have us sell them something for $79 that'll let it run. That's probably not happening.

The R&D and the device, the hardware, it winds up being fairly substantial to make this package. I've been reluctant to do it because there were reports of there being different versions and that we wouldn't work with all of them. Now, we haven't tested any of the small 170 kilowatt motors at all yet.

But so far, that appears to be all BS. We've got motors out of 2015, 14, 13s. Yeah, yeah.

And... Some are sport. That one doesn't make full power, but I suspect, again, instead of attributing things to malfeasance, it's usually incompetence or breakage. And so the truth is we're buying a lot of drive units and we get taken a lot.

I would say that there's a good possibility that people just don't know and that there's not a great deal of care given to a wrecked car in the six months between the wreck and when it's sold. But unfortunately, I think some of it is knowledgeable. Like flooded units.

Yeah. And so it's kind of a buyer beware market. And I'm about to cave on the whole thing and let you buy your own.

And then we'll outfit it. You'll ship it here. We'll outfit it and get it and test it.

And put the wafe in it and send it back to you because I've got $100,000 tied up in drive units and about four out of 10 aren't any good. And so that's an issue that we can't make up on volume. I have to buy two to make one and then put all this hardware with it.

It's a little bit of an issue. So we haven't really decided how to market and present this and do it where it works for our viewers and for us. But right now we're beta testing them.

We're supplying them. I want to see them turn. I want to hear them turn and do some testing.

We will put this on here sometimes. Let it run for an hour and a half. Now we hook coolant up to it.

Of course you do that. Yeah, sure. And a pump and so forth.

But just to see what's going on. But that's what we're running into. And this one we call growler.

That's awesome. Because of the very pleasant noise it makes. So when it gets into a car it might transmit through the whole chassis and people hear us idling down the road.

They'll think we're a long block, big block Ford. That's right. You know.

That's right. So we'll see how it goes. I want you to drive it until it comes apart and then we'll drop it and take it apart and see if we can see how the parts fit together.

That's a challenge I'll be happy to take. Stay with us. So life is hard up here on the Canadian.

But we're getting there. And I'm excited about some of the things we're doing with the controller and the Tesla drive unit and interfacing it to various displays and so forth. That I think is going to get to be pretty good.

We also have quite a bit of... I've done a couple videos on the Tesla batteries. We'll do a video on that. Column will be here March 13th.

And I'm probably going to wait and do it then because we've got a lot of very good news. We now have complete control of the Tesla battery modules and its BMS. And we can read every cell and every temperature.

And not only that but we have worked out the wiring, procured the connectors. And we can automatically set up a string of up to 64 of these. Assign addresses, form a battery pack and actually control a couple of contactors.

It's taken some software, which is open source, and a little bit of hardware that we're working on or Colin's working on to do the very odd 600 and whatever it was 12 and a half kilobits per second on this isolated UART bus. And to do a couple of isolated output MOSFETs that'll do up about 7 amps to effectively close contactors and take action. And of course our little module will convert all that to CAN messages that'll be easier for you to deal with in your solar project or your EV project by CAN.

I've kind of tasked Colin and another guy, but I may do it myself, with devising sort of a simplified battery management CAN protocol set of standardized messages that avoid all this bit packing and kid engineer self-aggrandizement stuff. They just get too tricky by half. In an effort to save space on the CAN bus and processor time, we can do a lot of CAN messages with today's hardware on that bus and processor time.

The Arduino DUE has more computer in it than the computer I started publishing BoardWatch on. It's dirt cheap. But I think to some degree it also protects the mystique of the CAN hacker, CAN programmer in just unnecessary and frivolous ways.

Let's come up with a rational protocol to define multiple packs, multiple modules, multiple cells, report the cells and the temperatures and with high-low on all of that. In a rational standardized scheme that then anybody can use in any device to read a battery pack. We've also done some work here on the overall pack and we now have control of the contactors.

We can close the contactors, read all the temperatures, read all the voltages. This kind of forces us to have a second set of contactors because of the way Tesla times the pre-charge on their contactors. They know what capacitance is going into in the inverter.

Your mileage may vary if you have a Koda or Siemens or another type of motor. And so we almost have to strawman the whole pre-charge contactor thing. And so the other issue is the connector is unobtainium and if it was obtainium it would be extremely expensive, that high voltage connector.

So Bill is putting together a box that contains the controller, two contactors, pre-charge resistors, some LED lights, a switch, the connector bars that go in to carry the high voltage, the HVIL pin and everything into one unit. You flip a switch, it will bring alive the battery pack and do a timed pre-charge based on your criteria and close two contactors to connect you to your system. But it will also disconnect it under criteria you set for high and low cell voltages and high and low temperatures.

And so it's going to be a box that plugs into it in place of the existing connector and then you'll hook cabling up to that. And so you'll be able to use the whole pack. Bill says nobody wants to use the whole pack, they're 1,330 pounds.

Man, I picture that thing wedged up against a wall, it takes up no floor space. The thing is six inches thick and all that enclosure, which is heavy, is also quite safe because of that. It's kind of hard for little kids to get into the battery when it's all sealed up in that case.

And in the case of an earthquake or that sort of thing in your facility. So if you've got a forklift, I think it would make an excellent way to put down a 85 kilowatt battery pack for your solar installation. That'll run most homes for about 60 hours and that'll get you through some stormy days and basically put you off the grid for all practical purposes while remaining connected to the grid for the minimum monthly charge in case you need to use some of that.

So I'm kind of a fan of treating it as an entire battery system. We're going to swing both ways and I expect to be able to talk about that more. Colin will be here March 13th.

I think we'll just shoot for that video the whole thing on both ends. We do have some further work on the DOKA that Bill has done and videoed and I'm going to do a quick roundup of the DOCA with Bill Bayer and what he's been working on. Let's take a look.

All right guys, I wanted to take this moment to give you a little update on where I've been on the DOKA project. It's actually coming along pretty good the last couple weeks. I've been able to really get some time to spend on it which is the thing that a project like this deserves and needs.

I wanted to kind of show you what I've done but I wanted to also talk a little bit about actually what's involved with doing a build. There's 300 and something videos out there EVTV has done which are all there for you to look at. I'm going to kind of say the same things that they say too but it's good to hit the high points again.

It's easy to get distracted by the installation of a Tesla drivetrain or having stub shafts made and all these really exciting things, big welding projects and the things that really stand out because it's easy things to talk about and it's easy to see what's been done. The thing that will consume most of your time during a build are just little things. Running wires, putting terminals on, discovering that you've cut a wire three inches too short and now you've got to figure out a way to extend it a little bit or rerun your wiring.

If you're OCD like I am, I spend an awful lot of time just indexing the terminals in the crimping tool. I guess it works in my brain that it's a little bit better connection one way versus another but it's also keeping unity throughout the entire build. I'll show you a close up of some of the relays and some of the fuse blocks that I've put in there.

Just keeping things straight, keeping the left side of the relay is the ground for the relay, the right side is the positive, the front part of the relay is the output to the fan or pump or whatever. It's stuff that isn't really all that important for doing. It does take some extra time but where it's more important is for several years down the road.

We all like to think that we're going to build this truck or doker or whatever and that's going to be the car forever but you don't know. You could end up selling the thing later, trading it for somebody else for another vehicle to do another build or whatever. I like to think that Jack's going to just give me this doker when it's done but the reality is that he's probably not going to do that.

So 5 years from now, 6 years, 10 years from now, somebody's going to need to know what to do with this thing and I'll put together a wiring schematic and that's good but I try and wire things so it's a little bit intuitive. In this case, I've run separate wire colors for each circuit. So the inverter motor, the water pump has it's own color and the cooling fan for the DC DC controller and the charger, they've got their own color.

Things like that. Just to keep it straight so someone down the road can figure out what's going on. And that's the thing that's really time consuming and it doesn't really, you can't really talk about it at a show, at a car show.

Because believe it or not, you're probably going to take your car to a car show. You're probably going to take it to several of them and even if you don't take it to an official show, you're going to show it to people. You're going to show it to your wife or your sister or your sister's boyfriend or something.

You're going to show it to people. You're going to open the hood and you're going to point at things because they're not going to know what any of it is and you're going to be the expert. You're going to be the smartest kid on the block for electric vehicles, certainly for conversions.

And it helps to know what you're talking about and if you do your build and you keep it straight and you keep it nice and tidy, that means a lot to the person you show it to. Because they may not know what they're looking at, but they can tell if it looks like their car. You know, if it looks like it was intended to be there.

Because everybody's been under the hood of their own car to change their oil or check their filters or something like that. And they know what a car is supposed to look like under the hood. And if you open it up and it's kind of a mess of stuff, they'll appreciate what you've done for converting a car, but they'll know that something's not quite right.

Let's put it that way. So I just wanted to kind of take a few minutes to sort of talk about that. And now I guess I'll try and give you a little close-up shot of what I've been doing.

All right guys, I've got it zoomed in a little bit into the treasure chest here. Show us a little bit what I've been doing. You should be fairly familiar with the layout in here.

We'll have our two batteries on our sliding rails. And then over here on the right, we've got the charger there nearest us and the DC-DC converter on the other side. That cable you see hanging down there, that's a crossover cable.

So the way these batteries are going to work, the one battery pack on the other side of where we're standing on the passenger side, that's going to be the first battery pack and represents the most positive terminal of the pack. Then it comes across through that pack along on this crossover cable over to this side, feeds its way through that, goes through the bulkhead here on the left into the back seat, underneath the back seat to the third battery pack, goes through that and then that's the the most negative terminal onto the pack. And then from there it goes through the shunt, IVT and into the pre-charge relays, pre-charge contactors into the motor, the inverter.

Let's see if I can do a little Blair Witch right quick. Over there you see we've got a little distribution block. What that is is the black and the red wire you can see that comes from the J1772 that's in the original charge port on the passenger side behind the passenger seat.

I don't like having a charge port over there. It's kind of awkward. I prefer having it either on the front bumper or the back bumper just to make it easier whenever you do charge your vehicle.

But you know that's how it worked out for this. And we've got an interlock with the Tesla. The software on the GEVCU to run the Tesla there's an interlock so whenever you've got the J1772 plugged in it makes it so you can't start the vehicle.

You can't drive away because that can be a real problem. If we look up here, here's some of my wiring. We've got two fuse blocks.

This first one being the switched 12 volt. And then you see I've got all the circuit breakers marked, all the fuses marked. That way you can do troubleshooting later on down the road if the DC-DC controller doesn't work for some reason, doesn't have power, you can come over here and check it out.

I've got about a 5 amp fuse on that just so in case anything were to happen everything stays protected, we don't catch any fires. And then over to the left I've got four of the five, ended up being five relays on this. We've got a relay for a vacuum pump.

We've got a relay so the GEVCU can give a negative signal to that relay for the brake lights. Likewise for the reverse lights. And then there's also a charge enable, it's an interlock.

That's what I was telling you about whenever you've got the J1772 plugged in. The AVC module will give a 12 volt signal to that and it will remove the ground from the GEVCU and tell the inverter to not do anything. You've got a J1772 plugged in and eventually I'll have that same 12 volt from the AVC2, that'll power the original charge light in the dashboard.

So if you try and go, you can't figure it out, you'll see a big red light staring at you and you should figure out what happened. All right, so that is this side. Let me readjust and we'll go to the other side.

Okay, reset over on the other side. We're over here on the passenger side now. Again, as you're familiar with, we've got our Delphi DC-DC converter there and a charger.

And zoom up a little bit, show you our, what I've done with the, this is the 12 volt hot, hot at all times side. So this is going to be, this will power the relays that I want to have running at all times, or at least have the ability to have them run at all times. The AVC goes in here, the fan motor for the DC-DC and charger, they go in here.

That kind of presented a bit of a problem for me. Um, we need a way whenever you have a key on to have the DC-DC converter will be on, of course, it's water cooled. So we need a way to have the, uh, the water pump for the cooling system on.

We've just got one circuit running through the charger and the DC-DC. So we need to have that pump on, that water pump on to keep the DC-DC cool. Um, but we also need to have a way to turn the same pump on whenever the J1772 is in, whenever we're charging.

And that was going to be kind of a problem because we couldn't run it to the same, uh, power source. If we ran it to the key on, then whenever you plug the J1772 in, it would feed power to the pump, but it would also back feed power, uh, backwards into the key on, and it would turn all your accessories on that are normally not on when the key's off. So what I ended up doing is I came over here and I've got two relays for the, um, the charge pump.

I've got a charge pump relay and a DC-DC pump relay. And so what I've done is I've run the, the inputs to the, the relays. So the low side, call it the low amperage side of the relays.

Those are separate. Those are, uh, I've got one that goes to the AVC2 module, and then one that goes to the key on. When the key's on, it'll activate one relay, and that'll get the pump going for the DC-DC.

When the key is off, but then you plug the J1772 in, the AVC will run power to the charge, or run power for the, for the water pump for the charger. And then the outputs of those relays are just in parallel. That way there's no way to back feed 12 volts where I don't want it.

So that's how I ran that. That's why we've got a whole mess of relays over here. We've got a, the AVC module is just there.

And then this is our 12 volt hot at all times relay. And if I zoom back, you'll notice that this side looks remarkably like the other side. I even went to the trouble of making the front of the fuse box go forward.

Um, so they're both oriented in the same direction. And that I didn't do for any reason other than whenever this does go to a show, there's going to be one guy who will notice. There will be.

There's one guy who will come by, look underneath here, and if it wasn't like that, they'll, they'll turn their nose. How do I know that? That's because I'm that guy. So I did that for anybody out there who's like me and they will appreciate it.

And likewise, uh, with the, with the little loops of wires on the ABC two, you'll see that they're about the same. Well, they weren't, but I had to fix it cause it was just driving me crazy and it's stupid and it takes time, but it's worth it to me. And it'll be worth it to somebody who sees it.

Um, so yeah. And then let's see, you see, I've got, uh, the, the, you know, the bulk of the wires they're coming down and I've got them in cable ties and then they run along the side of the, the side of the truck here on this. Well, you can't really see it, but there's cable ties holding it down along the side of these rails here.

And there's some on the other side and, and then the wires, they all feed along the top and they're all held in by cable ties. All right. I wanted to quickly introduce you to a tool you may not have seen before, but it's one that I find, um, really, really helpful.

Uh, anytime you've got to put a clamp, um, and there's no way to get behind it. So you, you, you've got to secure the clamp down. I don't really like using self-tapping screws.

Um, they kind of invite corrosion, water gets in there. It'll, you know, eventually it'll rust everything rusts. Um, and eventually it'll just lose its connection.

I certainly don't like them for, um, using as a chassis ground. Um, but really they're just kind of a pain. They're, they're just, they're not elegant, I guess you would say.

Um, so a tool that I've used, um, with previous builds and restorations that I've done, it's called a, um, what's it called? Rivnuts, rivnuts. So what it is, is you've got this tool here and it's got a threaded end on it. And, um, what you do is you take one of these, they're called rivnuts.

You take one of them, you screw it on this end here. Well, if it's the right size, you would, you would screw it on the end of there. Uh, you'd pre-drill a hole and then you'd put this inside that hole.

And whenever you pull this back, it works just like a rivet. And it just, it basically gives you threads, um, just blind threads that you can tighten things in. These are made of aluminum, so you don't want to make it structural.

It's not going to hold your motor down or anything. But, um, for things like clamps and, um, uh, fuse, uh, fuse blocks and relays and just small things like that, uh, this just makes a really quick and convenient way to cleanly, uh, install your components. And they're reusable.

Uh, I guess there could be some electrolysis between the aluminum and the metal. I'm not too sure about that. Um, I don't really, I mentioned grounds before.

I don't really use these for grounds either because being that it is aluminum, the aluminum to the, to the ground, the metal of the chassis, I'm not sure if there would be some sort of a reaction to that. So I don't, I don't use these for grounds either. But, um, but just for simple fastening things where you, where you can't get your hand behind it, um, pretty easy in maintenance.

Um, all right, that's over here. Let's look at J1772. All right, so up here at the right front of the car, I mentioned before the J1772.

We've got it where the original fuel filler was. I don't like having it on the passenger side of the car, uh, just kind of invites forgetfulness. And it just, it doesn't help you remember to unplug the, unplug the charger when you're done.

But you know, we've got the interlocks in there, so there's really, there's not going to be any way you could drive off with it there, but it is kind of a convenient, because you're going to get in there. You're going to try and start it. You're going to try and go, it's not going to work.

Now you're going to have to get out. You're going to have to walk all the way around, undo it. I know these are real big problems that the EV community has, but there it is.

But yeah, so J1772. And then if we get, zoom in here kind of closer, uh, I've actually reused the original, um, I've reused the original, uh, filler tube, the plastic filler tube. A reason for that is, uh, you know, the front tire's sitting here.

So it's going to be kicking back, uh, rocks and mud and you know, whatever. And, um, it's just going to, it could invite, um, it could break the, um, the insulation on, on the wiring, uh, and cause who knows, frame leaks or, uh, shorts or whatever. So it's best in an application like this to keep things as, um, protected.

All right, guys. Move this underneath the Doka now. What we're looking at here is a vacuum pump.

Got it mounted in a little bracket. It's got some rubber mounts on there just to keep vibration from the pump from transmitting through the chassis, making a bunch of noise. This particular pump is made by Dommel, D-L-M-E-L.

You can pick them up online. Summit Racing's got them for about 250 bucks, but it's really nice. It's really, really quiet.

A lot of the pumps you'll get are sort of diaphragm pumps, um, and they just make a, well, they, they, they're just noisy. If you've ever seen like a pancake, um, air compressor, uh, it's that sort of noise, just that brrrr. They're really obnoxious.

But this one is a, um, it's got veins in it. It's a vein pump. So, uh, it doesn't make very much noise at all though.

We've got this mounted back there. This is the, uh, original, uh, line that goes to the brake booster. Uh, this here is a one-way valve.

So whenever the pump creates the vacuum, the, the one-way valve prevents the vacuum from leaking out, um, or actually, actually prevents atmospheric air from leaking in, I guess you'd say. Um, so that's what that does. And then got the wiring going into the main harness.

Speaking of wiring, all the wiring that we saw in the, um, treasure chest, that comes down here. Um, I was talking about things that are time consuming, but are really important. Um, making little labels.

This is the, the, uh, signal, the negative, uh, signal that comes from the GEVCU to, then it goes up to the relay to make the reverse lights come on. This is the, the coolant fan. This will go to, uh, some sort of a thermal switch.

Uh, this will be for the motor and the inverter. Um, and we've got, you know, just, this is the brake. That's the brake.

This is, uh, the DCDC fan ground. So this will go to the, um, the cooling system of the DCDC and inverter. Um, and it'll turn the fan on whenever it gets to, you know, the, the, uh, temperature that we've set the thermal switch for, uh, this, these wires here.

I'm going to have a JLD404 in it. Uh, we're going to have a display, an EVIC display, uh, for the Tesla and it'll have, you know, it'll have, uh, amps and voltage and all that stuff on there too. But I just really like JLD404.

It's, it's a dumb little gauge, uh, just, you know, but it's, it's really functional. It's got the amp hour counter on there. Um, I've gotten used to it in my truck and kind of every vehicle, uh, in the shop here has one.

And it's just a really useful, um, it's a really useful instrument for, for your vehicle. Move forward here a little bit. What we're looking at now is the vacuum switch.

Um, what that does is, uh, it turns the vacuum pump on and off. On this particular, on this Doka, I've got the vacuum pump wired into ignition. So when the key is on, the pump comes on.

Um, and when enough vacuum is reached, this switch, um, turns the pump off. On my truck, I've got it wired a little differently. I've got it wired so whenever I hit the brake, the brake light comes on.

And when the brake light comes on, that turns power to the, to the vacuum pump. And I don't, I, it worked for my, for what I'm doing. But the problem is, is the first time you hit the brakes, you don't have vacuum assist.

So there really needs to be a, it really needs to come on when you turn the key on. Um, so you get your vacuum assist instantly. Um, it's not going to be a, a noise problem on this because of that vacuum pump I showed you is really quiet.

Um, but you know, even if it is a problem, it's, it's not that big of a problem. Let's be honest. Um, so that's what that is, uh, up underneath here, zoom out a little bit.

Those two terminals right there, that comes from underneath the back seat. That comes from the, uh, well, that's the, the, the outputs for the inverter. Um, so that'll come back to the inverter for the Tesla drive unit.

What else do we have down here? Oh, over here. We've got the, that's the DC, DC and charger. Um, radiator, coolant, radiator.

Um, I get to run lines. I'm going to run hard lines. I've got hard aluminum lines that I'll put a and fittings on it.

And, uh, I'll run those back to the pumps back to the pump for the, for the cooling system. And then finally, I'm hung. That's what she said.

And then finally, we've got probably too much light. Um, we've got one of the pumps here. That's going to, that's the pump that's going to be for the DC, DC controller and the charger.

And swing around on this side, we've got another pump. That's the pump going to be for the inverter. Um, straight above it, there's going to be our catch cans.

They'll be right above. I'm thinking at this time, they're going to be accessed through the access cover that's in the bed. I would rather have them accessed through this, um, this sort of door here.

This is what was originally you would get to the, um, the radiator fluid and, um, power steering pump or whatever. If it had one, you'd get to all that stuff through there. Uh, the problem with that is it's not in order for me to get the reservoirs high enough so that we don't end up, um, with air pockets in the lines, uh, you wouldn't be able to access the fill.

You wouldn't be able to, you wouldn't be able to fill them. So I'm going to have to have them mounted up a little bit higher, which isn't a big deal. Um, you know, on a, on an internal combustion engine, you have to, you know, worry more about the cooling system than you do on a, on an EV.

They don't get that hot. They don't really, they don't get pressurized. So, um, it's not really a problem.

And with AN fittings, it won't leak. So we should be able to fill it. All right, guys, here we are back under the doke again.

I wanted to give you a little brief update, what we've been dealing with, uh, this week. Um, when you're doing a build like this, sometimes you get really good ideas and sometimes they work and sometimes they don't. If you remember back a few months ago, when we started the doca build, Jack and I were standing in front of it.

And I suggested that we use the original Tesla radiator to do all the cooling, you know, to fulfill the cooling needs for the inverter motor. And I had found a pretty cool place in the back of the DOKA. It looked like it was going to fit sort of like it had eyes.

It was really nice. Um, however, once I put the frame around the radiator and got the coolant fans mounted, uh, it really wasn't an option anymore. It could have gone back there, but it looked horrible.

Uh, it would have been big and bulky and just in the way and, uh, I didn't like it. So, um, went looking for other places, other locations to put the cooling system and there just happens to be a Tesla Model S radiator sized spot, uh, pretty much underneath the, uh, the left side of the car, as you can see here, um, right between the frame rail and the, and the, uh, unibody chassis rail on the sides. And it just fits perfect.

It really, really looks nice. It, um, it tucks in there nice. I was going to have it, you know, you can see how I got it just sort of held up by some wire at the moment to sort of mock it up.

Uh, I'm going to have it at like an angle. There's plenty of room in the back for the fans. There's enough room for the, the air to come out of the fans and escape.

Um, and it just would have been pretty good. The hoses down here, uh, they'd be out of harm's way for the most part. Um, and I could have run that over and, and just tied it in with the cooling system.

The problem is it's really close to the back, uh, front tire. And as you're driving down the road, if you were to take a left turn, the tire would basically just spit up rocks and grit and dirt and water and mud all over the radiator. And, um, probably would have been okay.

I doubt it would have damaged the radiator. The radiator is normally in the front of the car where it catches all that anyway. Um, and I could have put like a metal grate over it.

Um, you know, I could have, I could have done things to maybe make it work, but something occurred to me that occurred to Jack a long time ago, but I, uh, I wanted to, I wanted to use the Tesla radiator as much as I could, but it's, it's just not going to work out for our application. Um, so I just decided that, uh, I could use the normal radiator. Seems pretty obvious, but, um, so I went online and found out, you know, what's involved with bleeding the system.

And, uh, as it turns out, it's kind of a pain in the butt to bleed, um, Vanagon, Doka cooling systems. Uh, there's a bleed valve that you have to have on and they say you need to have the engine revved up to 2000 RPM. And, uh, it's kind of a pain in the butt, but, uh, one of the things I think that I've done to maybe make that a little easier on us is I've got our coolant tanks mounted about a foot and a half higher than they are in the actual car, um, when it has the ice engine in it.

So that should help us out a little bit. Another thing that'll help us out is that radiator is designed to cool a two and a half liter or a 2.1 liter water cooled four cylinder engine, um, which definitely puts out a lot more heat than our, our Tesla drive unit ever will. Um, especially considering the only time our driving is going to put out any heat really is when we're going down the road, uh, with a tremendous amount of airflow over it.

So, and then being out in front, obviously it's going to catch all of that. There is a fan, there's an electric fan on the front of it, on the, on the, on the back of the radiator, um, that pulls just a tremendous amount of air. Um, but the reality is I don't know that we're ever going to need it.

Because like I say, the only time we're going to generate any heat is while we're moving. Um, and in the times we do need it, I'll have a little, a switch on there. We'll either get the JEVQ, uh, Jack's kind of working on it, um, thinking about how he can put a, an output, uh, for coolant, um, for a trigger for it to turn the coolant fans on.

Um, or otherwise I'll just get a little thermal switch. Um, I found some online actually for not too much money. We'll see how those work.

But at any rate, um, we're going to use the radiator. Let's go take a look. Okay.

Up at the front of the vehicle now, uh, hopefully I'm kind of in the frame. I'm, uh, say, so we've got the Doka radiator mounted up here in the front. Um, got the normal inlet and outlet for the radiator.

Uh, traditionally they would go all the way back to the engine through either a steel some steel tubing, or in the case of this one was some plastic tubing, uh, the plastic tubing over time, um, just disintegrates. And we had a, uh, you know, half of, half of the tubing actually stayed inside of the radiator hose when we took this thing apart initially. Um, and we don't really like using stuff like that

We, we prefer to use AN fittings and, uh, AN lines, and I'm going to run the coolant system with eight AN, which is basically half inch of tubing. And, um, so the way I'm going to do that is my friends over at McMaster car, they sent me a piece of one and a quarter inch pipe, and that'll just slip inside of here. I'll cut this off, you know, the appropriate length.

And then I've got these, these little welding bug bungs from, um, from summit racing. And then what I'll do with that is I'll take a piece of, just take a piece of metal plate like this, drill a hole in it, weld that metal in, weld that bung in there, and then take this piece and weld it onto the back of the tube, grind it down so it looks decent, and then slip that in there. And that will allow me to have the AN, uh, line run all the way back.

I bought, I got some, um, some aluminum, some half inch aluminum pipe, aluminum tubing, and that's what I'll use to run it all the way back. The aluminum tubing is not only cheaper than the braided line that we normally use for our cooling systems, but it's also more durable. You know, we're going to be underneath this thing, so there's going to be rocks and stuff flying up and hitting things.

And also, as the coolant's flowing through this aluminum tubing, the aluminum is going to dissipate a tremendous amount of heat, even before the water gets up to the radiator. So, when we've got two runs of aluminum, you know, we've got a run of aluminum pipe coming all the way up, this gigantic radiator, a huge fan, the whole system designed for a, you know, a two liter ice engine, I think cooling is not going to be an issue for us. Which kind of works out nice, because as it turns out, cooling is, um, the Tesla's pretty sensitive to cooling.

Yeah, let's leave it at that. It's a pretty sensitive system, and cooling is good. And even if we run into a situation where, because we don't have the same flow of coolant through the, you know, the inch and a quarter pipes and all that that the ice engine does, if we can't even fill up this entire radiator for whatever reason, if I can't bleed it all the way to the top, it's really not going to matter because it's way, way, it's way over capacity for displacing heat.

So, we're going to be just fine. And that's cool. I'd rather, overkill is always an option.

I think that's what we say around here. So, that's where we are. Electric fan up here.

When that thing's going, it's like an airplane moving. It's ridiculous. We've got our, we've got stuff falling.

We've got our heater hoses coming down here from the, from the heater core. I think what I'm looking at doing at the moment is getting one of those Eberspracher, even Spacher, Eberspacher, whatever, those heaters from the volt and plumbing that into the system. I've got plenty of room under here.

I could put another little coolant loop under here, run it through the Eberspracher and up into the heater core. Getting air bubbles out of that might be a problem. I don't know.

I'm not there yet, but that's an option. Another option is just one of those resistive heaters. One of those ones that we've been selling for, you know, forever.

They're fantastic. I got one in my truck and it gets warm right now. It's really good.

So, we'll see. We're not quite there yet. That's where we are on the front.

So, all right, here we are at the drive unit, this time out of the car. I've taken it out so we can put the Quaife torque biasing differential in there. I also want to get in there and just kind of look over everything, make sure the bearings are good.

We go ahead and replace both of the bearings on the differential, just as a matter of course. Once it's back together, we'll go ahead and replace the seals, the oil seals on the outside of two. No real big deal with this.

We did a video of it back in July. July? I think it was July. I've taken this thing apart.

It's just a whole bunch of screws and then you whack it with a rubber hammer a couple times and it splits. It's pretty straightforward. Going back together, it's not too hard either.

It's a little bit tricky because you've got to get the bearings lined up with where they seat in the case and they kind of have to go in there straight. If you get them cocked or crooked, it doesn't really go together very well. If you get it right, it just kind of slides together.

It's good. We've also got some sort of longer bolts. Whenever we have it together, we can use those longer bolts to go around the outside of the case and sort of cinch it in together, make sure it all works.

There are some oil seals inside here. This one goes on the back by the electrical connectors. I'm thinking it just reduces the chance of oil flinging off of the motor gear back here spinning around at 16,000 RPM.

Just reduces the chance of oil getting past it into those electrical connectors. And then there's a bigger seal up here that goes right there. That's for the water jacket and then a smaller seal here that seals this water coolant jacket as well.

But I go around with just a straight edge and kind of touch up all the surfaces and get all the old gasket material off. They just use RTV at the Tesla factory, so that's kind of handy. I remember back in the days of ICE engines, people would put gasket glue on and end up ripping the gasket in half and you'd spend all day just trying to clean that stuff off.

You do have to be kind of careful because the razor blade itself is actually a harder metal than the aluminum is. So I've got to be real careful to keep it nice and flat so I don't gouge the mating surface. Once it's all cleaned up, I'll get all the old oil out of the inside and then go around with another bead of RTV and put it on together.

But I don't know that we've actually shown the ATB assembled with the ring gear on it, but that's basically it. There's not really much to look at. Okay, the big news for this week is definitely something we got in earlier.

It's from our friends over there in China. If you remember, we had some issues with getting the stub shafts made to adapt the Tesla output to the Volkswagen CV joints. We contacted numerous manufacturing, engineering, fabrication shops looking for anybody who can do some machine work for us.

We had some drawings. We had some pictures. We had a lot of interest to help, but when it actually came down to doing it, we came up with nothing.

I had one company, Summers Racing out in California, that were all on board. They sent us some beautiful drawings. Then they had some coronary, I don't know, farction or whatever you call it of their computer system and they lost all their data.

Apparently, they lost our email and our phone number because I've sent numerous emails to them trying to figure out what they're doing. We've sent them some samples and they've just gone off the grid, which I kind of admire, but it's a different thing. Anyway, Jack went online and did what he does best and found some nice Chinese people that would help us.

They emailed back and they said all the right words and we decided to go with them. We said, all right, go ahead and make these things. We sent them a stub shaft from the Tesla and the little spider gear from inside the Eaton differential so that they could get the splines right.

That's all we sent them. They came back with some beautiful, drafted up drawings and we said, go for it. They did and they sent them back to us and we got them here Tuesday.

We put it in, the splines fit perfect, the distance, the depth, everything about them is great. Here's what they look like. They are really quite nice.

They're nicely machined and hardened and they look really good. The only problem was the holes for the CV joint. This is the outer part of a CV joint just removed from the drive shaft.

That basically goes right here. It's got holes that are on 86 inch centers. Whenever we got this shaft from them, there was only these six holes here, which are on 90 centers or 89.5 or something.

To this day, we're not exactly sure where they got that number. I don't have any idea. They just thought that'd be a good place for it.

It might be the bolt circle for the Porsche 930 axles. I don't know, which may come in handy later. We'll see.

Anyway, our good friends in Cape Girardeau over at Cape Precision Machine, Lucian and I'm terrible with names when you're trying. Tony, Lucian and Tony. They hooked us up and I brought them the CV joint blank and told them what we needed to do.

They hooked us up in just a day. They had the holes turned at the right spots. And as you can see, it fits and it fits beautifully.

It's just fantastic. Yeah, that's all I can say about it really. It just worked out really well.

You see, we can turn our wheels here. We've got the frame jacked up so the suspension arm is at its maximum travel point. So this is the farthest that the transmission will ever be in respect to the axle as the tire comes down.

So that's the farthest that distance is going to be. And you see, we still got some good travel there at the opposite end of it. What I'm going to do later on is I'm going to get this tire off and the spring out and I'm going to push the swing arm all the way up and that will represent the closest distance between transmission and tire.

And as long as I've still got some travel when it's at that position, then I'll know that everything is fine. The kind of interesting thing is the distance between here and here on a Tesla is just about the exact same as the distance between here and here on a Type 3. It would have been a Type 3 Volkswagen transmission. So basically, you know, the measurement was fairly easy and I think we've come up with a winner.

I think it's going to work really good once we get this thing going. I guess the bet is still on if the Volkswagen subchassel will hang with the power of the Tesla. So we'll find out.

But it's pretty exciting times. Yeah, it's good stuff. And here's something that Jack found online.

Not really sure where he found it. I think he got it on Amazon, but it's just a jack. It's a jack, but it also has a jack stand, but it also has a jack on it, which is just what a good idea.

You know, I mean, it's just a great idea, especially in this kind of application where I want to lift the truck up. So I'd have to bring a bottle jack in or lift the floor jack up on the lift so I could pick this tire up and then get a jack stand to put underneath of it. I mean, this is just an all-in-one thing.

The only thing I'm not a big fan of it is we got this little, we got this lifter bar, which is just, it's going to get lost. Anybody who spent more than about 10 minutes here at EVP knows that when you put something down, it's kind of gone within a half hour or so. We'll see.

But good stuff here. We're moving along on the Doca project and having a good time with it. All right, guys, that was the walk around of the Doca project.

So you see where we're at so far. I did want to take some time just to explain to you about building vehicles. I know you guys are out there doing it.

We don't really get a lot of updates from you guys. It'd be really nice to have some videos to see what you've been up to. But you're in the process of building.

It doesn't need to be a great video. If the only video camera you got is on your camera, that's fine. We'd love to see what you've got.

If it's just a whole bunch of pictures and you want to write something up real quick, we can read it off what you write and show the pictures while we're doing the show. We'd love to see what you're doing. It's an inspiration to us.

It's certainly an inspiration to everybody else. It's kind of easy to forget that there's other people out there doing the same thing, and they all need help too. If you've run into a particular problem installing a component, let us know what you've done.

Let us see what your solution was. We'd really like to hear about it. But as for the Doka, now I've pretty much got the 12-volt wiring done.

I'm going to have to connect all the wires on the bottom, connect it into the vehicle interface for the JEV-Q harness. And then it's a matter of going up in the cockpit, up in the driver, up in the upstairs and figuring out where the EVIC display is going to go, where the Power Key Pro buttons are going to go, the JLD404, where are we going to put that? And that's the real critical part. The stuff I was showing you, the little dumb things that I do, one person out of a thousand is going to see that.

Placement of the EVIC display, every single person is going to see that. So it's pretty critical to have that right, to have it look good and to have it... It's a driver interface, so it needs to be in a friendly place for the driver. And so that's what's going to consume my time for a little bit now.

And I think that's all I've got for you now, guys. Stay with us. That's our show this week.

I'm hoping it's, again, it's about an hour long, but I'm afraid it may be closer to three. As everybody knows, we do long, boring videos at EVTV. They're fun for us and for a few more.