RAID ON LIVONIA: The Challenges and Opportunities in Electric Vehicle Components

We failed to produce a show last week. This is partially a function of being under the weather a bit. I've had lung crud that has weakened me considerably. We had a bevy of daughters arriving for Thanksgiving celebrations. And Richard and Mike went to Detroit to manage the loading of a large and heavy body of electric vehicle components from the Azure Dynamics liquidation.

The demise of Azure Dynamics is the latest in a string of tragedies involving dedicated talented people devoted to the cause of bringing electric drive to our central need and desire - mobility. It is a rather encapsulated solution to a number of truly horrendous problems, but oddly difficult to implement. The "automobile" is so centrally entwined with our very daily lives that change comes hard. And whoever tackles the problem always somehow winds up coming in $20,000 too high.

For over a hundred years, the last and final refuge of the electric vehicles has been in local commercial delivery. We just had a visit from AMP Electric Vehicles where they announced they are basically phasing out their passenger vehicle efforts in favor of largish delivery vans.

AZD largely had it right. A small, economical delivery van purpose built by Ford for the market, artfully converted to electric drive, and with a relationship with Ford that would allow support from Ford dealers for the fleet. It was a carefully thought out approach and in many ways artfully executed.

In the end, my sense is they became to OEM like, costs got out of control, they became management heavy dealing with governmental compliance directives and Ford, trying to establish a global footprint, etc. For whatever reason, they rather suddenly ran out of capital.

And at $65,000, the little van again represented an underwhelming value opportunity.

I guess I was a little surprised Ford didn't step in and either take it over or bail it out. It was technically a Canadian company, which may have made that difficult. And they may have seen the value proposition similarly.

Frankly, I had no interest in this auction. Several viewers alerted me to it, but HGP's terms and conditions were just abusive, there was little means of telling exactly what you would be bidding on, the logistics were frightful. They wanted insurance and union riggers and no end of things just to move stuff out - at least on paper. It all had to be out about 30 minutes after you bid on it. You could not even tell you had won an item and indeed they were under no obligation to actually SELL it to you after you had won it. I simply passed on it.

But we had a whole other waive of viewers insisting we participate in an almost non-stop series of e-mails. A lot of the stuff was in large lots of 10, 15, 20, or more units. It wasn't broken down very well. It was a quick and dirty liquidation.

So I caved. Suddenly, I'm on the hook for about $120K I didn't precisely have laying around looking for a home. EVtv has been expensive enough and oddly lacking in cash flow after three years of pretty serious effort and two conventions. So on top of lumbar pneumonia, I was a little angry with myself for having suckered on this mess and it was a pretty surly several days in the EVtv Motor Verks shop. We're seriously behind on this Speedster for our friends in Tokyo. I've never really finished the environmentals on the Elescalade. And now a ton of cash out the door on who knows what kind of a lark. I was not a happy camper. And when I'm not happy, I am apparently very adroit at sharing with others.

We bid on an incomplete Transit Connect largely because it had a Johnson Controls battery on it and I got the thing for $6500. If it DID have a 28kWh battery on it, I can't really build one for that. And it might be quite important to see how they actually wired and configured things on the vehicle if we were going to deal with the components.

As it turns out, it was "incomplete" in that they had not done the final diagnostics and title paperwork. The vehicle is actually complete and purportedly runs. I haven't actually seen it yet but it should be here this week.

Siemens motors are always a conundrum. There was actually a bunch of them available from the Ford Electric Ranger fiasco during the last CAFE standards battle in California. But they had a large helical gear shaft on them that mated to the differential, an open front bearing allowing lubrication from the differential, and of course no differential. As it so happened, no inverter either. I watched these for years on eBay with longing as they gradually all went away at $2000-$2500 each. But too many problems for me.

I fear we may have repeated the offense.

The motors we received were actually miscounted. Instead of 56, we got 65. We didn't pick up somebody else's, the lot was just mischaracterized. The motors came on pallets in brand new sealed cardboard boxes. And as best I can tell, they are just excellent motors and ideal for conversions.

We did pick up 56 DMOC645 inverters. But I don't know they will ever be usable. First, AZD had continued development of this inverter and it truly looks OEM class now - nothing like the DMOC they used to sell with the elevator motors. But along the way, they had shed the CONTROLLER function and just reduced it to a CAN bus controlled inverter - moving all controller functions to a separate multi controller, the Vehicle Control Unit. (VCU).

But wait, it gets worse. Apparently they quit installing even the inverter software in the units because of mixups in software revisions. So they would install the bare inverter on the vehicle, and then install the software later when everything is in the vehicle. As we got them new in box, there is about a 100% chance that they have ZERO software in them at all. I've ordered the dealer diagnostic harness and software from AZDtec, but it's unclear to me that this will solve the problem. I don't know where to GET this software, or how to flash it into the EEPROM or flash memory of the Inverter if I had a copy in my pocket.

We're trying to talk to some ex-AZD employees and consultants, but it's unclear at this point if these will ever be anything but very stylish billet aluminum liquid cooled boat anchors featuring excellent power electronics components.

The Brusa Chargers were what really caught my eye going into the auction. My first EV had a Brusa charger. It was a bit of a challenge to install and configure. But after I'd gone through the process, I rather liked the result. It is endlessly configurable. Very small and light for its power. And obviously OEM class hardware. I actually have them in the Mini Cooper and the Spyder 550 as well now.

We bought 48 of them and they are brand new in box. Unfortunately they come with no cables at all. Fortunately, I think I've chased down almost all the piece parts for the cables. You always did have to wire up your own 23 pin Ampseal connector on the Brusas. Now the battery and mains cables as well. But I think I've chased down all the pieces down to the NTC thermistors for temperature sensing.

We paid $2900 for our first Brusa. They had gone up in price to almost $4000 for awhile, largely the exchange rate between euros and U.S. dollars. They're out there now at $2900. We will price them at $1995 and we've already moved six of them so I think they will go pretty quickly. They're isolated so you can gang them together for more power. And we've chased down the latest documentation and software.

Aside from the Transit Connect vehicle, the other surprise was the eGearDrive transaxle from Borg Warner. This was introduced a few years ago and has become THE transaxle for electric vehicles. Aptera was using it. Tesla Roadster was using the 03-01. Coda uses it. Obviously the Transit Connect was using it. It is light at just under 60 lbs. This one features an 8.28:1 gear ratio with Park Neutral and Drive using some very efficient helical gears and a splash lubrication system. It can do 200nm continuous and 300nm peak at up to 14000 rpm. And you can read the control lever position over CANbus to interlock other things in the vehicle.

This kind of gets you out of the adapter plate business. The Siemens motor just bolts right to it and the eGearDrive includes a coupler to neatly connect the motor shaft to the transmission shaft. You have to have splined shafts made to go from the eGearDrive to the wheels, but this is so commonly available you'll have no difficulty doing that locally.

So we have a superb motor I think can do peak accelerations at up to 150kw that is liquid cooled. A gearbox to go with it. A great little charger. And a DOA inverter that poses mostly a problem.

Predictably enough, I had already caught the forum ragpickers complaining about our pricing before we decided on it, much less posted. It is just too precious for words. This attitude has actually strangled the embryonic industry of those trying to develop products for the hobbyist/custom converter crowd from the beginning. And so I have no sympathy for or indeed respect for these bottom feeders. They're a negative to progress in electric vehicles in every sense and nuance of the word. A plague and a pox on the movement - entirely in it for themselves in the most craven fashion. I watched in Nausea as they literally drove Otmar and Zilla out of the business years ago.

I would like to enable our viewers to build better electric cars more affordably. But I'm very reluctant to kill all sales of other motors and controllers by doing so at prices that make life non-tenable for those trying to serve this market. So I would rather trickle charge these onto the market over a year or two than to dump them in a month. While this puts me a little underwater, I don't need a quick buck that badly. So it's kind of an impossible situation.

The controller is of course the problem. We need to locate, kill and eat the software for this thing. I have no idea what that entails. We're talking to anybody that will listen or respond.

That takes us to the Vehicle Control Unit. If we DID have software in the box, we still need a Vehicle Control Unit to interface CANbus torque commands to vehicle inputs such as throttle and brake inputs, interlocks, ignition, etc. It would be nice to also have outputs of battery voltage and current, motor and inverter temperature, rpm, etc both to drive analog gages and in CANbus format for Android displays and so forth.

I'm working this on a couple of different levels. We have contacted the company that actually manufactures the AZD Vehicle Control Unit and asked for a cost to develop a slightly more generalized device that would drive this inverter. If the economics of this make sense, we'll do that.

One of our viewers, Paul Dove of NASA's Marshall Spaceflight Center in Huntsville has volunteered to develop one there for us pro bono. Claims they have some talent on hand to do it. I like all that.

My nephew Derek is a superbly talented C++ coder. He's so good I actually just like to read his code. Very elegant stuff. Succinct and to the point, but clear - not convoluted. He actually groks the zen of object oriented design and makes it look easy. He's offered to do it as well.

The guys at RechargeCar have actually done an Arduino Mega knockoff with a couple of handy differences. They've put some basic protections on the inputs and outputs and put a CAN bus ON the board. Most of the Arduinoites are curiously unaware that their "sketch" thing has C++ underneath and can be fully object oriented. It doesn't have the STL library, which is a kind of limiting omission, but I've actually done some very object oriented stuff on an Arduino complete with the customary syntactic sugar C++ guys are familiar with. It actually works a champ. Arduinos can run some really bad code. But it's not actually REQUIRED as part of the spec. That's just what Arduino guys mostly DO.

So I'm thinking to add an ethernet shield and do all configuration via HTML form after the fashion of the EVnetics Soliton1. Make the CANbus interface object oriented where you can drop different CAN message digests into it to drive Rinehart inverters, Tritium inverters, Siemens inverters, SevCon inverters, and hopefully yes, this DMOC645 simply by loading a different message digest. This kind of assumes they all do exactly the same thing. They don't. But they do mostly the same thing in the same way. We can minimize the pain anyway in moving from one to another.

VCU's are historically proprietary and dedicated to a specific vehicle. I'm thinking open source and more generalized for any conversion vehicle. Open source hardware and open source software.

If we establish that we can sell AC motors and gearboxes to the satisfaction of Seimens and Borg Warner, and if we pull off this open source VCU, I think this pretty much opens the door to liquid cooled OEM quality AC drivetrains at more attractive prices. It even opens the door to actually making useful the coming cornucopia of OEM drive trains that will be showing up on eBay as these new cars get cycled through the wrecked car junk yard blender of actual use on America's highways.

Hopefully this knudges the movement toward universal adoption of electric drive an inch or two forward.

But we need help. If you can scout up any information on these components, particularly the DMOC645, you can certainly help us nudge.

In the meantime note that our Netgain and HPEVS offerings are pretty much plug and play. These AZD components are for the technically adroit with a hearty sense of adventure. They just are not yet there for the faint of heart.