The scandal started with the International Council on Clean Transportation attempt to show how clean diesels can be. They contracted with West Virginia University’s Center for Alternative Fuels, Engines, and Emissions (CAFEE) to actually perform some real-world testing. But the results they kept getting didn’t match the excellent numbers generated in the lab by the California Air Resources Board (CARB). After repeating the tests several times, the CAFEE guys determined that there had to be some kind of gaming going on with the software.
Note the total absence of the U.S. Environmental Protection Agency in all of this. Some days it is unclear just what this organization DOES now that they’ve started generating their own mining ecological disasters, but they aren’t actually doing any tests of cars. Indeed, they were busy filing comments with the U.S. Copyright office AGAINST the exemption to the Digital Millenia Copyright Act for automotive maintenance and repair. Their fear was that individuals would improve the fuel economy and power of their vehicles at the expense of emission standards by tinkering with the firmware. Point taken. But instead of individuals, it appears Volkswagen, the largest automobile manufacturer in the world, was the culprit.
The U.S. copyright office did a remarkable thing for a government agency. They did the right thing. And they did it in a very cunning way. On October 28th they announced they had GRANTED the exemption. We may reverse engineer vehicle software at abandon without fear of legal intimidation from the automobile manufacturers under the aegis of the DMCA.
Noting that it was not the copyright office’s job to enforce DOT or EPA regulations, they did offer to delay the start date of the waiver for one year to give other government agencies time to figure out how to enforce THEIR laws, which of course we should in theory still be in compliance. http://copyright.gov/1201/2015/fedreg-publicinspectionFR.pdf
Many have decried this one year delay, rather missing the point. The delay is essentially a tongue-in-cheek smackdown of the EPA’s “interests” here, knowing full well it has no impact on anyone anywhere ever.
The real world implications rely on the fact that there have been ZERO cases of any litigation of any kind anywhere over automobiles and the DMCA -ever. And there was never GOING to be.
In our topsy turvey legal “system” if you want to grant it that status, the entire issue was over the THREAT of litigation under the auspices of the DMCA. The DMCA was all about DVD movies, not automobiles. And no automotive manufacturer would EVER risk actually having this entirely specious legal theory dismissed by a judge in some night traffic court in Arkansas. The thing was ONLY useful as a THREAT. The last thing they wanted was any sort of ruling on the theory.
But it had an absolutely Satanic chilling effect on the whole concept of automotive software and right to repair. If it was anyone at a University or a corporation or really any entity with anything to lose by the expense of long drawn out legal expenses, a simple Cease and Desist letter CITING the DMCA was too often thoroughly effective at ENTIRELY shutting down any reverse engineering activity. I’ve even had EVTV hackteam members ask to kind of be held in the background as their positions could be awkward with employers and customers etc. if there were any legal implications.
Having adopted FOAD on the ROAD as rallying cry myself, it’s rather good to be me. But not everyone has positioned themselves to be able to effectively adopt this policy in blanket fashion.
So what is the effect of the year delay? Not only none, but comically none. Who’s going to issue a cease and desist letter threatening legal action under a law already scheduled to be exempted long before any legal action could actually be scheduled? Any lawyer would face sanctions for even filing such a thing. Assuming they WOULD actually seek a ruling, which they never were going to risk? It’s a brilliantly neat little copyright office joke telling the EPA to FOAD, while technically appearing to cater to their every whim.
I did finally get a video out and of course it is over 2 hours in length. Of course we recently had an “open house” version of EVCCON where I attempted to share with the 55 or so people who actually attended what I thought was going on, where it was going, and what we were doing about it. Byron Izenbaard has posted a HERO video of the opening session if you care.
But we’ve been kind of busy behind the scenes.
For one thing, our component sales, which provides the oxygen around here in the absence of the normal advertising support publications generally receive, have undergone a bizarre change. As we have begun to explore OEM style components, we’ve flushed out an entire universe of pretty serious people who are actually trying to produce electric vehicles of various types and at various levels. Detroit Edison, for example, went into production this week on their Lotus Elise-based electric sports car – the SP:01
But there are many others. While the hobbyist/DYI guys are all spending their ducats buying 3-year-old Nissan Leaf’s coming off lease, there are hundreds of these small operations springing up, particularly in EUROPE. They do tend to be small, but pretty serious about it and they like our current offerings.
Unfortunately, they want custom “quotes” and special accommodations and worse, international shipping AND they want a quote on the shipping. After a quote they send a Purchase Order, and then we are supposed to send them an INVOICE. And then they pay generally by wire transfer which we have to manually confirm with the bank. Then the actual International shipping begins. With minimal staff, this couldn’t have hit at a worse time. And I suppose it is the kind of problem to have, these are all $5000 or $10000 or $20000 orders. But they are still a problem. We’re kind of set up where you order yourself by clicking on the web site, pay by credit card and then we strive heroically to just get the right stuff in the right box. Three Siemens, two eGeardrives, and a UQM Powerphase 100 sent to Turkey is not our usual activity.
And of course, with the attention span of a four-year-old, my real interest is of course using the Arduino Due to tame all of OEM components worldwide, and of course starting at the top, with the Tesla Drive Train. With the contributions of a number of bright individuals, a lot of just plain old fashion luck, and a few hundred hours of just staring at the screen and trying things, and we have 99% of the Tesla drive unit figured out. No small part of this is Tesla’s continued support via their http://service.teslmotors.com which I must say is actually quite excellent as far as it goes.
But that doesn’t quite do anyone any good. For one thing, Tesla has eschewed the concept of a model year. Rather, they make more or less continuous improvements to the car. A simple example of this is that they have a Generation 1 and a Generation 2 version of their onboard chargers and distribution box and they are really VERY different. But it extends to the drive train as well. They already had a P version and a non-P version and then it appears their left hand drive is different from their right-hand-drive and now of course they have the all-wheel-drive D model. And so it goes.
Meanwhile, we have dozens of guys out there who have spent multiple thousands of ducats buying up the salvage drive trains and we hear from them. Basically it is a pretty demanding insistence that we give them a firm date when we’ll have free software available so they can make theirs work too. I don’t know quite how to react to that. It generally is delivered more as an accusation than a request and so disposes me to allow self-fulfiling prophecy to take its natural course. But the technicals of the issue kind of prevent me from having to deal with it at all. We’re not going to put out a control system that MIGHT work on your Tesla drive train, never mind that we are some distance from having anything remotely resembling a product. Right now I’m running some software on an Arduino Due with a CAN port, and providing control inputs as ASCII characters on a laptop. That’s not something you can drive to the convenience store for Ho Ho’s and Ding Dongs.
So our confidence level on it working is kind of based on buying drive units, hooking them up, and seeing what they do when we send the commands. Over time, doing a lot of them, and patching the code to accommodate the variants, I expect to build toward a universal controller that will detect and run any variation commonly available. But that’s kind of a blue sky take on a perfect world. It might be easy. And it might involve some further exploration.
I kind of hope to fund further exploration by selling them onesy twosey at a time as we get them working and get a controller device that could actually work in a car. But that kind of involves a custom harness, the controller, a display, the drive unit, all on a single pallet ready to ship to Belize. And of course tested live before shipment.
So we’re working on things like wheeled wooden dolleys to crane them onto so we can wheel them around, hook em up, test them, etc. As you see in this episode, we have taken delivery on a very expensive variant – the entire ass end of a Tesla model S. 74 inches wheel rim to wheel rim.
Well that already has us wrapped around an axle – so to speak.
For one thing, it already has a full wiring harness on it. So all our work to chase down connectors doesn’t really apply to this one. It already has a great factory connector for the encoder and for the inverter control signals. So we need to walk back up the harness to the end connectors we DO have and figure out the part number for the mating connectors, procure them, and wire them appropriately just to talk to it at all.
And there are some further targets of opportunity. The rear wheels feature TWO sets of brake calipers. One of them is electrically actuated and is what keeps the car from rolling off into the river when you put it in park or turn off the car. Nice feature actually. These calipers are connected to an Electric Parking Brake Electronic Control Unit or EPB ECU officially. This is connected to a different CAN bus in the car, the CHASSIS CAN bus, and so we need to do some captures and reverse engineer the CAN codes to control the parking brakes. But of course, we didn’t actually get an EPB ECU with our drive train. More connectors. And of course we need to actually get the control box.
In this episode, we do examine this entire assembly. I also talk a bit about our control progress using a PowerkeyPro 2400 keypad to enter PRND type commands to my little control box. After 105 days and 93 e-mail messages, starting July 17 when we were all in violent agreement to purchase 48 of these and they expressed how very deeply delighted they were that we were, I’ve given up. The device is great and works fine and is very reasonably priced. But the company, like so many American firms today, is just totally dysfunctional. Days and weeks go by between e-mail messages. Their products don’t actually match their spec sheet. The spec sheet on the web site is the WRONG spec sheet (somehow my fault?) and on and on and on and on in an absolute litany of incompetence and apathy culminating in the amazing announcement at the end that they’ve sold the entire product line to ANOTHER COMPANY ANYWAY.
We really prefer to order from the Chinese in CHinglish. American firms simply do not have personnel on staff that can do this work, and I’m starting to believe this is true of EVTV itself. I’m now up to four full time and one part time guys trying to get this shipping to work and they are just not ABLE to get the right thing in the box one time in a row. I’m looking to hire some Chinese students graduating Southeast University here soon. It’s my only hope.
Meanwhile, I also showed the Andomeda Interfaces Electric Vehicle Integrated Controller or EVIC displaying data such as temperatures and current and voltage and so forth. This is actually a touchscreen display but in our version done for GEVCU, it doesn’t actually have any touch inputs. But I would like to have something reminiscent of the Tesla instrument cluster.
Our first wag at this looks something like the below:
I shipped my new Windoze Laptop with Visual Studio 8 and the Andromeda SDK to Collin Kidder in Sparta. In less than a week he had this up and going though kind of staggering through the process. He then shipped the laptop back and now we’re on the same page both able to compile this software. I’m pleased to report that late Sunday morning, November 1, I was able to actually control the Tesla Drive Unit using the buttons on the display. Naturally, some further work is recommended, but it is promising. The performance of Windows CE is problematical. But we’re looking at it.
Note that the Alset Drive allows you to increase/decrease regenerative braking in 5% steps. You can also turn CREEP off and on. It has both an Odometer and a Trip meter and keeps Watthours per Mile on the Trip. And of course PRND. The right side of the dial is power in KW and the left side is miles per hour of course. State of Charge is shown with a progress bar in the lower center of the dial. Hopefully video of all this in the coming week.
I did decry the state of the onion vis a vis electric car sales noting the apparent missing-in-action status on the 2016 Nissan Leaf and 2016 Chevrolet Volts. Technically, I suppose they are out there, but effectively the numbers are so miniscule as to not exist. And I compared sales of Leaf and Volt with sales of Tesla for the first 9 months of 2014 compared to 2015.
This is almost too pat. The real world is naturally a bit more complicated. The Volt and the Leaf were I suppose early and often as plug-in vehicles available for sale go. There are new e-cars introduced almost daily. The guys with the best numbers on this stuff are at http://insideevs.com. We scoured some of their back issues to glean basic plug-in car numbers. These are not strictly battery electric vehicles but includes a number of hybrids, but all able to charge by plugging in and of driving at least part of the distance with the gasoline motor off.
This first graph shows the January-September Sales values or 2011 to 2015. We can readily see that things improved every year through September 2014. But the first nine months of 2015 they WERE down rather markedly. As this is cummulative among all plug-in vehicles, there is no doubt that my prediction of things going bad this year have woefully proven all too true.
But you have to be very careful with numbers. Like reading messages from God in cloud formation, it’s all in the interpretation.
In this second graph we look at sales by month for the entire period from January 2011 through September 2015. We see that the bulk of the hit in 2015 was really in the early part of the year and it actually looks to recover here in the latter days of autumn.
I also graphed all U.S. passenger vehicles against this by month. And I was struck by the similarities in the graph. They are on two different scales, with U.S. vehicle sales forming a bit over 100:1. But the ups and downs of EV sales are not entirely different from general vehicle sales.
And really, when you back away for a birds eye view, we are taking a hit and I do think it is related to depreciation and reliability and questions the public has over a new automotive technology. But it’s not all bad. The apparent losses of sales for Nissan and Volt are a bit simplistic.
For example, not shown is BMW and the i3. They sold 3104 i3’s January-September 2014 but 7,893 during the same period in 2015. That looks like/sounds like/smells like 154% year-over-year growth rather dwarfing Tesla’s tiny 52%. I can readily envision some part of those sales coming from otherwise would be Leaf and Volt owners.
And just to bring this full circle, there is the VW e-Golf. In 2014 they sold ZERO eGolfs. But during the first nine months of 2015, they sold 2555 e-Golfs – which by definition not a single one were either a Leaf or a Volt. Calculating the percentage growth here is a little challenging. But let’s have a go at it. You have heard me refer to the number Brazillion a number of times and some have even inquired how many were in a Brazillion. Let’s just say that VW has enjoyed a Brazillion percent growth in the past year of their e-Golf vehicles while sales of their diesels may have diminished somewhat. And we can define for the future that any number divided by zero produces the numeric 1 Brazillion. I’ll leave it to you to work out then what 2 Brazillion represents.
But I would say that any company with a Brazillion percent growth in one market, while getting killed in another while under threat of criminal prosecution of their executive staff, might consider focusing future efforts on the more promising growth market.
I often feel like the wet blanket of EVdumb with my mantra regarding the technology adoption curve. The problem is I’ve been playing with these adoption curves for about 35 years now and haven’t quite caught any of them in the act of failing in the hypothesis. Generally, the early adoption segment of the curve begins in earnest at about 2.5% penetration of market share. With 13.5 million U.S. vehicle sales this year, that would be about 337,500 plug-ins or EV’s depending on how you define it and what you’re tracking.
Let’s say we confine our “market” to light passenger cars at 7.5 million per year. That still needs 187,500 cars. We’re at 90,000.
I started a small newsletter about bulletin boards and online services in 1985 and sold a publishing and trade show operation 13 years later in 1998. Along the way, one of the issues was that the proponents of the Internet wanted it so badly, they kept declaring victory over everything at any positive moment. And they also had a propensity to write it off as “who killed the Internet” at every negative development. In the tinkerer and innovator stage, success is NOT assured, but even if it occurs, it is marked at that point by fits and starts, wins and losses. New companies and bankruptcies. A bubbling cauldron of entreprenurial activity. Dreams lived and hopes dashed on the cruel rocks of economics and timing and the winds of war. It is the violence of storms in the petri dish.
When I combine this with the sudden influx of dozens of new startups SERIOUSLY wanting to form real companies and build real vehicles in varying quantities around the world, it dawns on me that we might finally be at the BEGINNING of the electric car phenomenon. The fork lift motors in the 30 year old Datsun in the garage giving way to a new breed of serious entrepreneur who while duly impressed with Elon Musk, does not consider the Tesla Model S as the final say so in the history of the automobile. So I see not so much doom and gloom as a hint of dawn of a new age.