This week’s show is finally a bit shorter as we get caught up.
Speedster Nippon went on a truck. This is one of the best builds we’ve ever done, and quite a pain in the ass to perform. It points up why we DON’T want to be automobile manufacturers, even on a small scale.
The issue was of course government and paper in nature. The state of Missouri, onerous in many ways, is a bit generous when it comes to automobiles. Neighborhood electric vehicles and scooters of 50cc or less require ZERO licensing or regulation within the state. You can just buy them and drive them on the street. Similarly, we have an EXCELLENT law on the books regarding historical vehicles. You buy a license plate and register the vehicle ONCE, for $25, and the registration is good FOR LIFE. This is probably a nod to some historic car collections that have been traditional in Missouri.
Registering the Beck Speedster replica is very interesting. The state actually registers it as a 1957 PORS even though it might have been manufactured 3 weeks ago in Bremen Indianna. They do actually moark the title REPRO several inches down the page, but they allow registration under the historic vehicles law.
Speedster Duh was registered in this way and delivered to DELTA FORCE last July. They purchased a follow-on build for November 1. We missed it rather badly.
Actually we made it – just. But along the way, they rather insisted that we were to register the second vehicle as well. There was actually a great deal of discussion regarding delivery dates at the meeting here in Cape and it was specifically decided NOT to register it here as it would add to the delivery date. I forwarded some e-mails reiterating this and they admitted it wasn’t in the original agreement. But they wanted it registered here anyway, at additional expense and delay. Why?
[jwplayer streamer=”rtmp://s3einxnpkaij93.cloudfront.net/cfx/st/” provider=”rtmp” file=”news060713-iPhone.m4v” hd.file=”news060713-1280.mov” image=”http://media3.ev-tv.me/news060713.jpg” width=”950″ height=”584″ html5_file=”http://media3.ev-tv.me/news060713-1280.mov”]
I can only surmise that it has something to do with registration in Japan. And similarly Japan, home to many car collectors, grants certain advantages to owners of classic vehicles. I can only assume that Delta Force found SERIOUS advantage to having the Missouri Title state that it was a 1957 Porsche.
And so it goes. Tesla at one point was fined $244,000 by the EPA for not filing a vehicle emissions certificate – on an electric car. There was no contention that it HAD any emissions, only that they had failed to file for the certificate. Regulatory madness.
And that brings us to a key element of EV-dumb. You do NOT really want to be a vehicle manufacturer.
Most car buyers are almost totally unaware of the business of automotive manufacture. Almost everything about it would shock you and indeed most of it would shock people IN the automotive industry. Let’s start with the basics: There business is actually NOT designing and building automobiles.
Actually about 80% of the engineering talent at any American automobile manufacturer do not design cars at all. They design a machine, approximately a mile and a half long, that makes cars.
This machine consists of people, robots, sprayers, and a vast logistics system to deliver parts from all over the world to precisely the right gate at precisely the right time and in precisely the needed quantity. It costs billions of dollars to construct and many hundreds of millions of dollars each year to modify. Stuff goes in one end, and cars come out the other.
The very tiny handful of employees who actually design and test automobiles are under constant edict to “design for manufacture” which means they don’t care how the car runs, they don’t care what is required to maintain the car, they are MOSTLY concerned with what has to be done to MANUFACTURE the car.
The PARTS that goe into this machine, are a tiny part of the cost of your car. GM can buy and produce ALL the parts for a $60,000 Cadillac Escalade for about $8,000. You of course, could also purchase all those parts from the local dealer and build your own Escalade, but the parts would total about $300,000. So you see the markup on vehicle PARTS is quite interesting.
The labor to build the car? Lost in the rounding errors. When GM was facing bankruptcy, they disingenuously pointed the finger at the Union workers who assemble the cars. The head of the Union was widely criticized for claiming that they could work for free with no benefits for the next 20 years and not have any effect on the bankruptcy at all. The media howled. The GM executives scoffed. And we all went along with it. But the Union leader was exactly right. The labor costs to build the Escalade get lost in the rounding errors. The assembly workers are a tiny part of the “payroll” which is mostly high dollar salaried engineers in the invory tower.
We see many estimates discussed as to how much it COSTS to manufacture the Nissan Leaf or the Chevy Volt. The automotive executives seem vaguely secretive and evasive about this don’t they? Actually, they are not trying to be either secretive OR evasive. They don’t precisely understand the question or what you are talking about. And while they know quite a bit about it, they don’t know the answer to the question as stated. They exist to answer very different questions.
The cars don’t HAVE a profit margin. If they did, you would be shocked to hear it. If you take the cost of the parts and labor to build the car, and compare it to the price the manufacturer receives, the “margin” is probably on the order of 600 or 700% – improbably large. But it’s kind of nonsense. The cost of the MACHINE dwarfs EVERYTHING. And so the way they think of profitability of a model is the break-even. If we introduce this car, it turns profitable with the 63,343rd vehicle produced and sold. In their world, if they sell 63,342, the vehicle model didn’t make any money at all. And that if they sell 64343, for each car after that, it is literally raining money. What each vehicle margin is or how much they made on it is almost nonsense. It’s burdened by the breakeven – the cost to build the MACHINE and the overhead just of having a corporation. Net/net/net, it is all about how MANY they sell. There is no specific margin on any particular vehicle except in historical review. At any point in time, you can calculate BACK but as sales occur each month, that almost doesn’t make sense.
Complicating THAT is that they are going to sell the SAME vehicle next year, and additional costs are a function of what changes have to be engineered into the car, and thence more importantly into the MACHINE. How MANY years they can sell the same model, and the degree that it changes year to year, are huge impactors to profitability.
And that goes to predictability. They have to forecast very accurately, for each model introduced, how it will be received, how many it will sell per year, and how long a life that model will enjoy. All rather impossible figures to come up with.
Now, if you have a financial CREDIT LOCKUP stop ALL financed vehicle sales entirely, as we had in 2008, can you IMAGINE what this does? That’s what really happened to GM, Ford, Chrysler et al in 2008. They were not only bankrupt, they actually had NO IDEA if they were bankrupt or HOW bankrupt they were. They were kind of suspended mid-air, with BILLIONS ticking off on the clock at a DOZEN of these machines in a surreal time warp where NO cars were sold but the machines kept eating like monsters.
This is why Bob Lutz proclaims the automotive business as one of the most complicated in the world, and curiously difficult for new entries, such as Tesla, to manage. Never mind titles and emissions certificates.
From my point of view, you don’t need to know any of that. It is unattractive on the face of it:
1. High capital costs.
2. High employee count/labor costs.
3. Extremely high government regulation.
4. Ultimately low margins.
It is a thoroughly mature thoroughly unattractive business. You could literally be much more profitable devising the ultimate bobble head doll to put in the rear window of the car, than you could manufacturing, at any size or any scale, the car.
Tesla has a couple of opportunities with NUMI and the Tesla electric car. It’s true it is a disruptive car. But it is more important that it be a disruptive MACHINE with a disruptive technology to DESIGN the machine. That means very high tech very silicon valley techniques to computer design cars and robots and assembly procedures. Annoyingly for Musk, the actual production of the car got down to humanoids working overtime into the night in December to put cars together and fix problems. It’s not really going so well there. The supplier logistics do NOT lend themselves to new age techniques. In fact, GM kind of has THAT part down to an art form – beating up vendors. If some 80 IQ truck driver shows up at the wrong gate, out of the 26 available, 10 minutes late on ONE delivery, the vendor finds their profit for the MONTH gone entirely. Read the fine print. It’s clearly spelled out in 4 point type in the purchase agreement. And the vendors are NOT unaware of it. They know ALL ABOUT IT.
Once you are aware of this, you understand how thoroughly annoyed I am with absolute rank amateur CLOWNS when they inform me in unctiously superior, nay CONDESCENDING tones that “they only sell to OEMs.” They are TOTALLY CLUELESS what this sounds like to me. Or what an actual GM supplier might think of them.
A good friend of mine recently actually shut down a company he had taken over 20 years before – after his father’s untimely death. He layed off ALL his omployees, many of which had worked there since before he was born – beyond friends – essentially family he had known all his life. People who had NO PROSPECTS of ever having employment anywhere again for the rest of their lives. Sold off all the equipment for 10 cents on the dollar. The business was beyond “not fun anymore.” It was a brutish existence living as a menial serf on the GM and Chrysler landhold, eeking out an existence one day at a time. The spectre of what might have happened in the narrow escape from GM’s bankrupty clinched the deal. He managed a tiny pittance by selling his contracts to competitors. Maybe enough to live on for a year or two and start something else on a shoestring. Thanks enough for 20 years of 100 hour work weeks.
On a cheerier note we have NADA. The North American Dealers Association. You know them as the ill fitting plaid sport coat crowd with the patter selling you the cars and going to see if management will approve the negotiated price. You rest easy in your career knowing that if all else fails you can always get a job in a car lot selling Buicks.
Here too, the business has almost nothing to do with selling cars. They have to sell cars to keep the franchise. And often they make $300 on a sale. Chump change. It barely keeps the plaid suits employed. But it does give them kind of an ironbound lock on YOU as purchaser of the car.
Out back, behind the acres of shiny new cars, are eight “service bays”. Depending on vehicle line and geographic location, EACH bay can generate as much as $2.5 MILLION in annual PROFIT, not revenue. Once they sell you a car, they own you. THe automaker pays them for what’s covered under warranty. And you pay them for what is not. They get it from both directions.
Ergo Elon’s tearful assertion that it is wrong to profit on service in the automobile business. While charging $600 per year essentially for software upgrades himself.
All of that goes to a prediction I made 20 years ago that the Internet would absolutely and INEVITABLY lead to the collapse of distribution networks. It has taken a long time, and it will take decades more. I am astounded at how resistant and persistent those distribution networks have been. But in many products, alcohol, wine, vehicles, they are cast into stone with protective legislation usually at the state level. That said, they are an act of gravity defiance, and ultimately all will tumble to the ground. The advantage of selling direction, both to the manufacturer and to the end user, are just insurmountably too huge to discount.
Ironically the BEST example of the collapse of distribution networks IS itself a distribution network – WALMART. It will first eliminate all other distribution networks, and finally eliminate itself. In demonstrating how much more efficient it can be than the existing networks, it will ultimately demonstrate the absolute efficiency of no distribution network at all. In the interim, in a very strange way, they give us “buying power.” The ability to drive manufactures to bankruptcy on price.
Sounds pretty gloomy Jack.
Actually not. Mature industries drive toward zero profit and zero opportunity for new entrants and employees alike. But disruptive technology is, ….. well… disruptive. When an airplane is invented, it gives rise to a LOT of needs, airplane tires, and airplane lights, and wire, and fabric, and runways, compasses and airports and airport beacons and the BULBS in the beacons that were never needed before at all by anybody. It’s a great invention, but it needs ALL the infrastruture to support it. These needs initially run both wide and deep in all directions. After about 110 years, it becomes MATURE with all the little holes of opportunity filled in, and wiht a century for our brilliant leaders to figure out how to milk it to the uttermost farthing with taxes and regulation.
When Henry Ford first massed produced the automobile, there was a sudden need for EVERYTHING related to automobiles. Imagine how much was made and how much opportunity in such esoterics as ROADS, and gasoline stations and spark plugs and spark plug wires. Bill Lear, inventor of the Lear Jet, made his first fortune by developing an AM radio that could be installed in a car, allowing radio reception in spite of tne noise from the ignition system. He called this new venture MOTOROLA and it had NO relationship to cell phones at all.
Today, we have two very strong areas of disruptive technology. Let me be clear. There is no if. There is not really a when. It is inevitable as sunlight and rain. Electric cars will subsume ALL of the automotive industry, in its entirety, and world wide in the coming few years. Not hybrids. Not market share. It will EAT IT ALL.
And similarly solar power is not going to be an adjunct to the production of electricity, it will subsume it ALL in coming years. Scoff as you might, about 1350 watts of power fall on each square meter of the earth. Photovoltaics allow us to harvest it with no moving parts and almost no maintenance – anywhere the sun doth shine. After that, it’s all about cost and efficiency. How much of that 1350 watts can we harvest and at what cost. Intuitively, comparing it to a monstrous abomination that eats coal, spews sulphuric acid, and costs hundreds of millions to build, what’s the question?
These are disruptive innovations featuring ENORMOUS opportunity. Probably NOT in the actual manufacture of the cars and probably NOT in the actual manufacture of the solar cells.
In this weeks’ show, I used two graphs to illustrate recent history in the Solar Industry.
This graph shows you the solar module pricing and the INSTALLED solar costs from 2000 through 2012. You will see that the modules have fallen in price from about $5 per WATT in 2000 to about $0.74 cents per watt now. I actually priced some Korean panels this morning in small quantities at $0.63 cents. The INSTALLED cost, which includes the labor, the panel racks, the wiring and the inverters has similarly fallen from $12 per watt to $4 per watt. Both of these are EXPONENTIAL curves of declining price. There is actually a hypothesis, similar to Moore’s law, that has held remarkably accurate for 30 years, that with each DOUBLING of cell manufacturing capacity, we will see a 20% decrease in costs.
Existing panels are about 16% efficient, meaning we get 0.16 X 1350 or 216 watts per meter peak. In the lab, we are already over 40% efficient. That’s 540 watts per square meter. How efficient can these black monoliths whose job it is to lay there be? Well historically we can pretty much bet not 100%. Beyond that, it’s all about innovation.
This graph shows installations, in Megawatts, of Solar Power in the U.S. by year. Note first that it is NOT cummulative. Each of those numbers is the combined utility, commercial, and residential installed capacity for THAT YEAR.
This curve is also exponential in growth. And out of context, when you look at this graph, it would appear that as an “adoption curve” we have missed the boat and widescale adoption is already underway. Don’t you wish we had gotten on board earlier in the game?
Actually, earlier there was no game. And a case could be made that there is no game yet. Small numbers look ENORMOUS compared to zero. And be careful with these charts. It’s all a matter of scale.
I didn’t include it in the video because it is too much of a horror for young ears. What do you imagine that 4200 MW of installations in the U.S. is as a percentage of total WORLD installations this year?
About 11%. We aren’t even one of the significant players in that space as a nation.
And let’s further examine, what percentage of all electricity produced in the U.S. would you imagine Solor represents?
Again the number 11. But this time 0.11% according to the U.S. Energy Information Administration. Our grossly vertical curve, in the big picture of electricity production, is so flat you can’t find it without test instruments. We don’t yet have both tinkerers AND innovators onboard yet, and we are HUGELY before early adopters. But as you can clearly see, on its own terms it is taking off exponentially. Can you predict the future here? And how did you do that? And how hard was it?
And how many of us use electricity and how deeply do we depend on and want it? And what if I told you that you could put a few panels on your roof, measuring about 12 feet by 12 feet, and power your entire house with it? And it cost $1500? Of course it would require inverters, and chargers and BATTERIES and any end of other things to operate.
Same question with regards to automotive personal transportation. And what effect would it have on our established utility industry would it have if significant numbers of 100 million households actually installed Solar? Or drove electric cars?
And what do we need? Infrastructure. Not just charging stations, …..EVERYTHING. These are totally disruptive technologies and we don’t have ANY of what we need yet. First, we don’t know what we need, and swecond, nobody has invented it yet.
The American adulation of invention and innovation ONLY works in arrears. This is why an IBM executive would predict that with six large computers they had all the computing power mankind would ever need. Why a patent official would announce ti was time to close the office as everything that needed to be invented HAD been, circa 1904. Along with many other idiotic predictions such as “If God had meant man to fly, he would have given him more money….”
Uh-Oh. Actually that one is probably true.
But you get the idea.
We can only examine innovation and invention with clarity in arrears. At that point, it was obvious. Looking to the future, it is never obvious at all.
But every generation has their frontier of opportunity. I would represent that electric drive and solar power, far from being passe failed techologies of the past, are actually the most globally important of the future, and we’ve hardly started yet. As these are BIG things they have had long development and adoption curves.
But you’ll hardly make your fortune in Solar Panels or manufacturing electric cars. Anymore than George Westinghouse manufactured railroads. He invented the airbrake for the railroads. Nobody knew we NEEDED an airbrake until we HAD railroads and had already killed several thousand passengers, and brakemen, with them.
Bill Lear did not invent the car, he invented the car RADIO. Later the 8-track tape player. And of course the Learjet. A personal sized jet. Which at the time made about as much sense as building your own personal sized aircraft carrier.
Today, we have pontoon boats. And they are for barbecuing, but not precisely for landing aircraft. They mostly haul beer and potbellied post middle aged guys in Hawaiian shirts. Kind of a red neck yacht club. Otherwise ALMOST identical to a personal-sized aircraft carrier. Better if electric.
At precisely the time that a signficantly large portion of our population is becoming absolutely UNEMPLOYABLE in any capacity, we face an absolute pinnacle of entreprenurial opportunity the like of which the world has NEVER SEEN, but only for entrepreneurs and only for those who can innovate. Those standing in awe of governments and large corporations inevitably fall into the first group – permanently redundant unemployable.
Elon Musk stands at the cusp of facing an $85 billion entrenched automotive industry, an even larger entrenched oil industry, and of course the legislators of all fifty states. How does he think he can win?
Simple. He has them surrounded…. And oddly you do too…
I see EVTV’s role in the future as Summit Racing to the EV builders. And I think you will very soon find irritation at all the things you CAN’T do with your store bought electric car because it still belongs to the manufacturer. Tesla has a patent on the CHARGE port for Christ’s sake. The batteries. The software. Etc. You will drive, care for, repair, and own that vehicle in exactly the manner Elon Musk decrees, or do without – even if you’ve already paid for it. The cars inherently become MORE proprietary and less accessible to the owner in some very strange ways.
As such, the custom car movement in electric vehicles will be MUCH stronger and more advanced than the Hot Rod community of the ICE engine sort ever was. I predict it will dwarf SEMA by a LOT. Conversion shops. Pimp My Rides. Race Teams. Car shows. It will be huge. WIHT LOTS of opportunities for jewelry, gadgets of all sorts, add-ons and improvements. A whole cottage industry in digging out scrap from the wrecked factory cars fo use in the shop.
Both fronts, solar and electric drive, offer opportunities so vast they are incomprehensible. Transportation and energy will be the hot topics for the remainder of my life.
On a housekeeping not, in a panic over the infuriating collapse of CA60FI cell stocks, I might have just overordered by a few. I have a thousand cells reserved due to land in LongBeach June 15. I should have them sometime between June 20 and June 30. Annoying as it is, they are probably going to want me to pay for them. I can. But it throws all our numbers out of their neat rows. All because of my fit of anger and shooting off my mouth.
You get to bring it back into nice neat little rows by purchasing cells at a 17% discount prior to noon Friday. NO dealers. Limit 100 cells. Not likely to happen again soon. You’ll save a slick $1500 which I could have put to good use, on a 100 cell pack. Party like shop dogs. http://store.evtv.me/proddetail.php?prod=CA60FI. Don’t forget I have a minority interest in your freakin car then, please. You can buy me back out with photos in our new database if I ever get it coded up. We’re going to publish the 100 BEST EV BUILDS WORLDWIDE, hopefully by the end of the year.
50 thoughts on “Where the Sun Don’t Shine…”
Hey…. Watch the a Redneck jokes… You will wound our inner child…
I liked the show this week and the write up….
Really would love to do a solar powered pontoon boat…. Just imaging the reception of explaining that the solar panels actually do in fact keep the beer cold….
P.S. You will know that the EV’s have truly arrived when you hear two bubbas in Overalls saying, between tobacco spits, “If Ida just had 50 more amps I woulda won that race….”
@ Anne – intrigued by the early quitters among your cells. I have far fewer cells in my CALB test pack and those were pretty close to being in balance out of the crate. I have bottom balanced them but they weren’t far out. Do you suppose your outliers were undercharged at the factory or maybe have some kind of internal leakage which caused them to self discharge?
What do you think Jack?
I think he will find they are full capacity perfect cells that were somehow at a different SOC than the others in the cartons. It looks to me that this should be automated and they should all come out at exactly the same SOC, but they don’t.
In fact, each cell has its own personality and we start naming them if we get to fooling with them too much. Hey there “supersport” get “recalcitrant bitch” inline there with Rudolph, dancer, prancer and Deadhead.
Most likely they do the initial charge to do the magic that needs to be done and then discharge to a set voltage at a fixed current to find the capacity so they can document it. There are a couple of ways they can then put in the storage charge. A fixed rate for a fixed time or at a fixed rate until the cell reaches a particular voltage. Either one will put different amounts of energy into the cell because of errors in instrumentation. If you do resting voltage SOC measurements you find that near the middle of the SOC where they ship the cells a voltage variation of 0.0004 is good for about 1% difference in SOC. The internal reference on the common microprocessors is only good to about 1% and a 10 bit A/D can only resolve 0.004 volts when given an input range of 0-4.095 volts. And worse still the resolution error is + or – one bit making them really only 9 bit devices (0.008 volts). So unless all the cells were tested on the exact same device there will be some cell to cell variation in state of charge. And temperature makes a difference as well. So if you assume that for some reason they actually consider it important to set the shipping SOC to be identical, about the best they could do if they equipped their charging gear with 0.1% voltage references (plus or minus 0.003) and on top of that the error in a 12 bit A/D converter of 0.001 volts fives an uncertainty of around 0.004 volts which is about an order of magnitude worse than would be needed to get better than a 1% variation in the state of charge. And I strongly doubt that they care that the shipping SOC of every cell is identical. I would expect 5% variation in shipped SOC if they charge at a fixed rate using a cutoff voltage to stop. SInce it seems to be a little better than that they are probably using a constant current source and charging for a certain amount of time. The variation then becomes how accurate is the current source and how accurate is the timer. My GBS cells had a low shipping voltage of 3.3037 volts to a high of 3.3089 volts. This would be about 11% variation in initial SOC.
Thanks Doug that’s helpful
Oh my !
Its so much fun to read your Blog.
Where did the upload link go ?
I was thinking about uploading some electric TT-Ford Bus fotage.
Did you get your cells bottom balance ? I am gonna try to see what happens when i throw in 4 more cells, in total 46 Cells in the Renault. My goal is 48 Cells over time 🙂
Hi Allan !
It will be hard for you to get 46 cells working in the Clio. The controller in the CLio will not start if voltage is over 147-148v and if you divide that on 46 you get 3,217v /cell which will not be near full charge on your thunderskys. I used 43 cells for a while but I pretty much lost regen because controller also reduce regen current when pack voltage goes over 148v.
There could be a way around it but you would need to do a little modification of the controller.
What range do you get with 42 cells?
It makes sense. when the car has been for charging and would like to take a trip, it take a little time before i can get the car moving. Well I thought that it was the The fans that hat to clear the air around the battery box, from the time with NI-Ca battery.
Never looked after what was wrong
Long trip the car run´s 200 Km and i have about 10 Ah left, but you need to be careful with your Ah´s (economical driving), and it’s tires with low rolling resistance.
In the town it is about 150 – 170 Km.
The model 1997 -1998 is switching the main pack on when the auxiliary battery voltage drops, so if your battery is not charged in 14 day it will take a part of your power, maybe 10 Ah,I have not measured it yet, and i have the EMW instrument which is using power from the auxiliary battery all the time.
Do you know which wire they measure 147-148 volts?
PS: my battery pack has been discharged to 79.5 Volt with out any loss of Ah. However in the winther time, I am misisng about 10 – 15 Ah and when the temperature is about 15 deg. They are coming back again. I have been running 7000 KM
Hi Allan !
Register on the swedish forum
or email me directly for more info on how to fool the controller.
I have a motorcycle on craigslist going for enough for 50 cells………..
A long time lurker, first time poster! Love the show, love the blog, love the movement.
Thanks ever so much for going into Dante’s Ninth Circle of Detail on the HPEVS offering over the past 3 shows — it has helped me immensely to make sense of the best power plant for my recently acquired glider/roller, a 1989 Porsche Carrera 911 (in the same GT Silver as Brian’s Ghia no less). I was also waiting ever so patiently over the past year for the new 144V Curtis (and for the right roller!), and had pretty much thought that it and the AC75/76 was what I am planning to run with… until I watched the show(s). I am now going with the new 35×2, and have been talking to Bill R about how it will fit, etc., as they have an ’83 SC 911 in the shop. They are planning to upgrade it from the AC75 in there now, as soon as the Vette is done and dusted, probably long after the last EVCONN folding chair has been stowed away for next year.
One thing he did confirm, though, is that you can run 48 cells max, yielding a 160V pack. You had mentioned 46 on the show, so I’m not sure if we are getting different info, but I’ll take the added cells if possible.
An interesting calculation comes out of that, as it related to the 3.2L in my roller:
1) 160V x 1000A = 160kW = 214HP.
2) Porsche 3.2L (1989, stock, US) = 214 HP (@5900 RPM; http://911evolution.com/carrera32/carrera32spc.htm ).
I know it’s apples to oranges, but kinda interesting how that adds up.
At any rate, I anxiously await the arrival of your 35×2, and can’t wait to see it go into Speedster Lite.
BTW: On one of your “creature comforts” directives (#3: “tunes and iPhone”) given to Brian on the Ghia, check out:
Blaupunkt Toronto 410 BT
I am considering the Blaupunkt, as that’s about as OEM as you can get when it comes to VW and Porsche, but the Sony has better features and reviews, IMO. Cutting out a big messy double DIN hole in the dash or center console isn’t in the cards for me.
Hope to meet you and Brian at EVCONN.
For a Ghia, gotta love Blaupunkt. It just fits.
We received our first 35×2 yesterday. We’ll be talking about it this week. We’ll also be adding it to the store most likely sometime tomorrow so order early and often.
Bill has been all over the place on the voltage. The real answer is 170v. But here’s what apparently is happening. Full battery if you come out of the garage with regenerative braking, it apparently spikes the voltage and the controllers fault. At 46 cells, this just can’t happen. At anything over, it CAN. Your mileage may vary. I expect we’ll be provisioning battery box space for 50 cells but I’m prepared to learn the hard way that 48 or 46 is the max.
Oops… I mean the Blaupunkt Toronto “420” BT:
I hear you on the “fit”, and I also decided that a Blaupunkt would be more at home in the Porsche, so I ordered one last night on Amazon. They’re a buck-fifty right now. A nice weekend project while I am still doing my “road research” before pulling the 3.2L.
Thanks for confirming that the jury is somewhat out on the max cell count for the new Curtis. I think I will plan to wire up 46, but leave room for 48 (or 50).
good show and write up. Two things (not car) came to mind.
1 did you read about the EV Soul from KIA and what are your thoughts on it making it (to market and lasting)
2 on the database mag front is it just for cars or is Anne going to have some boats in there?
I’m really enjoying your blogs at the moment. They are fun to read.
Any chance of a bit more of an explanation about what happened with the development of the Pulsar charger? Was it a bit more complicated/expensive to make than first thought? Will it turn up, but later, or is it dead?
Despite having funded it 100% up front, and after we had hammered out and agreed to about a 10 page specification on just what it would do, Ryan Bohm has gone off out of control and not communicating for months on end, and designed a monstrosity I wouldn’t hook up to two AAA alkalines. It’s dangerous at best and a mess stem to stern.
May 13 was the LATEST date he missed. He offered to return my money and I accepted. Unfortunately, he has since gone dark. No check. No PulsaR. And the bit of a user guide he sent confirms that he has built something entirely different than we discussed. You’ll have to use your mOTOR windings to charge, your car, it will only work with series DC motors, it has a broken controller INCLUDED in it, and I’m just furious.
Sure. It will turn up. Just not on EVTV. And who knows when. He was CERTAIN he could deliver it by November 15, 2012.
I’ve learned a hard lesson here. Generosity is probably not a winning gambit in EV space. There appear to be any number of James Morrison’s out there….
sorry to hear about the Pulsar problem. just curious, why not get the service of say Evnetics or other controller builder to build it?
I DID get a controller builder to build it, and I paid $20,000 up front for 10 of them to sweeten the deal. Never delivered. And they through the design spec we agreed to in writing out the window as their first official act.
That’s really too bad that the PulsaR ended up being vaporware. I was really hoping that it would become reality by the time I started my build, basically waiting for the new HPEVS and 144V Curtis pairings, and for the perfect roller — both of the latter pretty much happening in the same week.
Along those lines, there seems to be really only two chargers left in the running: the Elcon (EVTV edition), or the Brusa NLG513. What is the best option for a 160V pack of 48 cells? ‘Best’ being subjective, I realize — but money aside, which one is more suited to a lower voltage pack? I was set to buy a Brusa in the online store, but held back because it seems to be suited for a higher voltage pack. I appreciate any wisdom you can share.
[BTW: My previous post wasn’t posted for some reason. If I am breaking forum etiquette, just let me know. An EVTV noob here, but excited to be part of the 100K army]
You kind of have it. It’s all about voltage. The BRUSA has an 11.5 amp limit I think, regardless of voltage. Not bad at 350 volts. But we would charge 48 cells to 168 volts. I so happen to have a TC Charger here at 4kw/168v/30a. At the max voltage, it will only do about 24 amps, but that’s over fast as twice a charge as the brusa. The Brusa is smaller, more elegant, but you would need two of them for that pack, which actually works fine except for being $4500 in charers for 22 amps, compared to 24 amps for $1895 with the TC Charger and controller.
I’m sorry to hear that things haven’t worked out with the PulsaR. It sounded like a neat idea but I did always wonder if it would be tied to series DC motors since that’s what Ryan is known for. I thought that he and George were going to do some AC stuff some day but I don’t think that’s happened. So, I guess I always figured that the PulsaR was more suited to people who are going to use DC motors.
The fun thing is that nothing really stops a motor controller from also being a charger. Well, technically you do want more transient protection if you’re going to use it as a charger but still the hardware is capable. The DMOC, if properly programmed, could be a 110kW battery charger from 240V three phase industrial power. That would certainly top off the average battery pack in a real hurry. Anybody want to try it? 😉
Manufacturers simply don’t care what the SOC is when they ship! First charge and discharge cycle is called the forming process and is part of the secret sauce of each manufacturer. The second charge and discharge is used to measure the capacity of the cell. This info is used to sort the cells by capacity. (In the 18650 manufacturer I worked at they broke it into 6 bins of capacity and price. Large format cells only have 2 grades typically) Finally they add a set amount of amp hours to all cells. Since the cell capacities vary and a constant Ah is added, the resultant SOCs vary a fair amount. Check the Ah capacacity that the manufacturer states for cells in a group versus the voltage when they arrive. You will find the lowest received voltages have the largest capacities…..
I had the opportunity to test 80 cells they here 160Ah Thundersky´s , in 4 of them I noted that they were a bit of all the others
Good cells out of the box were all at 3,299V and 3,302
the bad cells over time drop in voltage ,
I had made a test to determine if LiFePO4 had self-discharge , so of the 120 cells I did the following
After 22 months in storage I got 10 cells, and did a capacity test , all but 1 cell were at 50%(+-2%) SOC
After 26 months in storage I got 20 cells and did the same tests, all but 2 cells were at 50%(+/- 4%) SOC
After 30 months in storage I got 10 cells and did the same tests, all were at 50%(+/- 4%) SOC
On the others 40 cells not tested after 30 months in storage all were at 3,299V and 3,302, except one cell that was at 2,444V !
the bad cells, 4 of 120 , were showing sefl-discharge, it is believed that impurities in active materials and/or in the separator my cause “soft-shorts” and lead to self-discharge.
These batteries were manufacture by Thundersky in January 18 of 2010 (that is what the paper said )
That is very interesting Celso. I must check the OCVs of my CA100s which I’ve had for some months
Its interesting about taking the Calb cells and dropping them down straight away, the Instructions on the thunderskys was to charge them at CC to 4.2 before using them at all, I have taken that to mean even before bottom balancing them, and taken them down after that initial charge.
If anyone in the UK has any blue Calb 180s I could do with one or two as spares for my Smart. This had 24 Thundersky 160s in it and having swapped to 25 Calb 180s the slightly lower voltage is a detrimental move and an other cell would make a nice difference.
I was worried about the big dip in voltage of the CALB Cells but they do just keep giving as Jack commented on recenetly.
All within 0.02 of a volt after sitting around for 9 months
I saw something on the BYD website recently to the effect that they have observed no degradation of their LiFePO4 packs with consistent fast charging of a taxi fleet. Check out their e6
Two weeks ago I took the first drive in my EV conversion – the culmination of 1.5 years of effort. Last September I ordered enough CALB CA180 cells from you to make the local UPS delivery man mad at me, and I can’t say enough good things about them, now that I finally have them in action. Living the EV dream.
Nice one Joey – do you have any photos or youtube vids?
Joey, your EV Porsche 911 project (http://eporsche911.blogspot.ca/) was sure a major inspiration to get me going on my own — many thanks for that. I’m not sure I’d be attempting this right now, if not for the detailed photos, steps, diagrams, tips, component selection, etc. You are finishing as I am just getting started, so like an EV-911 baton being passed.
BTW: You got a gold one, I have a silver one. If someone here converts a bronze one, we could all show up at a future EVCCON and take our places at the podium…
Any chance of some more batteries becoming available?
Regardless, Thanks for all you do!
Excellent and most excellent. We are curiously unsympathetic to UPS here at EVtv. But they seem enamored of us for some reason, weight for the most part I would guess. Everything here is heavy.
The cells perform very well. And I hope you found both the process and the result of your EV conversion enormously satisfying.
The video of my first drive is posted on my blog: http://eporsche911.blogspot.com/2013/06/video-first-test-drive.html
Very nice ride Joey,
You have to be very proud of that build….
Oh, what a feeling! Congratulations on your successful project. Any chance we’ll see it at EVCCON?
Thanks everyone. And thanks to Jack for running the trustworthy EVTV store, because writing a big check (big in my world) for batteries is stressful enough without wondering if your order will actually show up. I wanted to bring the EV to EVCCON, but we already booked a trip to Disneyland during the same week. I can’t say no to the kids at this point. Next year, I hope to be there.
A very nice build. I did take a peek at your blog and your video. That is indeed a gorgeous 911. Sell the kids for whatever you can get for them and join us at EVCCON 2013.
That said, your well intended efforts at battery balancing may have been counterproductive. I thought I was the only one on the planet to actually spend the effort to TOP balance a pack on purpose and by hand. You may be the second. In any event, it is not a good thing for your very handsome and expensive set of CA180FI cells.
The cells vary slightly in capacity right from the factory. You may have cells as low as 189Ah and as high as 203Ah in that set. Most will be clusted around 195-197Ah – about 90%.
First, charging the cells in parallel, even if you leave them for several days, does not precisely “balance” them with regards to state of charge. The resting voltage of the cells varies in tiny values, while the differences in SOC can be quite large – watch Anne Kloopenberg’s investigation of cells on receipt in this weeks show – just posted. All had almost identical open circuit voltages, but the number of amp hours actually stored varied from 18Ah to over 50Ah on his CA100FI cells.
So there is insufficient voltage difference to cause any useful balancing current flow between the cells.
But worse, in balancing them at the top of charge, you are moving ALL the variation to the BOTTOM of the discharge curve. If you ever inadventantly overdischarge the pack, the stronger cells quickly eat the cells of less capacity or at a lower SOC, reversing the current through them and destroying them quite irrecoverably. Driving on the top half of your pack, you will not likely ever notice this. But you’ve built a bomb in your battery pack, and sooner or later it is going to go off.
This can have grave consequences. If you DO go on a long drive and reverse a cell or two, in a pack that size you may not notice it. The cell is irreparably damaged, but you get back to the garage just fine and plug in the charger. It charges to a specific voltage as hard as it can. With a couple of damaged cells, they will not take a charge in the normal way and can indeed go into an overtemp situation that tends to cascade – leading to a battery fire. If they do, they will achieve a high enough temperature to set the good cells adjacent into thermal runaway as well.
End of car, garage, and perhaps house.
Before even driving this car further, I would urge you to individually discharge EACH cell to as close to 2.75 volts resting as you can get them. This is termed “bottom balancing” and it is much more accurate that top balancing anyway. The cells are perched on teh vertical wall of the discharge curve and at rest. At that point, small variations in cell voltage DO indicate variations in SOC.
Once all the cells are manually balanced at 2.75v each, you can charge the pack normally. We usually charge these cells to 3.50 or 3.55 v X the number of cells in the pack. This undercharges them slightly and avoids the huge divergence in cell voltage that occurs when attempting to charge a series string at 3.60 or 3.65v.
We do put out a little document on cell care available at http://media3.evtv.me/cellcare.pdf. Please read through it.
Thank you for that LiFePO4 care document. The EV club at the local community college just received their pack of CA cells and we had a meeting they day they arrived so I insisted on giving presentation about how to treat them and how not to treat them. I’m going to make sure each person gets and reads a copy of the document you created.
As I recall, you’ve been on a long road of learning about these cells the hard way. The EV club is fortunate to have you.
I’m pleased you found the cellcare.pdf useful.
Yes but at least I haven’t lost any cells, thanks to you. I’m glad you found my blog and contacted me about my extremely high top balancing voltage of 4.00V. Next month will be 2 years since I balanced my pack (top due to the limitations of the Zivan chargers) and I don’t detect any signs of the cells drifting. I do stay off the bottom so rarely go to 80%DOD. If I get to where I’m needing to push my range I’ll figure out a way to get my charger to stop on time and bottom balance my pack.
David D. Nelson
Has anyone inquired about the Rimac motor? They are advertising that they are selling their components now and say on their website ” Don’t worry – at Rimac Automobili „custom“ doesn’t equal „extremely expensive“ They are cooling the rotor of their permanent magnet motor with light oil.
Hey Jack, nice show once again as always ! So, I wanted to chime in on something you mentioned about UL listing and the Anti-Islanding of the chargers. I’ve been involved in solar power for about ten years. About four years ago one of the major inverter suppliers built a battery based, grid tied inverter that was guilty of the Islanding effect. It was a major issue as several hundred of them were in service at the time this was discovered. And thus UL stepped in made some major changes to anything that involves batteries and grid tie. But you probably already knew that, didn’t you !
@Anne and Jack (and Celso if you are around) I thought after three rounds of tests I was beginning to understand these LiFePO4 cells, but you just casually lobbed in a bit of a hand grenade. Have you got a view as to what is happening with these low charge outliers? Your spreadsheet seemed to imply that two cells at the same rested OCV could be at radically different states of charge (like 18% versus 50%). Or were these not the voltages at the start of your test, but the voltages when they left the factory? My experience to date (with the Headways and CALBs) is that they are pretty close in SoC out of the crate, and also that the voltage and SoC curve, whilst shallow, ain’t flat; but I’ve been working with far smaller numbers of cells and may just not have had any outliers.
Would you check something? Would you put your 18% outlier on the shelf for three months, record the OCV every week and at the end of three months do a full cycle to see how much energy you get out and how much you can put back in? If OCV and SoC have declined the smoking gun is likely to be a defect that is causing self discharge.
I dd hear from a gentleman this week that had a cell that was slowly bleeding down in voltage. Clearly discharging. Actually at about 4 mv per day. He said they were CALBs but did not say whether SE or CA series. He didn’t get them from us so I don’t know much about them.
Again correct me if I’m wrong, but what I got from Anne’s test was VERY minor differences in voltage, and very major differences in the amount of energy contained. I don’t quite understand this.
And yes John, one of the peculiarities of thsse cells, or maybe all cells, is that the more you test, the more questions you have and the more there is to be tested.
Bang on I think Jack “…VERY minor differences in voltage, and very major differences in the amount of energy contained…” Presumably a self discharging cell in a bottom balanced pack would be quietly eliminated by Darwin: forced into cell reversal by their peers, trashed and weeded out as a failed cell. I wonder how often it happens?
I have been watching my cells for quite a few charges now. They do seem to have their own personality. I have two that shoot up in voltage right at the end of charge which I cut off at 125.5V (36 cells). My PLC stops the charge when these two cells hit 3.6V.
The first time I completely charged the pack they shot up to 3.86V just before the constant voltage stage terminated….
I am not sure why they do this and it does not appear to affect their capacity. In fact they seem to have a higher capacity that some other cells that do not go up as much in voltage at the end of charge??? I kind of expected the weaker cells to do this….
I am going to bottom balance my pack again and see if it affects the top end charge of these two cells and also to see if they bottom out the same as they did before…
Hi Jeff – yes that is consistent with what I’ve observed: the cell voltages on two of the three packs I’ve tested writhe around like a bundle of snakes in the CV portion of the charge curve. With the pack of CALBs I’m testing now I stop charge soon after the start of the CV phase. The thing that has me rubbing my eyes at Anne’s data was two cells at the same voltage but very different state of charge. That is seriously weird
I think one must be careful to draw any firm conclusions from Anne’s initial reporting on the discharge cycle of his cells. There a few variables in his test set-up that could contribute to some of the admittedly puzzling results he has observed.
John’s request that Anne observe voltage variation over extended periods on the outlier cells is a good place to start.
On John and Jeff’s points, please bear in mind that one is amplifying any cell variation at the top end of the charge cycle if adopting a strict bottom-balancing protocol. Great care should be taken during the CV part of the charge cycle as a result. Generally we don’t even bother.
I have good direct contact to the CALB factory, and plan to visit there in the next few months. I have not queried the exact protocol they follow during cell prep, but depending on Anne’s further results, may well attempt to establish first hand what is actually done at the factory. One early observation though, is that their production batches don’t necessarily correspond with their shipment batches. One must therefore not assume that cells in the same box were produced in the same batch, or at the same time. You can see this phenomena on the battery no. in the test report, not the barcode!
Out of interest John, I also have good contacts at BYD, as we are going to use their Solar PV modules on a grid scale PV facility here in S.A. later this year. I actually visited their facilities in Shenzhen last month and was able to test-drive the E6. It is a reasonable vehicle, although performance is not exactly in Tesla territory.
I also had the opportunity to view their latest prototype EV, due to be launched in China next year. This is one to look out for as it is part of a JV with Daimler, and the influence of German engineering in the design and execution is evident.
They also have a grid scale LiFePo storage facility (approx. 1MWh) linked to a pilot PV facility, which is quite impressive. They are however firmly on the cell level BMS train, although they have no real choice due to warranty constraints.
I have had some interesting discussions with their technical staff on the merits/risks of BMS vs. non-BMS; although their BMS design is probably the most elegant I have seen as BMS systems go.
@Anthony. Yes I agree absolutely about care in interpreting the results. Interestingly my observation is that cell voltages don’t actually diverge much during the CC charging phase. On one recent cycle chosen at random all 8 cells were within 24 – 28 milliVolts of each other right up to the start of the CV phase with no discernible trend. And I’d agree about the benefit of dropping the CV phase altogether in practice
I’m very interested to hear more about your BYD and CALB experience. I have some material from BYD which suggests that the e6 can be charged from a 3 phase AC source at 40kW using the controller/inverter backwards. On paper the e6 looks very capable with masses of torque and a decent range.
The E6 can also come with a built in TV! 😉