Many thanks for your patience last week as I played a bit of hooky to attend my daughter Jacquelyn’s graduation from Colorado State University in Fort Collins Colorado. She received a bachelor’s degree in Journalism and Technical Communication – kind of a chip off the old block. She’s grown into a fine young woman of 24 years.
As a result, this week’s show kind of piled up on us. I cut about half and we are still at 2:41 minutes. Our record was 3 hours but this was still a long hard slog. I finally got it up Monday, for you all a holiday. No rest for the wicked.
Tesla is of course the big news. Their stock closed at $97 Friday and opened at $113 this morning. We are anxiously awaiting their Supercharger announcement, even though it was announced months aga. Perhaps in announcing it again it will get better. More than their car, this fast charge station deployment across the U.S. is actually the disruptive end of Tesla Motors in play. It could potentially devastate the plans of other automakers who will no doubt be left out of the party. That would indicate they can announce their own, and more effectively build their own. Nissan announces their own charging network monthly wiht fast charge stations at all their dealers across the country. Unfortunately, it would appear the dealers never get the word as they are not installing any such thing anywhere. But Nissan does keep announcing it.
This places BMW and General Motors in an interesting but unenviable position.
Although my personal use of electric cars is rather a local affair, and I think that is well and good unto itself. But fast charge enables cross country travel and removes one objection to electric vehicles rather effectively. My change in this came after the realization that virtually ALL the batteries in use can do this very well NOW. We need not wait for better batteries. The problem is getting a point of fueling that provides sufficient power to actually accomplish the feat. These little 25 kW and 50 kW “fast chargers” are nothing of the sort.
While I have waxed poetic on the concept of 500 convenience stores with Tesla superchargers and Ho’ho’s and Ding Dongs, in this episode I talk a bit about the possibility of battery module swapping being part of the announcement. This comes at an ironic time with BETTER PLACE announcing bankruptcy this week. But a 10 minute swap is feasible with the Model S, though not the Roadster of course. THAT would be pretty fast charging, but it is NOT the real import of such a move.
Tesla has a bit of a problem looming with their battery packs. These packs are made up of 14 modules, each with 7 “blades” and each blade containing a variable number of cells but nominally 69 in an 85 kWh pack. Each cell is now over 3 AH and indeed the latest Panasonic boasts 3.4Ah per cell. And therein lies the rub. In riding the wave of consumer electronics cells with all the advantages of economies of scale, and being able to simply deploy the latest by substituting in the same form factor, pretty strong reasons to do this, there is a little wee difference between pocket flashlights and consumer electronics and the Tesla Model S.
Panasonic gets kudos and undoubtedly increased sales when they bump their energy density to 3.4 Ah per cell. They also get kudos for increased power density when they can then do 10 amps per cell. And of course they gain more sales every time the price goes down. But these cells offer relatively short cycle lives of 800 cycles or so. And no one in consumer electronics really cares. They are the type of cell you replace anyway. The 800 cycles is generous in that context. And no one would pay extra for a 1200 cycle version of the same cell. They undoubtedly WOULD pay a premium if they could get 4AH. Or 15 amps. But because they are both inexpensive and replaceable, nothing for cycle life.
Cars are a bit different. And California is kind of mandating a 10 year 100,000 mile warranty on vehicle battery packs. Can the Tesla Model S meet that? Dubious at best.
From their latest SEC form 10Q filing:
The range and power of our electric vehicles on a single charge declines over time which may negatively influence potential customers’ decisions whether to purchase our vehicles.
The range and power of our electric vehicles on a single charge declines principally as a function of usage, time and charging patterns as well as other factors. How a customer uses their Tesla vehicle, the frequency of recharging the battery pack at a low state of charge and the means of charging can result in additional deterioration of the battery pack’s ability to hold a charge over the long term. For example, we currently expect that our battery pack for the Tesla Roadster will retain approximately 60-65% of its ability to hold its initial charge after approximately 100,000 miles or seven years, which will result in a decrease to the vehicle’s initial range and power. Deterioration of the Model S battery pack is expected to be less than the Roadster; however, such battery pack deterioration and the related decrease in range and power may negatively influence potential customer decisions whether to purchase our vehicles, which may harm our ability to market and sell our vehicles.
Normally, capacity ratings are done to 80% of the initial capacity. That the Roadster could be down to 60-65% of capacity after 7 years would indicate huge warranty claims int the future for very expensive battery packs. The admonition that deterioration of the MOdel S battery pack might be less is more of a function of the Model S being younger and so we can rewrite this later than it is of any change to the cells themselves.
Elon has gone on the hook in an earlier announcement that he is personally guaranteeing resale better than the Mercedes and BMW models. What would happen to the residual value of these cars if it got out that their battery capacity could be as low as 60% after 7 years, with a battery pack price of $40,000 or so? I’m guessing the word DEPRECIATION would no longer be adequate. And so Tesla would face huge claims on this residual value, along with HUGE claims for warranty on the packs themselves. This could total staggering amounts of moolah in getting well.
Were that the total plan, I would view the sale of each Model S as a ticking financial time bomb. This is corporate bankruptcy disaster by engineering design. After punishing the short sellers, does Elon Musk have a secret plan to become one?
Hardly. Let’s take a look at another extract from the same 10Q document:
Other factors that may influence the adoption of alternative fuel vehicles, and specifically electric vehicles, include: …. our capability to rapidly swap out the Model S battery pack and the development of specialized public facilities to perform such swapping, which do not currently exist but which we plan to introduce in the near future;
Tesla’s Hair Club for Men. In doing a battery charging infrastructure announcement, I believe Tesla intends to introduce the concept of a battery swap club. Given that they’ve added a $600 per year fee just for software updates, it is no stretch of the imagination for them to introduce something, even as high as $200 to $300 per month, to join a club where you get unlimited battery swaps.
And that rather neatly ties up the whole battery life cycle problem. The modules can be rebuilt rather easily, more easily off the car. Not only rebuilt with current cells, but even upgraded when newer better cells become available. By having dozens or even hundreds of swap sites across the country, let’s say you do NOT opt in for the swap club and the additional expense. Seven years down the road, if you register a complaint with the company over failing capacity under warranty, the fix is not so terribly expensive. They send you a piece of PAPER allowing a one time swap at a nearby facility. Your pack just goes right into the natural flow of packs that are recharged and if not up to spec, rebuilt. This effectively reduces the cost of warranty claims to about nothing. They could probably even get away with prorating the warranty with full replacement for the first three years, and partial payments required for each year thereafter culminating in full price at 10 years.
With such an infrastructure and program in place, the depreciation is kind of back to normal vehicle depreciation. Indeed the life expectancy of an electric motor and single speed gear drive could be expected to be much greater than an ICE engine and as features are mostly software updates in these cars, you might expect LESS depreciation than normal.
So in celebration of Better Place’s bankruptcy, I rather expect battery swap to be the main surprise in this week’s supercharging network announcement.
We were invited for a test drive of the Model S at the new St. Louis service center. I took two cameras and cameramen but the results looked like the Blair Witch Project during an epileptic seizure and so you are not graced with the results.
I don’t know quite how to react to the test drive frankly. It was mixed. The vehicle is just drop dead gorgeous of course. They actually had six or eight vehicles on hand. A shiny black one with buff interior is virtually the color of my Escalade and I fell swain if not swoon over it on sight. It’s a very pretty car.
It IS as predicted a four door sedan, favored by Lutheran Ministers world wide. Not precisely my style. Worse, it was harder to crawl into than the Speedster 356 with a very low roofline. Inside, the seats were actually quite confortable and visiblity reasonably good – out the front. Not so good to the rear. As we were in an 85K Supersport package, with the 4 second 0-60 acceleration, I guess I was prepared to be astounded by the handlign and acceleration. I have to admit, I was not. It drove like a well powered electric car. But the FEELING was nothing extraordinary compared to the eCabra, the Speedster, or for that matter the Spyder. I have to assume that all the raving is from people who have just never driven an electric car. It was indeed the SAME feel for the most part. Nothing shocking nor notable. Very good of course. But nothing new or different for me. So a little disappointing I suppose.
In fairness, I did not accelerate 0 to 60. More like 0 to 50. I was instructed to do so, on a rainy St. Louis residential side street one block past a 20 mph school crossing. I just looked at the little girl giving us the test drive. Punched it up to about 45 and quickly glided through 50 all quite under control and assuming I would have perfect brakes. Which I did.
So I really have nothing to say AGAINST the car, it was in all respects great. The 17 inch screen was truly a marvel. But I walked away thinking $100K, and it is basically JUST an electric car. With the range, and with a nationwide network of charging stations, undoubtedly worth it. But a car nonetheless. So a tiny, VERY tiny little bit of letdown. Like seeing a movie that everyone has told you would be great and your expectations got JUST a little bit out there past what a movie can do.
On the other hand, I got a little bit of lift as to how we’re doing with OUR builds and how the many viewers are doing with THEIR builds. If you cannot afford a Model S, even after the stock runup. Fear not. For a lot less, you can have most of that feeling, without a 17 inch screen.
A few years ago, a viewer approached us with a build already underway and stopped by to show us the battery boxes he was building for his S10 pickup – wherein he planned some excellent Optima lead batteries. I simply berated him mercilessly until he finally sent $16,000 to James Morrison in Washington for a set of CALB SE180 cells. They were never delivered. And the poor guy was tapped for ANOTHER five or six thousand in legal fees before recovering about $400 total from the “lawsuit” so clownishly handled by some of the other victims. He did purchase some cells AGAIN from US more recently and in fact his truck won BEST BUILD at EVCCON 2012 and now he’s a bit of a celebrity up in St. Louis showing off the vehicle at every opportunity.
But I’ve always felt badly about the expense this fraud cost this retired guy. We’ve been predicting a short squeeze on Tesla stock for OVER TWO YEARS at EVTV. It was most gratifying when he stopped by last week to note that he HAD been watching, and that I was not to worry any longer about the lost battery money. He’d more than made it back. Good on ya Dale Friedhoff.
No, we are not going to rename our show TESLA TV. In fact, I was very excited this week but not really about Tesla. More so about High Performance Electric Vehicles.
I’ve always liked this company and this product. We have done a lot of builds using their AC-50 motor with the Curtis 1238 three phase controller. It offered a 75 kilowatt AC induction motor solution with regenerative braking and a fairly easy to setup controller that was actually very “tunable” to dial in the “feel” of your car in a way that most controllers simply cannot achieve. They work. They drive the car fine. But with the HPEVS system we were really able to make small changes that made SMALL corrections in vehicle control feel. And the motor, dispite lacking liquid cooling, always ran cool. No brushes. No dust. And it was smaller – about 8 inches in diameter. It was just a great little package for anything under 2500 lbs.
The Curtis 1238 controller is nominally 108Votls which is kind of limiting. The 7601 version bumped power up to 650 amps which was welcome. But it was still kind of a weeny AC solution – though also very reasonably priced. A great value for the money. We quickly discovered you could ACTUALLY run it up to about 120vdc without complaints for a smooth 120vX650a or 78 kiloWatts.
Bill Richie, of HPEVS had been teasing us with tales of a new 144volt model that was coming for now TWO YEARS. Every time I asked he told me they were about a month and a half away from being available. Apparently, that’s what Curtis had told HIM. But the months turned into years and I had frankly given up on it. Once word came out that the current was dropped to 500 amps from the 650 amps, I really lost interest. No gain in power and actually a LOSS in low end torque. We can do MPH with transmission gears. I like torque at the bottom.
Well the controller finally arrived last week. And all that time Curtis delayed, apparently HPEVS put into play. The result is an entire SERIES of differing motor windings in a new larger 9 inch frame designed to complement the controller abilities. Better, the Curtis controller always did feature a program written by Brian Seymour of HPEVS in the Curtis Vehicle Control Language or VCL. Concurrent with the introduction of the new motors, was a spanking new version of the program.
And so there are several new features/upgrades of note. You might watch the video for a full rundown – the contactor issue went away. The dedicated 1311 Programmer issue went away. You can now officially IDLE with an automatic transmission. And there is a USEFUL CANBUS connection to get all the motor and controller performance data out so you can design a pretty display. Excellent, excellent, and most excellent.
And all very understandable. But a huge amount of work went into these new motor windings. Even better, and I actually secretly think they GOT this idea from us and our Dyno graphs, Brian Seymour actually BUILT a capable motor dynomometer in their shop to test motors and controllers on. The company always did buy Leeson motor parts but they always wound the motors themselves, almost a family tradition going back 30 years when their Dad was rewinding washing machine motors for the neighbors. So you could ALWAYS get a custom wound AC-50 if you knew to ask, and of course were willing to pay.
But the advent of an inhouse dynomometer brings this to the fine art level. Now they can try different windings and immediately see the results on the dyno.
At this point, that actually makes HPEVS something of a one of a kind jewell. A small custom motor shop that DOES like to sell to DIY home builders, hobbyists, as rainmakers for their more lucrative small run OEM business. They power the WHEEGO for example. They started in custom golf carts. And so they are kind of dominant in a nicely carved out niche of SMALL motors.
But they are kind of branching out into larger power levels, and the news is that this opens the door for HPEVS solutions for cars up to 4000-5000 lb class. That’s exciting. And so I’m predictably enough excited.
But the story is almost hard to tell. I fear many will fail to pick up the import of all this, because the telling of it gets a little technical.
The basis of magnetic drive is that a current passed through an inductor causes the expansion of magnetic lines of flux in what is termed a magnetic field. It actually takes electrical energy to setup this field. When you cut off the current, the field collapses and actually CONTINUES the current through the wire, until the field is gone. So this field stores energy. IT also interacts. OTHER conductors nearby will have currents induced in THEM as well as the field expands and collapses. And this happens ONLY when expanding or contracting, not when fully on or fully off.
This then is HOW we INDUCE currents into windings in the rotor of an AC INDUCTION motor.Those currents, in turn, cause a magnetic field in the rotor. And by manipulating these currents in the stator or “field” windings at an AC rate, and noting that the rotor follows but never quite catches up, the two fields interact causing an axial torque to be applied to the shaft of the motor.
Since this field is invisible, and since we can’t see or feel it, it must be weak and feeble. Actually no. You can feel something similar using some neodymium magnets. As you try to push two fo these magnets closer together, the fields of each magnet interact with the other as a function of distance. The closer they come in proximity, the stronger this interaction FEELS and at a still quite visible distance, you just can’t push them any closer together no matter how hard you try.
Actually, all the locomotives, power plants, and steam ships in the world acting in concert do not make enough power to put those two magnets together. It is almost infinitely powerful at close ranges.
In the case of the AC motor, the rotor and the stator are typically separated by an “air gap” of less than a tenth of an inch.
And so we get almost MAGICAL amounts of power out of a bit of current passed through a coiled wire. I never really get over it.
This power can be felt at stall – zero rpm. But as we turn the shaft, and speed increases, the rotor begins to generate it’s own counter electromotive force (CEMF) sometimes referred to in the vernacular as BACK EMF. Basically, the motor is becoming a generator and offering a counter voltage. The resulting current through the motor is a function of the applied VOLTAGE minus the COUNTER voltage from the spinning motor.
In most motors, it takes quite a bit of speed in the form of RPM for this CEMF to become significant. And so for the first couple of thousand RPM we have a quite stable amount of torque or axial force at play. THIS is what gives electric cars that marvelous feeling of continuous acceleration. And compared to an internal combustion engine, it is truly a HUGELY wide torque band. But it is not infinite. As RPM goes up, at some point the CEMF begins to be felt, and the torque of the motor, and the motor currrent, begins to actually fall.
We call this point the knee of the curve. The point at which torque begins to fall. In most motors, it falls quite rapidly.
Because they have no commutators or brushes, many AC motors are rated for speeds up to about 14,000 rpm. And people become fixated on this largely useless specification. Because the torque often starts falling off in as little as 2500-3500 rpm. And by 6000 rpm there is rarely anything left to do much with. So from 6000 to 14000 the only way to increase rpm is by not having a load on it at all. It is a useless area for electric vehicles.
So torque is almost entirely a function of the amount of current through the motor. But current is almost entirely a function fo the applied VOLTAGE. And if we apply a HIGHER voltage, across higher RESISTANCE windings, we can have the same current, but it will extend out further along the rpm line. Instead of 2000 rpm at 100 volts, lets’ do 3000 rpm at 150v. By using a higher applied voltage, we can change the point at which back EMF is felt and becomes a factor, and so move the knee of the curve out to higher rpms.
This would be a good place to talk about the term HORSEPOWER. Basically the relationship between torque and horsepower is given by the formula HP=ft-lbs X RPM / C where the constant C is a value of 5252.
So if we have 100 ft lbs of torque at 1000 rpm, we have 100,000/5252 or 19 horsepower. But if we have 100ft lbs of torque at 3000 rpm we have 300000/5252 or 57 hp.
And so you can see that your motor will exhibit higher horsepower, for a given torque level, at a higher rpm.
To see the import of what HPEVS et al have been up to, let’s look at some power graphs. Here is the familiar AC-50 motor at the nominal voltage of 108 volts and 650 amps offered by the Curtis 1238-7601 controller:
In this first graph, you can see the AC-50 at 108 volts. As our current rises to 641 amperes, the torque remains reasonably steady, starting at 120 ft.lbs and declining ever so slightly as the rpm rises to 107 ft lbs at about 3800 rpm. From that “knee” it dives to about 60 ft lbs at 5000 rpm and 40 ft lbs at 6000 rpm By 8000 rpm we have just over 20 ft lbs of torque. Our maximum horsepower occurs just after the knee at about 76 horsepower at 3800 rpm.
If we take that SAME motor and run it at a lower 500 amps we only get 86 ft lbs of torque. But it is fairly flat out to 5500 rpm. Because of the higher 144 volts we have moved the knee from 3800 to 5500 and so got 80 horsepower out of it despite decreasing the top torque by a third. And we still have over half our torque at 8000 rpm. This is how to get a much higher top speed of course.
The AC-50 had of course been wound specifically FOR the 108v curtis. Let’s take a look at the same frame, still at 144v, but now rewound and relabled the AC-51 to get our torque back up a bit, while still having a little higher RPM value:
By decreasing the number of windings and so their resistance, we can more quickly build up our current and so we start with 108 ft lbs of torque considerably higher than the 86 ft lbs in the AC-50. But we also maintain that torque out to about 100 ft lbs at 4500 rpm for 88 horsepower instead of 80.
HPEVS has also moved up from an 8 inch frame to a 9 inch frame with their AC70 series. Let’s compare the AC75 and AC76 windings at 144v.
With the AC-75, we start at 124 foot lbs of torque. We can maintain that out to 119 ft. lbs at 3900 rpm and our peak current actually occurs a bit after that at 4500 rpm. So does our horsepower at 92 hp.
When we go to the AC-76, we are wound differently for HIGHER TORQUE and so we start out at a much higher 168 ft lbs of torque which we can hold out to 153 ft lbs but the knee is now back down to 3000 rpm. And our peek horsepower is 88 at 3200 rpm.
So why two different motors? Well, a heavier vehicle or truck for example, you might want more torque like the AC76. A lighter vehicle or sports car would be very spry with the AC-75 and the higher 92 horsepower at 4500 rpm.
Think of the AC76 as a V-8 and the AC-75 as a high compression Porsche flat six or flat four.
They actually have an AC-74 whose graph is as yet untested that they are winding for super torque using the Curtis 1238-7601 at 650 amps.
And so HPEVS is basically developing an entire array of motors and controllers you can use to “tune” your driveline to your vehicle. Since we have already noted that the Curtis Controller is already the most capable controller in the field for tuning performance and feel to the vehicle, HPEVS would seem to be playing that hand to that strength in kind of an act of genius as far as positioning the company. The combinations you can do to tune perormance and feel become astounding.
And just when you think you’ve got a handle on it, look to the past. They used to have a popular motor called the AC-35. Let’s take a look at this remnant of their line:
Here, we have the motor with the earlier Curtis 1238-7601 108v 650 amp controller. This little motor puts out an impressive 128 ft lbs of torque, actually 7 ft-lbs MORE than the larger AC-50. But the knee is at 3000 rpm instead of the 3800 rpm of the AC-50, and so a horsepower of just 70 at 3100 against the AC-50’s 75 at 3800 rpm.
But again, if we use that SAME AC-35 at 144 volts and 500 amps, our torque drops to an initial 94 ft-lbs, but now it maintains that all the way out to an even 5000 rpm – still 88 ft-lbs for over 82.5 horsepower.
In fact, if you look at this graph, you see a remarkable thing. An AC induction motor with an almost entirely flat torque curve from 0 to 5000 rpm. You just don’t see that much. In fact, I’ve NEVER seen it anywhere.
In fact, we still have a third of our torque at 8000 rpm!
Oh if only it were not so small.
Ok, how about this. How about we siamese them? Let’s put TWO AC-35’s in ONE case on ONE shaft. What would that do?
Well it would require two controllers for one thing So you wind up with an amazingly compact 20 inch long motor with TWO 3-phase connectors on it. By the book, it should do 170 horsepower and 190 ft-lbs of torque entirely flat from zero to 5000 rpm – the perfect motor for an automobile.
But the graphs have not been published yet and it has not been dynomometered. So let’s go back to the beginning. When we send current through a wire, it causes a magnetic field. And that expanding magnetic field also INDUCES currents in nearby conductors. What happens when I put TWO sets of field windings in one short case, with ONE long rotor with windings in it?
Answer: I don’t know. Never done that before. We’ll see. I’m guessing the resulting graphs will NOT be identical to an AC35 times two. Perhaps close. But not identical.
But I am curious. If only I had a really lightweight car to stick THAT in. Like a carbon fiber Speedster at 1130 lbs for the roller. I would have to give up my dream of keeping it under 1550 lbs I guess. And indeed, I think a roll bar of some heft might be in order here. Instead of the lightweight A123 120v pack, I’m guessing 50 of the CA60FI’s would be in order. But I’m betting that little MF will get up off its ass if I tell it too. I wonder if it could beat an HPEVS Siamese AC-35 Corvette at EVCONN 2013.
I guess we’ll find out. I ordered one Monday.
We have been saying for some time that the holy grail of DIY custom electric cars has been the 150 kw AC induction system at under $10,000. I may have mentioned that to Bill.
Let’s see. 170 x 500 = 85,000. How about 170 kiloWatts at an MSRP of $9700. With chill plates call it a smooth $10K.