Well, the transmission on my 2004 Taurus seems to have given up the ghost at 76.5K miles this past weekend — the diagnostic computer implicates the torque converter. The bad news is that, ff course, this is just 3000 miles over the expiration limit for my extended warranty. But the good news is that, once again, the Taurus got us all safely home from a long trip before needing to go into the shop. That’s honorable failure in my book. Nevertheless, you have my vote for taking your time to be as careful and thorough as you can with the Roadster’s transmission. I am literally feeling your pain (soon to be matched by a pain in the area of my wallet). Too bad the Whitestar isn’t here yet: now would be a good time for me to switch.

Ironically, from what we can see on the outside, the Tesla battery system appears to be one of the most reliable components. The transmission and the NHTSA appear to be Tesla’s biggest challenges.

Darryl, just post the clarification message at Better Place’s community at projectbetterplace.com/community/index.php

Tesla is the best! Support Telsa forever!

Not to worry about battery rumors or any other kind of rumors. Just put more and more miles on those pre-production Tesla Roadsters and let us know how they’re doing. Meanwhile, let’s start a “juice-y” rumor about the oil industry. Two summers ago, gasoline cost over $3.25 a gallon when oil was about $50 a barrel. Now with oil over $90 a barrel, gasoline is under $3 a gallon. Why? Here’s the rumor: the oil industry is becoming weird, because of all-electric cars.

It’s good to hear about Tesla and Th!nk’s battery partnership. There was much speculation.

I guess that Ener1 is the one providing the batteries AND the battery managemetn systems then…

Oh, and articles like this display time and time again how so many articles are inaccurate, or at least have a factual mistake in them. (I’d say about half, or more, of teh articles I read about EVs have incorrect statements.)

Another possible source of battery concerns was your own blog.

Here:
www.teslamotors.com/blog4/?p=59
was written:
“# Tim Johnson wrote on August 8th, 2007 at 6:53 pm
## I noticed that the battery was de-rated by 5%.
## I thought I’d read it differently in the whitepaper, but I checked and that said the same 53 KWh.
## At least you guys are consistent, but it seems slipping that into the whitepaper without annotating the
## change to a paper stated as written in 2006 is a bit duplicitous.
## Not that big of a deal, but to me that sort of behavior is rather irksome.
”

This edit in your whitepaper to change the kWh rating of the ESS may have had some speculate that you had to change batteries. There was also mention at a similar time about “durability concerns leading to delays” which could have caused some to think that battery durability was an issue.

The current version of the whitepaper on your site says “stores about 53 kilowatt hours of electric energy”.
An archive.org version says “stores about 56 kilowatt hours (kwh) of electric energy”
web.archive.org/web/20070421083156/http://www.teslamotors.com/display_data/TeslaRoadsterBatterySystem.pdf

The change from “over 250 mile range” to “over 200″ also added to the speculation that different batteries may be under consideration.

TEG2-

While I haven’t read the whitepapers in a while, I read (and reread) the blogs fairly often. (and I’d imagine that the kind of journalists who we’re discussing wouldn’t want to dig into anything as “technical” as the whitepapers; they’d stick to the blogs.) The blogs have been quiet clear: the decrease in the range had to do with an increase in weight. The battery pack is the same battery pack that was always planned.

On a related note, there really wasn’t much of a decrease. The Roadster is EPA rated for 245 miles/charge.

-Jim

Thanks for putting the truth out there, Darryl. One thing that has always set Tesla apart is your transparency. But there was a moment there, when Micheal Marks took the CEO spot, that Tesla Motor’s direction as unclear. The course corrections that Mr. Marks put in place where not fully explained, and that lead to speculation.

Perhaps a post in the Think category from Mr. Marks would go a long way toward dousing some of the speculation fire. I imagine he came in saying, “Whoa, boys! These are all good ideas, but let’s take a deep breath and work on one thing at a time.” Well, maybe not in so many words…

“Software Executive Shifts Gears to Electric Cars”:
online.wsj.com/article/SB119362189984274462.html

Sir:
Perhaps the error was yours. When you have been aggressively producing positive PR and you have accepted cash for future delivery, you have in effect created a support community with additional stake holders, other than just company entities; some have just an interest in new technology; but, others with a monetary investment, could be damaged by negative news. So, how do you announce production delays to this community after you have concentrated so heavily on positive PR? I suggest that when you delay the project, you also give a reason for the delay and publicise it well. Otherwise, your community is left to speculate, rumors become real and your credibility is called into question, as it is now.

Might I also suggest that you meet your announced date for “streeting” the cars with no additional delays. Each date announced and not met affects your credibility within the support community. And please remember: It only takes one “Ah S..T” to negate 100 “At-A-Boys.”

In a recent WSJ interview, Martin said that Tesla was delayed a year from the original schedule because you had to engineer thermal management into the battery pack to prevent potential safety concerns. That could have been a source of the confusion. It turns out this delay pre-dated the 2007-1/2 plan and was not the cause of the recent delay that pushes it into 2008.

The transmission failures you describe look similar to when the car broke down on NBC’s Today Show several months ago. NBC News reported that Tesla said that it was the result of (something to the effect) of extreme and abusive testing. Are the breakdown problems a bit more generic than that?

On the point of rechargable battery packs: this could work if
a) the packs are simple to exchange [within a few minutes]
b) the packs are widespread [many, many cars using the same pack], and
c) the packs are leased instead of owned [so you’re not screwed when your empty-but-brand-new pack is exchanged for a full pack in its 500th cycle].

One day maybe. Right now it looks more appropriate for a Th!nk than for a Tesla. With time, who can tell?

I always thought the delay was mainly due to transmission, especially after Tesla changed suppliers. I’m no engineer but I imagine it must be pretty difficult to make a smooth, reliable change with no clutch and 13,000 revs on the clock in first gear. If only they hadn’t promised 0-60 in 4 seconds we would all be happy with just second gear and a 0-60 of, say, 6 seconds. I’m just guessing here but that sort of number doesn’t seem unreasonable. The Wrightspeed only has one gear I believe and achieves 0-60 in 3 seconds, albeit with a lighter car and a lower top speed. Ah well, nobody said it would be easy, right?

Fabrizio Ferrari:

it is still stated that “There are also issues of safety with existing lithium-ion batteries that have become unstable under extreme temperatures.” Is that true? If so, how Tesla solved it?

Tesla put all the batteries in a wine-case like container thingy. In this container thingy (called the ESS) there are loads of electronics managing the batteries and fuses and everything. The batteries are housed with some sort of fireproof material also, so it is all designed that if a battery were to expldoe, nothing would happen.

“I always thought the delay was mainly due to transmission, especially after Tesla changed suppliers. I’m no engineer but I imagine it must be pretty difficult to make a smooth, reliable change with no clutch and 13,000 revs on the clock in first gear.”

Try driving a motorcycle or a high performance drag car that has a sequential transmission. With a blip of the throttle and slight pressure on the shifter is it possible for extremely fast & smooth gear changes with minimal or no use of the clutch.

From what I’ve heard (please correct me if I’m wrong) the problem is that Tesla wants a completely electronically controlled gear change. That is; the gears separate, the electric motor changes to its proper speed with the new gear selection, re-engages. Literally no clutch is used.

Why did they go this route? I guess because on paper it should have worked. The problem is that whatever the design says, it takes too long for the shifting to occur (ie a lot longer than manually shifting gear and a whole lot longer that the performance paddle shifters out there). If they went back and used the design most others use for paddle-controlled sequential transmissions, their problems would likely be solved.

Or at least, that’s my view from the outside looking in.

Hello Tesla;
Any more comments about the transmision delays/problems? Is this the real reason for delay in delivery or are there other more serious problems we have not heard before? I have to believe the battery pack is OK because it passed all those gov mandated tests… Any thing else go wrong?

By the way, I think lowering the ordering price tag to 5K is a great marketing ploy. I can now just order one on my credit card and wait for the delivery date as required. Much easier than dropping 50K cash as before. This should really make the “sold” volumes improve a bunch.

Have they? Does anyone know the current order book? Is 2008 sold out? 2009? What are the numbers? To bad the blog is not two-way with more comments from a Tesla person.

Cheers!

Consider www.zeroshift.com. They are in the UK so Lotus can introduce you to those brilliant minds.

“If the transmission uses a clutch for shifts, what do you do to engage/disengage the clutch?”

I won’t try to answer the rest of your post because I don’t have a complete understanding of what Tesla has and I don’t want to spread incorrect information. However, I can answer the above.

Semi-automatic transmissions (aka paddle-shift/clutchless manual transmissions) typically use an electronically controlled hydro-mechanical unit to engage/disengage the clutch. In other words, instead of you providing the power/timing to actuate the clutch thru the use of the clutch pedal, the car provides it.

When I previously stated that Tesla should use the same designs as others, I meant the same semi-automatic transmission that companies such as Ferrari and Toyota have used, not that they should put in a clutch pedal for the driver to use. From what I’ve heard, Tesla’s transmission takes a couple of seconds to shift, Ferrari’s most current transmission takes 100 milliseconds.

There seems to be a lot of interest in the transmission, especially because it looks like it is the primary cause of the production delay. Any chance we fans could get an in-depth blog on this issue?

The transmission has 2 clutches that mechanically isolate the transmission during shifting. They are electro-hydraulically operated by the computer in response to the shift lever setting. When the car starts (powers up) you can hear the hydraulic system pressurizing pump whine a bit.
The new transmission shifts right quick. Way under a second. Zak told me that they are still refining the millisecond choreography of the process.

It makes we wonder why a transmission is needed at all in an electric drive? Can’t the motor be adapted in such way that it has some additional coils for the low rpm speed-up and which are skipped at high rpm levels? Take the stepper motor for instance, it has several coils. Can’t this aproach be combined into the electric motor with those additional coils being skipped at higher rpms? I thought the goal was a KISS style system. The most simple that comes to my mind is to make the transmission’s 2nd gear and clutch (if any) simply obsolete.

Tesla Motors get issues with interesting engineering problem. As I can see following difficulties would exist with all stages of pure electronic gear change:

1. eMotor have to be deselerated just a little to nullify the torque it apply to gears before eMotor gets disconnected from wheels inside gearbox. Either accurate rotation angle tracking would be needed (hard to do) for gradual gears disengagement should be done or very precise timing of rapid gear shift synchronized by dynamic event of short moment of zero torque on the flip from acceleration to deseleration would be required. I would select rapid gear shift approach here.

2. In disengaged stage eMotor RPM has to be rapidly changed almost twice. This should be fairly easy to do if short but high electric power spike could be applied to eMotor or absorbed from it. This stage should not be too different from normal acceleration or regenerative braking process - just control firmware difference.

3. Most difficult problem I guess would be to synchronize not only rotation frequency of eMotor connected gear and wheel connected gear but also exact angle of mutual rotation. Here I believe comes most of the real difficulties.

In stage 3 I think it is absolutely unavoidable that angle of rotation should be precisely controlled. But induction motor in contrast with DC brushless does not allow easy control of rotation angle because of “slip”. So dynamics of regulation for both velocity and angle with such a eMotor should be very difficult for any feedback loop. Also angle regulation would be by general feedback loop nature relatively much slower compared to RPM regulation. Typically with any feedback loop better precision comes with slower regulation. So it is very believable that Tesla Motors engineering team is fighting now difficult trade off between acceptably small time of gear change and required precision to avoid high degree of gear damage during gear change process making gearbox unreliable. One likely overlook which could have happen is that mass manufacturing process introduces unexpectedly high and unpredictable variations of mechanical dynamics of all rotating parts inside gearbox. As a result required accuracy versus duration of gear switch could be unachievable without some intelligence in the gear switching feedback loop to learn particular dynamics of gearbox and apply advanced dynamic prediction algorithms to reduce deviations that feedback loop should compensate hence improve timing versus accuracy trade off.

I never enconter any information that established cars manufacturers have successfully done fully electronic gear switching. Popular solutions of gear switching uses some sort of mechanical friction to synchronize rotation frequency between gears. Such friction and hydralic systems used in automatic gear switching also radically dump gear collisions when gears are finally shifting into lock that requires slight rotation angle adjustment by mechanical forces. Reproducing all these good old gear switch behavior with pure electronic feedback loop is very risky problem to take for mass manufacturing.

But if solved pure electronic gear change would manage extremely clean, low maintenance, no oil change solution for any car. By itself such a technology could be a value to sell in itself to any car company. So here I see typical risk taking of electronic start up company nature.

Williness to take such a challenge induces in me respect to the engineering team. The sad fact is that the problem managed real extra risk for Tesla Motors business existence. I wish the challenge would end up manageable after all.

From my experience I could mention the following techniques to improve feedback loop precision versus regulation timing behavior :

1. Have fairly realistic prediction model to predict gear motions dynamics. If adequate computational power exist and model is accurate enough it would reduce deviations for feedback loop to compensate hecne precision of regulation.

2. Kalman filters approach exist known from missile guidance and other military self-guidance applications. By defiinition it is the optimal filter for fastest feedback compensation. The filter would require accurate enough model of gear motion dynamics but it would manage faster regulation than non-specific feedback loop of low order typically used for general stability.

3. Manage frequent (at every car power up for example) self-calibration of parameters of model and feedback loop. This would resolve the issue that sophisticated models of gear dynamics would be too specific to particular car and too variable within car lifetime.

Tesla Motors team likely knows all that but mentioning it does not take much resources. Of course already selected design constraints certainly rule out arbitrary out of context ideas so nothing could really help.

In case it would be of any use I wish Tesla Motors engineers good luck with the problem in hand.

To Jurgen Goritz :

Exact reasons why Tesla Motors decided to use two speed transmission are not known at least to me. Use of that transmission allows maximum traction limited acceleration to sustain under largest fraction of eMotor power. This is first gear. But in that case second gear is needed to get maximum speed. I guess it is possible to use twice larger maximum power eMotor (two of them for example). In that case I think top acceleration and maximum speed would be managed withing gear change. Most likely weight saving consideration enabled the decision to use single eMotor, gearbox and differential. Now it is way too late to make any major change.

And this one:-

www.zeroshift.com/demonstrations3.html

David Kosowsky wrote:

“I’m in the 2 e-motor camp (not that this is feasible for 1st generation Roadster at this point). Seems to make more sense to have a second, smaller e-motor up front for traction, regenerative breaking, and efficiency. I don’t think wieght would be affected that much by trading the transmission for a small e-motor. This would be even more advantageous for Whitestar and other future models. One motor is “tuned” for low velocity acceleration and the other tuned for higher velocity acceleration. One for lift-off regeneration and the other for break pedal regeneration. Perhaps some of the more technically oriented folks could comment on the potential issues with this. ”

What?

Is there such thing as a 3-Phase AC motor being “tuned” for certain power bands? Two motors would certainly cost mroe than having just one motor, and then you’d need a transmission to connect the two, and not to mention a larger/or another controller to control the seperate motor. Plus, you’d need more wiring, and it would mos likely take up more space. I can’t think of any benefits.

It would be great if Tesla please clear up the tranmission situation, from start to finish, in some good detail. I’m an avid watcher of Tesla Motors, but am completely confused about what is happening here! Your transmission story would be especially good for the media to hear since they, along with us, have been wondering what is the source of your delays, and then you won’t have silly reports saying the batteries are cauing the delays.

About two eMotors configuration :

Most naturally each eMotor would be connected to its own rear wheel. In that case gearbox and differential can be removed. By the way I am not talking here about wheel motors. It would be just two eMotors of current design instead of one located approximately at the same position as current eMotor. Each eMotor would be connected via fixed reduction gear and proper shaft to corresponing rear wheel. In this case it would be no issue of sharing power from two eMotors together via fancy mechanical linkage.

Features of such configuration would be:

1. Weight increase would be at most ~70 lbs as claimed to be weight of single eMotor. I guess it would be some extra weight of modified inverter electronics because inverter have to have totally different schematic. But it would be some weight saving of removing gearbox. So overall weight increase would be ~50 lbs I would think which is marginal increase on top of 2700 lbs of total car weight.

2. It would be no gear switching needed. In case if to maintain car dynamics gear switching should be kept its effect could be electroniocally emulated.

3. Fixed gear ratio should correspond to current second gear. Current first gear manages twice bigger RPM reduction and twice bigger torque compared to second gear. So with two eMotors it would be possible to manage the same torque as today coming from two eMotors but via second gear reduction ratio. As a result same top acceleration and regenerative braking as today would be managed.

4. Reliability issues of gearbox would be avoided. Also because it would be no gear switch practical acceleration to 60 mph and to 100 mph would be somewhat improved. Also better traction control could be managed because torque distribution between rear wheels could be realigned actively and in wider range compared to differential. Differential behavior should be emulated electronically so control firmware would be more complex - which is manageable.

5. Car mechanical reliability would increase because of no gear switching in mechanical way.

6. eMotor cooling would be better because each eMotors would work at most at half of current top power.

7. Cost would be modified by exchange of gearbox cost by eMotor cost. Extra IGBT transistors cost would add up too. Extra control firmware development cost burden would be required. But all these cost factors I guess would hardly add more than $5000. But this adder would come with better reliability and car dynamics.

8. Surely control firmware and electronics should be very significantly modified for such configuration but ESS and all other major car components would stay the same.

I do not know why such configuration was not selected. I personally see it somewhat better than current Tesla Motors architecture. But it is surely way too late to even think about such kind of changes. Now the only choice is to make all selected components and existing car architecture work together at expected performance. If gearbox dynamics versus reliability will not be manageable to fix it would cost many months (may be total recertification so hence more than a year and extra millions of dollars) of production delay at the very least to exchange gearbox or switch to 2 eMotors. Major features of design surely were freezed with first VP arrival. Most likely Tesla Motors hope to fix issues without any drastic changes to car design so nothing like I described is expected even remotely.

Joseph:
>> Is there such thing as a 3-Phase AC motor being “tuned” for certain power bands?
It’s efficiency can be tuned for certain rpm bands. For example one low-rpm efficient motor in the back and one high-rpm efficient motor in front.

>> Two motors would certainly cost mroe than having just one motor, and then you’d need a transmission to connect the two, and not to mention a larger/or another controller to control the seperate motor.
Motors aren’t free of course but you have to compare other options. Two motors give you 4 wheel drive. You DO NOT need a transmission between them, that is the biggest plus there is. Front motor runs only front two wheels, rear only rear wheels, no transmission, no central differential, no long drive shafts. Two eMotors give more starting torque so the need for transmission is even smaller. All you need is a front and rear differential and halfshafts. You get very adaptive 4WD for the added cost of a smaller eMotor and around 40kg of added mass.

>> Plus, you’d need more wiring, and it would mos likely take up more space.
A wattermelon sized motor in front is not an issue at all.

>> I can’t think of any benefits.
I can think of lots of them . 4WD rocks

After checking out the website and looking at the animations, I echo the comments of admiration about zeroshift. Where can we find accounts of real-world experience with this mechanism that sounds almost too good to be true?

Also, I was able to get my Ford transmission reconditioned without spending too much money, so I’m back on the road again, hoping for Roadster sightings. Let’s hope the reinvigorated transmission lasts me until I can find an EV (with luck, a Tesla of some kind) that meets my needs.

P.S. to above possibly wacky electro-magnet clamping idea: the first and second gears would always be meshed with the gear on the electric motor shaft, and both gears and gear shafts would so always be turning. But on the short shafts that each of these gears is welded onto, a mag clamping device clamps down-this device is also around a shaft going to differential gear, linked to drive shaft. In other words-at the end of each short gear shaft is a shaft of the same diameter, right in line with it, like a fraction of an inch away, or whatever. At this junction of in-line shafts BAM! , as John Madden would say, the electro-mag sleeve (which is “normally” sprung open on springs) , grabs onto both gear shaft and the shaft adjoining it that goes to the differential gear, making both shafts become/turn as one. Of course the steel involved would be very high strength to hold up over time.

I wrote:

“I can’t think of any benefits [about using to eMotors}”

Warpedone wrote:

“I can think of lots of them . 4WD rocks”

4WD is inefficient, and would steal away plent of range. I don’t think Tesla Motors will go through the trouble to make something that reduces the range of their cars.

EETIM wrote:

“But better still is a no-transmission single motor with lower RPM and higher torque.

To eliminate the transmission, you need high current. Now we are back to the battery pack.”

Hmm…The reason for the two speed transmission is b/c to go to higher speeds, you must have higher RPM. The
eMotor cannot safely go to such high RPM, and so another gear is needed. the lower gear is used for higher acceleration instead.
I don’t believe more amps (current) gives higher torque for the kind of eMotor Tesla Motor uses. Even if, I don’t think higher torque helps increase top speed. I can’t say I am 100% sure about what I am saying though about the amps and torque thing. I am sure about the first two sentences I said though.

Searching throughout this entire site, the only information I’ve found so far on tesla’s battery life tests is that they warranty the
battery for 100,000 miles. I’m no guru, but I think my 10 year old son would definitely know to build a smaller pack with the same
cells that powers the car for maybe 10 miles, cycle the batteries maybe 4 times a day on a test track somewhere depleting about 80 percent of the charge each cycle, then logging that data into an excel spreadsheet for the customers to download. 4 cycles a day * 30 days a month. That’s 120 cycles per month. After 5 months of discharging the smaller version of the battery, I’d be real surprised if you could go even half the range of the battery as you did when the pack was new. I see no data from tests whatsoever tesla has performed like this anywhere on this site.

I’m glad the big shots over there know for sure that the batteries will still hold a charge after 500 cycles, which is less than two years.
Having customers do performance tests on this vehicle isn’t proving anything to the public other than what we already know about
electric drive systems. That isn’t what is threatning whether or not this car lives up to expectation, It’s the life of the battery. You can
count on that.

I’ve seen very intelligent people make boneheaded mistakes like this. Especially engineers. Let’s just assume that we”ll get 100,000 miles out of the pack without actually cycling the pack for those miles on several prototypes and not showing the customer we did this check, who just spent 100k with us. Someone from tesla, please correct me if I’m wrong about this.

You’re wasting your time on the stupid little nit picking rich clowns who care more about how big the lcd display is inside the car than
the life of the most critical part on the car and the most expensive. I’m skeptical of the life of this battery that has been made from the
same cells that fail in my laptops every 100 cycles. I can make a prediction the customers who use this car for long commutes are going to be pretty ticked when they find out after 100 cycles they can only go half the range.

I think the choice to use a BMS and heating system for the batteries was right on, as well as the EQ system,cell format, etc. but without checking the life cycle on a few prototype packs, you’re taking a huge risk with other peoples money. This could look really bad for EV makers. I wouldn’t spend a dime on another EV until I see some solid data from multiple sources to validate claims of battery life in the roadster. Way too much emphasis on performance and not durability.

What good is it to have the fastest car in the world if it only lasts 100 drives? To the other 90 percent of the public out there who don’t have the loot to spend on this car, performance isn’t going to mean jack squat when you’re driving to work in 45 mph speed zones and sitting in daily stop and go interstate traffic. Inevitably, This is probably going to be another endeavor to cater to the rich. I don’t see any evidence anywhere about making an affordable car, which you could already be doing with a much smaller pack using 1/3 of the batteries you put in there with a much smaller, much lighter chassis.

My toyota yaris would be the perfect size, with a 200 volt battery system and a 100 mile range. Build people what they need, not another toy some rich guy is going to keep along with 20 others in his garage to drive once a month. So, now, I wonder when tesla is going to finally stop wasting time building rich people toys and get serious about making the car I mentioned above. The average american in five years after the oil peak, will most likely not have 25k around anymore to buy the whitestar model. I hope I am wrong.

Best of luck

Re: B.A. Wow! I really wish some people would spend a little time and read the white papers before running their mouths off about issues that were addressed months (if not years) ago.

To EETIM :

Somehow people get exited with all latest battery technologies with Li-ion like LiFePO4, A123, Altairnano etc. As I could see all these new Li-ion flavor batteries are targeting:

1. higher power density per unit of battery system mass,
2. higher recharge cycle count,
3. higher temperature abuse resistance

4. lower energy density per unit of battery system mass as a trade off to get improvements mentioned above,
5. higher cost per energy capacity because new battery technologies did not mature enough on the mass market to be at lowest price (cost) point

than conventional Li-ion batteries with cobalt. Emotional perseption of all these new exiting battey technologies as certainly superior is understandable. But if we try to use rational arguments it is easy to see that advantages 1 - 3 are more important for smaller energy capacity battery system because for typical driving it would have higher recharge count and it would need higher power density. Disadvantages 4 - 5 becoming more important for larger energy capacity battery system.

It is fairly clear that ESS is not limiting Tesla roadster performance characteristics and it delivers decent range. By today standards ESS of Tesla Motors is really large battery system in which factors 4 - 5 dominating factors 1 - 3 mentioned above for Tesla roadster usage. So battery technology choice for Tesla roadster stays optimal from the engineering point of view. We have it like this:

1. Normal cars good old lead-acid batteries have energy capacity ~0.5 kWh. Here we have well established old technology. These batteries at highest power used to start up ICE engine. Power required here is 1 - 2 kW but for short time ( 2C - 4C peak discharge ). Sustained power is ~0.2 kW ( about 1/2 C discharge ).

2. Current hybrid cars like Toyota Prius etc has 1 - 2 kWh battery capacity. Here about decade old battery technology of NiMH batteries is well established. But this is still new to car market. Power required here is 10 - 20 kW. So here we got about 10 C discharge requirements. The problem was solved by using outgrow battery with total capacity of ~5 kWh used with shallow charge/discharge. Relative to outgrow battery we have ~4 C discharge rate. Cost impact of outgrow size battery was managed by the fact that energy capacity reequired was relatively small.

3. Plug-in hybrid cars require 5 - 10 kWh energy capacity in battery system. Also power requirement comes to 100 kW - 200 kW because now you have to sustain full car power electrically. Managing outgrow size battery here is cost prohibited. So you need a battery with discharge 200 kW / 10 kWh => 20 C. This is hard to do so ultracapacitors are frequently considered here as power boosters. But anyway 10 C battery with decent energy density is critical. Such battery systems are only now popping up, make excitement but are still several years behind coming into car market because of normal engineering development cycles. Best predictions target year 2010 but most likely it would be year 2012 at best I would think with all the typical delays etc managing 5 year delay which is short by car industry standards.

4. No punishment full electric car needs 50 - 100 kWh energy capacity for 100 - 200 kW power. So here we come back to

To Bruce Allen :

Information you are asking about reliability of ESS is also most interesting for me. But I am not Tesla Motors customer so I do not think I have any rights to demand anything from Tesla Motors in any way like you sort of did in your message. Strictly speaking Tesla Motors should not explain anything to anybody except potential customers. So I would think that only people approaching Tesla Motors to really put money on the table to enlist for a car should have access to information you requested. If you are such a person I would expect that Tesla Motors established some private method for communication about features you are concerned.

Tesla Motors is now trying to solve very important technological problem and without all the good will from at least some public they hardly could manage it through. Right now they are fighting for their very business sirvival because if they keep delaying their vehicle production further than they already did whole positive publicity they gain would crash and in this case we forget EV cars for a long time because of EV bad publicity reconfirmed as always happened before. I would like to see that Tesla Motors come through with real car production so that such outcome would not happen.

In general typical laptop batteries you deriving your experience from are not temperature managed in any way. Chemical reactions (basis of any battery) are very sensitive to temperature so I consider believable in general that carefull temperature management could deliver required reliability improvement. This is what Tesla Motors claim and it sounded OK to me.

Certainly proper testing data should be presented to justify real world car reliability. But at this stage of the game it is too late for these data to change the outcome. If Tesla Motors already made a mistake of relying on theory without proper testing of ESS lifetime they would not sirvive that as a company and EV concept reputation would be finally damaged much more than before Tesla Motors started. If they did proper testing it is done already. In this case it is not important that they prove or publish anything because car would work at the end. For good or for bad Tesla Motors gain exceptional publicity so in public image their success or failure define whole EV concept success or failure. For this reason I personally refrain from critical discussions until car manufacturing would finally start in real life. So far it did not happen and until it happens in real world nothing matters and nothing positive is done regardless of Tesla Motors self-image. Whatever they do now we never know actual ESS reliability anyway until we see roadster delivered to public and gain about a year of its usage by customers. At that point hopefully some alien use of the car would happen so we would know all its weak points against marketing claims. Whatever Tesla Motors publish in the blog no business self-inflict bad publicity. So even if some customers were not happy it is hard to believe we would see their opinions in this blog. So sure all customer experiences published here should look sunny as they do. This is hardly much convincing but Tesla Motors marketing is doing what they can under the situation they have.

#Bruce Allen wrote on November 4th, 2007 at 7:05 am
#[speculation about TM’s battery claims]

A couple of points:

1) Tesla Motors warrants the battery for 100,000 miles. If its engineers are wrong about the battery’s life cycle characteristics, the company will be in a boatload of financial trouble. Because the company actually wants to make money, and because the battery is such a vital part of the car, I would be surprised if the engineers would make made such a basic mistake.

2) TM has done (or is doing) long-term durability testing at a test track in England. As I understand it, the cars were driven as close to 24/7 as the battery would allow. The results of that testing hasn’t been made public, but presumably the battery’s ability to perform is part of that testing.

3) If you drive the car daily for about 165 miles — its worst case range per charge — for about 18 months, you will indeed exhaust the battery’s expected life cycle. But that means you’d be driving roughly three hours daily, and your nether regions would probably be permanently imprinted with the Tesla logo stamped on the Roadster’s seats. I would expect that most of its owners will drive closer to the national average of 30-40 miles daily, with full battery cycling happening once weekly or so, giving the battery pack six or eight years of real world usage.

4) Keep in mind that your laptop doesn’t have a separate climate control system for its battery, nor all the monitoring computers, etc., that the Roadster has. While few of us have any experience with carefully monitored battery packs, we might imagine how such care matters. After all, care makes the difference between a rose bush that blooms and one that doesn’t.

vfx, that is a pretty absurd comment. How is this any different than a car stopping at a crosswalk or light with streams of pedestrians in front? A car drives then stops. If this were grossly irresponsible then driving any car would not be possible. In case it makes any difference, the car is using a mechanical breaking system that is already proven.

Here is a link to an interesting article on the state of the art in ultracapacitors including a comment on EEstor technology. Still a way off, but Anatoly might like to comment on the capacities being touted.

www.spectrum.ieee.org/print/5636

# Ben Weiss wrote on November 5th, 2007 at 8:58 pm

## In any case, the Roadster works just fine in second gear, at all speeds.
## (You may not get the blistering acceleration, but you’ll still have 0-60 in about
## 5.6 seconds, which is none too shabby.)

Indeed. even 6 seconds is still pretty fast, but not exceptional.

## The two-speed transmission does
## allow for the combination of insane acceleration and reasonable top speed,
## which is critical for the sports-car market niche. Presumably, the WhiteStar
## would be able to live without the super acceleration, so I wouldn’t be surprised
## fi they drop the transmission entirely there.

Weight affects acceleration more than anything else. Whitestar will be heavier then Roadster and 7-8 secs to 60MPH is pretty much minimum you want for sporty family sedan. That might mean that gear ratio with single gear might not be enough to give it good enough top speed. Of course upping HP/torque a bit would solve that problem, but if it is as easy to do as it is to say they would have done it to Roadster already (for higher top speed).

Timo,

The top speed of the Roadster is limited not by HP/torque, but by RPM’s. Redline in second gear takes place around 130MPH; if they had added a third gear, presumably the car would have been drag-limited around 160MPH.

# Timo wrote on November 7th, 2007 at 8:32 am
## Roadster isn’t very light car: 1220kg. If you remove ESS and eMotor from that you still have 700 kg car. That isn’t really very light.

Say what? I thought the Roadster uses the lightest street legal, production chassis technology possible today.
What other road car (with full features like air conditioning) for sale has a chassis weighing ~700kg ?

The Whitestar is supposed to be a sporty family car, right? I so NO reason whatsoever to need speeds above even 85MPh in a family car. I’d say that if you’re trying to do 0-60 in 5 seconds and driving 120 MPh with kids in the car you deserve to be sent to a padded room until such time that you can prove that you do actually care about the welfare of your children.

A swift acceleration profile that doesn’t jar the bones and a top speed suitable for passing on the freeway can both be achieved easily with no transmission. This would reduce complexity, weight, and cost… seems like a no-brainer to me.

I disagree with Tim W. The car should be able to do 0-60 in under 5 seconds. Great acceleration should be a trademark of electric cars. Especially from the Tesla Motors brand. Tesla Motors should be all about quick acceleration and great handling. Start with a clean slate. Don’t inherit the bad habits of the Automobile giants. Electric cars should be DIFFERENT, and FUN!!!

Secondly, there are plenty of safe family cars out there that do 0-60 in 5 seconds or thereabouts. The BMW M-series, and the Audi RS-series comes to mind. Those cars are probably the safest cars out there. If I had kids I’d trust those cars. They have better breaks - to stop in time. Great handling - to avoid an accident without slowing down (moose test comes to mind). Great acceleration - to merge safely, or to avoid getting hit. In Europe this is referred to as “active safety”, while good brakes are “passive safety”.

Thirdly, ever hear of a term called “power reserves” or “reserves” in general. Just cause a sedan has the handling of a sports car doesn’t meat that one has to use it all the time. It makes one secure to know that in case something happens reserves are there.

# Tod G. Collins wrote on November 9th, 2007 at 1:49 pm

## So now here comes Fisker Automotive Inc. www.fiskerautomotive.com
## with a very sexy series hybrid 4 door sedan. Fisker will display a prototype
## at the 2008 North American International Auto Show in Detroit in January.
## However, I find that to date Tesla Motors is NOT listed as an exhibitor at that show.

Beautiful concept, yes. But only concept. There is no info about its performance, or anything, just range. For 80000$ car you want performance. How good miles / gallon is distant second. Whitestar will be pure BEV not hybrid, so it probably will not have as huge range, but as a BEV is true future car. Not some transition phase from ICE to EV.

Timo, I agree that Tesla should “give us something” they already have. They’ve told us that they need to complete and deliver the Roadster before they can start on the Whitestar.
They should do exactly that and not get distracted with the Whitestar since the Whitestar will not happen if the Roadster doesn’t.
Go Tesla!

With soo few Tesla service centers announced can anyone at Tesla elaborate as to when other centers are plaqnned? Might there be another one in the near South such as Charlotte or Atlanta? Either location has plenty of wealth to support this type of vehicle…. I am surprised that there will be a New York City center since most of NY types with the money use limo around town and Long Island. Not sure the NY demographics is a great fit but then there are always the NJ folks to think about.

If there is a way to be involved in a Tesla center in the Charlotte area, I would be very interested and would like to explore the possibilities.

Cheers

Unfortunately Tesla has to come to grips with the fact that the current ESS is not a good design for mass production. A better approach would have been to use the current ESS for prototype development, but target a production battery using any of the currently available prismatic type configurations from a number of companies that can be produced in a size and shape optimized for vehicle use. The current ESS, while a marvel of engineering, is an unnecessarily complex Rube Goldberg solution due to its use of a cell designed for universal applications.

Common sense tells you that anything with 6800 plus parts is not going to be a reliable mass produced item.

##Don A wrote on November 13th, 2007 at 11:46 am
##Absolutely! Everyone knows that is impossible to mass produce something
##like the average modern automobile which has over 15,000 parts.

I agree, automobiles will never last very long. 15,000 parts is obsurd But besides that point, automobiles will never sell…who would ever buy something that contains explosive liquid in it

##OzoneLevel wrote on November 13th, 2007 at 6:48 am
##Common sense tells you that anything with 6800 plus parts is not going to be a reliable mass produced item.

OK, enough playing around. OL, if I can call you that:) Yes more parts means more CAN go wrong, but it doesn’t mean that more things WILL go wrong, or that it can’t be mass produced, and mass produced reliably…possibly even more reliably than the “prismatic type configuration” you mentioned, it all depends on how the systems are mass produced. (Poor mass production techniques vs good mass production techniques). As Don mentioned, automobiles are a good example- thousands of individual parts coming together in a mass produced vehicle that is reliable (well, at least Honda does a fairly good job with reliability). So I say again, back up your statement with facts. You can’t just throw a statement out there and expect everyone to consider it a “common sense” statement.

Tesla web page comment: Any chance of updating the web page so that we can turn the Roadster 360 degrees? The still pictures are OK but being able to rotate them would be supper.

flabby makes my point. Current automobiles contain thousands of parts and have a known reliability factor, some good some bad. Now take those same number of parts and add another part with 6800 plus subparts. Simple math will show you that failure rate will be higher.

Here we go again. Today’s (11/14) New York Times “Small Business” section has page 1 article (”Taking A Whack At Making A Car”) that states Tesla “…was delaying production –reportedly because of problems with the battery technology– …” Why do Times writers keep putting out the wrong reason for a delay in the Roadster’s delivery and why is it always the battery? Maybe the answer is that marketing a modern car that runs on batteries only is difficult and any delay must be due to problems with its gasolineless power source. Hopefully, by Spring, 2008, one of America’s happiest sights will be Tesla Roadsters zipping by everyone of those “Reg. $3.49.9″ (or more) signs.

OzoneLevel

“Common sense” and “simple math”?

www.vicorpower.com/documents/quality/Rel_MTBF.pdf

Once again, NY Times is misleading the readers on the reasons for Tesla’s delays:

“Tesla has had its struggles. It originally said it would have cars for sale by mid-2007. But it recently said that it was delaying production — reportedly because of problems with the battery technology — until the first quarter of 2008, when it will sell its first 50 cars. The company said it would have 600 more cars ready by the end of next year.”

www.nytimes.com/2007/11/14/business/smallbusiness/14cars.html?pagewanted=2&ei=5087&em&en=2e67443d7a0147b7&ex=1195275600

Is somebody paying these writers to post wrong, potentially harmful PR-wise information?

www.autobloggreen.com/2007/11/17/video-riding-the-pch-and-more-in-a-tesla-roadster/

From what I gather, the base components that make up the roadster or a similiar EV can be quite inexpensive. KTA sells kits, although you don’t get the AV 100,000 watt controller, for under 5000. I would be interested in just purchasing the vehicle without the battery system. Since the lightweight chassis of these types are not available to the DIY people.

Why not make the battery system more modular, like the memory on a PC mainboard. Like the base car only costs say 30k, but you only get a 10kw battery module or you could opt to not get any modules. There would be additional slots within the ESS to simply insert more modules for more range. This would get around the battery cost until you come up with a way to make it affordable for everyone. This modular design would safeguard possible problems with certain groups of cells going bad in the unit. If you ever did have cells in the ESS die, just that module could be removed and not the whole pack. Having to replace the whole ESS seems wasteful to me in the event something did go wrong.

If I had the money, I know there is a way to make the car lighter than 2500 lbs using an alluminum alloy chassis. I would bet this approach would be much cheaper and easier in the long run. Something in size proportion between the smart car and the toyota yaris. This IS the ideal size for what normal people who work for a living need. I would build this car myself if I had welding skills and material suppliers. The IDEAL car that people will buy, is the size I mentioned above, 100 miles per charge. 1500 cycles @ 60 percent DOD, loading at the C/1 rate. a motor capable of hitting 80 mph, and the ability to add modules to the ESS. Now you’ve got price flexibility.

Once you’ve got your material suppliers in place, this shouldn’t take you that long to build. Of course, I’m not factoring in any of the saftety tests, but learning from the roadster this shouldn’t be a hang up anymore either.

Darryl Siry

You wrote:
“The primary cause of our delay has been and continues to be issues with durability and reliability of the transmission”

Any news about this? Have the problems been solved? Will the Whitestar have any gearbox at all if this is a problem to you?

Gordon Green: I think Tesla’s interests would be better served by delivering Roadsters to customers. The last thing the EV world needs is another company hyping up a non-existent product.

# Raymond Michiels wrote on November 27th, 2007 at 1:23 pm

## Not updating the blog for three weeks, not even after the LA motor show might make people think…

…
Somehow I have started to feel this same. Maybe Whitestar is just waporware and Tesla is quitting with pockets full of money after few Roadster sells. No info about Whitestar. No progress updates. Nothing. This makes me wanting to turn my focus on some manufacturer that does have the muscle to build one of the EV:s I probably do want. Like GM, Ford, Mitsubishi, Toyota, etc. Even that those are not as innovative companies as Tesla has been this far. I have serious urge to warn people in other forums and media that Tesla might be just another waporware-company. They have nice and expensive toy car with few sells, but their talk about next generation cheaper practical car is turning out to be false promise. Warn people not to buy here, wait for some other company that can deliver what they promise. Warn them not to buy Roadster either, because with new competition Tesla might be ready for bankruptcy pretty soon and you might not see your car ever and your money might be gone.

Tesla: GIVE US SOME INFO ABOUT WHITESTAR!

In case anyone missed it on the home page, Darryl recently had some comments posted here:
venturebeat.com/2007/11/23/how-to-invest-in-alternative-fuel-vehicles/

Getting people to start thinking about alt energy vehicle investment opportunities might be inline with thoughts of an IPO…
A year plus of pre-paid customers sounds like a win so far.
Lets just hope that nothing stands in the way of delivering completed Roadsters soon.

Construction is well underway on the LA sales / service office.
farm3.static.flickr.com/2364/1910046980_0834752053.jpg
farm3.static.flickr.com/2211/1897386042_0be0dbfaae.jpg

I think we have a bunch of addicts suffering from withdrawal symptoms here

Consider this: Tesla has sold out their 2008 model year. The cars will self-market if they start getting delivered. The blog’s original purpose was to attract customers, and with this proof of demand, attract enough financing to finish the job. Tesla has all they need right now, so the blog is, to be honest, hardly necessary at this point.

Raymond Michiels: I’d wager on “busy”. I actually suspect Tesla’s been busy long enough that people are taking days off again to avoid burnout. Either way, progress is made.

Timo: I think you might be overreacting just a little bit. Even if everything you say is true, Tesla proved there’s a market for EVs so it’s just a matter of time before somebody builds one

# Ryan Lamansky / Kardax wrote on November 28th, 2007 at 11:25 am

## Timo: I think you might be overreacting just a little bit. Even if everything you say is true,

I don’t think they are, it is just how I feel right now.

## Tesla proved there’s a market for EVs so it’s just a matter of time before somebody builds one

I know. GV Volt is one of the other ones.

What bugs me is that some time ago there was discussion about why Tesla doesn’t get their factory build and then there was answer that building the factory isn’t a big deal, problem was in designing Whitestar. So they have _something_. There just isn’t any info about what that something is. Nothing. Except that is will be pure BEV, sedan, probably four door, and it will cost somewhere around 70k-50k$.

Even just Teslas own minimum requirements for Whitestar would be enough. Raw sketch of what it should able to do.

All discussions here about Whitestar has been pure speculation. I, for example, made one based on Mazda 6 sedan model. It was pretty amazing what the result was, but it still was just pure speculation.

TEG2: I think that article you linked has messed up Roadster and Whitestar. A few lines later there is this:

“The result is that Tesla has collected deposits from 600 customers, which amounts to a $30 million interest-free loan. Of course, Tesla’s customers might have judged the car cool enough with only a single-speed transmission or uncomfortable seats or lame headlights.”

That obviously talks about Roadster, not “midrange sedan” Whitestar.

Dear Darryl,
What is the biggest obstacle to producing a wholly electric commuter car such as the defunct EV made by GM in the 1990s? I understand that none of the bigger companies i.e. Toyota, Honda, etc… are pursuing this line and there is great demand for a wholly electric car as opposed to a Hybrid. Toyota’s Prius is a best seller simply because it is the closest to what I am describing. The hybrid initiative, together with the Hydrogen one are not really as effective, efficient and cost-effective (to the consumer) as the wholly electric car. Is there any reason why Tesla Motors is not attacking that market? You would become the No.1 automaker. What personal computing and economical Japanese cars have shown us so far is that real innovation is derided at and the “leaders” will not adapt until they are not the “leaders” anymore. So why doesn’t Tesla Motors take the COMPLETE lead?
Yours sincerely,
Alejandro Szita

Malcolm Wilson wrote

## Hilarious.
## I remember back in Feb - March 07 there was a four week delay in the new blogs and people
## were posting nonsense about tesla disappearing or being vaporware!
## Chill people!

Scratch that. Tesla’s still here but the heart and soul of the company has been removed.

www.autobloggreen.com/2007/11/30/martin-eberhard-takes-a-further-step-away-from-tesla-management/

Dear Mr Musk,
Please take a giant leap for mankind.
www.spacex.com/media.php?page=56

I’m sorry. The first thing that came to mind after reading the news was “They killed Kenny, you…..”. Apparently the word describing an illegitimate child is picked up by the halibut filter (I certainly didn’t mean it as a personal attack, just a reaction). I really am quite upset though and I agree with Dean; you will lose your fan base rapidly if you do not explain this decision. We have all grown quite fond of Martin and this is shocking.

# Roy wrote on December 3rd, 2007 at 9:07 pm

## Bruce Allen: You are way off the mark. Kit cars either rely on all the chassis,
## doors, windows, seats, dashboard, suspension etc of the base car being
## modified or are bare bones like racing cars. Their cost is a small fraction
## of the actual cost to build a new car from scratch. Second, the battery pack
## size was chosen to provide the desired performance at 1C to 2C,

This is interesting point. 6800 batteries with 4 volts each and 375V nominal voltage of the pack gives figures of about 94 serial and 72 paraller. 200kW power output would require 533A current which is about 7.5 A each battery. That is a lot. I think. halibut, my old knowledge about electricity if fading, I can’t figure out what is “2C” current in terms of amperes.

## if a smaller 1/5 size battery system was used to save money, the acceleration
## and maximum speeds would be reduced proportionately, not just the range.

So even that we could get 4 times better batteries for energy density we still would need about same amount of batteries to make same performance unless power density of the batteries raise proportionally as well.

I think one of the good things in Altairnano and A123 batteries is that they have very good power densities which allows smaller battery pack reaching same performance. Those obviously reduce range too much for pure EV:s, but they are good for serial hybrids with small battery packs.

Hi Guys
I so want a Tesla (early adopter nerd!) but not at £80k.

Even at £50k I have worries about battery range over time (not distance).

I put together a spreadsheet to work out battery car range based on charge, average speed, air resistance, drive train losses etc and I get a range for the Tesla of 182 miles at 60mph. (I used Elise Cd figures and weights along with published Tesla weight info)

And Yes I have too much time on my hands but I had been looking at range models for designing my own car and used my model to look at battery technology and donor car options.

Based on the Li-ion loss of charge over time from here: www.buchmann.ca/Article9-Page1.asp
you lose about 12% of charge a year based on 25 degrees C storage and a storage charge of 70% (gets worse if you store it changed with more juice according to the site). (I averaged the 12% based on the graph for 40% and 100% charge). I also optimistically calculated the % loss each year based on the previous years charge (not year 1) and assumed the % degradation is the same each year. Here’s what I think happens to the range:

Time Range/miles
New 182
1 year 159
2 years 138
3 years 121
4 years 105
5 years 92
6 years 80
7 years 70
8 years 61
9 years 53
10 years 46
11 years 40
12 years 35
13 years 31

That would worry me lots, but it is not really addressed anywhere I have seen by Tesla. (If I missed it please show me - I so want to be wrong!) Oh and this is just degradation over time and does not include recharge cycle degradation from actually driving the car.

I have quotes from China for 7000 2000mAh laptop cells (same electrical spec as Tesla’s) at $1 per cell. So factoring in other costs a refreshed battery pack is going to cost upwards of $10k or £5K before any mark up.

Using 30 13.8volt Altair Nanosafe batteries, given the available space etc give my model a new range of 131 miles dropping to 111 miles after 10 years from Altairs quoted 1.5% annual degrdation. Problem is they are much more expensive and difficult to get! I supect the Lighning car must be using around 60 batteries to get a sensible range unless my sums are way off which is more than likely.

If someone has better info on Lithium battery degradation over time please tell me, but based on what I have found I am put off a Lion powered car!

Would this affect your approach as a purchaser?

I don’t want to diss Tesla as I love what they are doing but to me spending that much cash on a car is a car for life and so life expactancy is IMHO important. (iI have a an Elise I have run from new for 10 years and only something like a Tesla would replace it!)

All the best
Phil

Batteries are the major parts of the car. So this article is really very informative.

Battery is the most important part of your car.

thank you for clarification. thanks.