Monday, July 5, 2010

A 200 mile EV or a 13 mile PHEV? You choose.

The big news of the week (after Brazil loss in the World Cup and the iPhone 4 antenna issues, I suppose) is the IPO of Tesla . With a IPO price set at $17 per share, Tesla saw its shares increase to $30 at some point. On Friday, it was back down to $19.2 a share, but I think we can all conclude that this was a successful IPO. The company got some much-needed cash and the early investors cashed out. The 1st week run reminds me of another greentech "success" story- A123 Systems.

A lot has been said by various analysts on the problems with Tesla (e.g., They have not yet made money and have no chance of making money for the next 3 years), but I think the IPO shows that its possible for a small company to compete with existing players. The same can be said for A123.

Just because a company has a successful IPO does not mean that it is really successful. Tesla has a lot of problems to deal with, chief among them the fact that their cars are a tad bit expensive. Similarly, A123 continues to bleed cash and competition is increasing. Its not clear when one should consider a startup to be successful. Is it when they start becoming profitable, or is it enough if the investors, founders, and early employees make money?

Anyone who has worked at a startup knows that its a roller-coaster ride. Its not the proverbial "two steps forward, one step back". Its more like "ten steps forward, nine steps back". Everything seems magnified. A million things have to come together to be successful. Often times one has to change direction (remember that A123 was not a LiFePO4 company when they began) and this can be hard to do. Suffice to say, start-ups are not for the faint of heart. For all the guys who went through this ride, getting to an IPO will probably be considered an amazing success (well... I suppose 6 months from IPO would be a more accurate date because that is when you can sell).

For the rest, being profitable may be the criteria for success. Obviously this is no easy task. A lot has been said about the ability of a Tesla to take on, say, a Toyota (or a Tata, depending on the market) or a A123 to take on, say, a Sanyo (or a BYD). All these are valid questions and make for interesting speculation. But I think the approach taken by Tesla and that by Toyota exemplify the differences between a start-up versus a traditional giant.

We all know Tesla's approach well. They want to commercialize a pure EV with a 200 mile range. They buy laptop batteries with energy approaching 180 Wh/kg and make battery packs with energy approaching 150 Wh/kg with a total energy of 56 kWh for the pack. Assuming that their car design gets them ~250 Wh for every mile*, they are pretty much using all the energy of the battery with very little guard-banding (meaning, they use close to 90-100% of the battery capacity).

Contrast this with the news that Toyota is coming out with a Prius PHEV using a Li-ion battery. Total driving range on the battery-13 miles! Toyota argues that most commutes are less than 10 miles, but a look at the battery specs is revealing.

It appears that the Toyota battery pack is ~330 pounds and has a energy of 5.2 kWh which means that the gravimetric energy of the pack is ~35 Wh/kg! I can only assume that this is useable energy (meaning the battery will have more energy but only 35 Wh/kg is used)

Granted that Toyota would want a battery that lasts 7-10 years and so unlike Tesla, they probably are using a battery that has a lower energy than 150 Wh/kg for the pack. But one would have to think that they are atleast using something that should be greater than 100 Wh/kg. Which means that Toyota is really only using 35% of the total battery capacity (at best). Talk about guard banding!

Just so we are all on the same page, you can get 35 Wh/kg from a Ni-MH battery. One is left wondering why Toyota would want to use a Li-ion battery with such a low State Of Charge (SOC) range of operation. One can only speculate, but it would logical to think that this is one way to get the life to be better. They will operate at a lower voltage and not allow the SOC to swing too much (remember the battery rules: don't charge them too high, don't swing them too wide...).

Moreover, if the battery is only charged to a partial SOC, then if there is a safety incident (leading to, what is referred to in the industry, as a spontaneous disassembly. For the normal person, this could be called an explosion) then the lower state of charge helps decrease the impact of the incident.

All this makes sense, but what is telling is that Toyota is being very very safe in their move to a Li-ion from a Ni-MH cell (by starting with a battery that is comparable). One wonders if this is more a PR move to tell the world that Toyota is moving to the latest and greatest battery, rather than using these batteries to actually get more performance.

Compare this to Tesla which is buying laptop Li-ion batteries (which are typically the highest energy density battery you can get your hands on) and trying to squeeze as much from them as possible. One is going for incremental, the other revolutionary, one prefers an appliance-like vehicle, the other a "sexy" ride, one could be considered boring, while the other could be considered a bit brash. No prizes for guessing which one is which.

Its easy to see Toyota's point of view. All you have to do is open the newspaper (I use "open" to mean clicking on a web link) to see their recent trouble with, this time, the Lexus brand. Toyota is got to be thinking that the last thing they need is a battery-related issue. Better to be safe and boring than sexy and sorry, I suppose. They cannot afford another recall.

This difference between a startup and an established player probably resonates across all areas, not just batteries. Remember Amazon in the late 90's taking on the big box retailers, or any of the open source softwares (Firefox or Linux) taking on Microsoft.

The only difference: If you screw up your internet software all that happens if your browser crashes or worse, you are infected with a virus. If you screw up your car, things can be a little bit dicey!

Time (next 3-5 years) will tell if these newer kids on the block will succeed in being profitable. Personally, I'm keeping my fingers crossed.


Disclaimer: I don't own shares in Tesla, A123, or Toyota (as far as I know. My retirement plan is a complete mystery to me). As a matter of fact I make it a policy of not investing in greentech. My instincts tells me that they are a good buy, but I have a policy of doing the opposite of my instincts so...

* The previous version read "250 miles for each Wh". Its actually 250 Wh for each mile.


  1. "I make it a policy of not investing in greentech. My instincts tells me that they are a good buy, but I have a policy of doing the opposite of my instincts so... "

    A very smart investing policy, Venkat. Investing with their instincts is exactly how most small investors lose their shirts.

  2. You do a great job of explaining what is going on in the world of batteries. Some part of your posts go a bit over my head, but I'm trying to educate myself and for the most part can keep up with everything you are posting. I just wanted to take a moment to let you know I'm still enjoying your blog! Thanks for doing this.

  3. Venkat SrinivasanJuly 6, 2010 at 7:00 PM

    Cleanenergywonk: Thanks. I follow the George Costanza principe (from Sienfeld fame) :-)

    JC: Thanks for the post. It does help to have feedback like this. Questions, if you have any, also help.


  4. Hi,

    Love the site - thanks for taking the time to do it!

    Quick question: the Prieto Battery is another one of the startups that I remember reading about late last year. Now it looks like they've got a web page up (I think it launched today).

    Have you heard anything from/about them? Would you be able to comment on what they're doing and whether or not you think that their design is novel:

    If you can't comment then I understand.

    It sounds like they've got a commercial prototype right around the corner.

  5. Venkat SrinivasanJuly 7, 2010 at 8:13 PM

    On the Prieto Battery: Its difficult to judge. I have seen some of this work in the past, but that was related to making an anode using electrodeposition. So we only have the website to go by.
    In general as Jordi Cabana was pointing out in his post high surface area is not a good thing in anode for lithium batteries (see This is one issue to look out for.

    Other than this, they are trying to make a 3D battery. I know a bit (probably quite a bit!) on this topic. 3D designs by themselves are not novel. The devil is in the details on how its made. Its not easy making a battery, so you have to give them some time before a prototype comes along. Dont expect something soon.


  6. Thanks Venkat! I'll remain cautiously optimistic.

  7. Thanks Venkat,
    I've been advocating (politically) that a market for electric cars sans batteries would be appropriate for this point in the market. This gives one the option of suiting the product to the required commute or shopping trip or size of town.
    Plus changeout to a scooter, motorcycle, or solar energy weekend.

  8. This raises a question or two... it seems that we're looking at extremes: Toyota make batteries to last longer, while Tesla pushes to almost full utilization. Then does that mean that even if we settle for something between, we are still going to hit the roof? If we speculate further, even if we narrow down the nitty gritty by getting the best engineering pack, can we get 1.5x or 2x more energy density with reasonable performance with current technology?

  9. Anonymous on extremes: The Chevy volt is supposed to cycle to 50% State of charge range. Somewhere between these extremes. In general, they way this will evolve, companies will start with small SOC swings and keep widening as they lean more (this is for the big auto makers who think life is crucial).

    As consumers we make the choice to buy a car will less reliability because we gain something (usually lower cost). My last car was a saturn. It used to drink (not leak) a quart of oil every 750 miles. The doors would not lock and two of the windows did not roll up. I knew going in that Saturns were not recommended by reviews for reliability, but the price was right. Similarly, I suppose we can imagine cars where the batteries are cheaper, but dont last as long. But this assumes that battery cost decrease to a point where we can even start to make these kinds of compromises!

    How much will current technology improve: This is a bit of a loaded question (what is current technology mean, for example, is not clear). In general I believe that a 1.5x of today's technology can happen, but will require time. 2x can theoretically occur, but requires luck (read: some innovation that we cant schedule). For those who are paying attention, I see this as a change in materials. See by post titled "In batteries, 2+1=1..." at

    The post is a bit technical, but details ways to get to these large energy densities.


  10. Hi Venkat -- catching up on some back reading here. But have you got a units type mistake here when you describe the Tesla battery pack? It's about 150 Wh/kg, and total energy of 56 kWh, so a mass of say 373 kg. But then you write:
    "Assuming that their car design gets them ~250 miles for each Wh, they are pretty much using all the energy of the battery with very little guard-banding (meaning, they use close to 90-100% of the battery capacity)."

    But I think you meant something more like their design gets them 200-250 miles for the whole pack of 56 kWh, which would be like 3.5 - 4.4 miles per kWh? I don't have a feel for what kind of vehicle efficiency (miles per kWh) is sensible for these things, and I know that the Tesla roadster is a very light vehicle. Gasoline is 36.5 kWh per gallon, and if you get 30 miles per gallon, then you're getting about 0.8 miles per kWh, which is just a rotten efficiency, but of course the energy content of gasoline is about 13 kWh/kg (clobbering batteries). So -- did I do that right? Is the Tesla getting 4 miles per kWh?

  11. Venkat SrinivasanJuly 15, 2010 at 2:56 PM

    Dr. Pablito- The sentence should have read 250 Wh for each mile. Typo. Good catch. I corrected the article and added a footnote. Thanks for the catch. Appreciate it.


  12. Great blog!

    I wonder if the larger battery pack for Toyota also serves to reduce the C rate (effective discharge rate) needed to achieve a certain power for acceleration or high speed driving.

  13. interesting! although I have no expert, but I want have to know more and more, on your blog just interesting and useful information. Keep it up!