It's the Batteries, Stupid! - Page 7

AntronX · 12-22-2008, 09:01 PM

I've been told by a local battery tech expert that Tesla Motors went with lower capacity 2.0 Ah cells over newer and higher capacity 2.6 Ah+ cells because of failure rate. Higher capacity 18650 lithium-ion cells have thinner ceparators between sheets of lithium and over time, cycling causes small metallic impurities present in the cell to rupture that ceparator and cause a short. I have heard that 2.2 Ah cells had 1 in 10 million failure rate, which was unacceptable to TM.

Also, higher capacity often means lower power density, meaning less amps can be drawn from the cell than with lower capacity design. In automotive application, power density is just as important as energy density. Finding optimum balance between the two for specific application is the key. Also, cycle life declines as capacity increases, unless cell chemistry is redesigned.

Joseph · 12-23-2008, 09:27 AM

"I have heard that 2.2 Ah cells had 1 in 10 million failure rate, which was unacceptable to TM."

That would be one failure per every 1,463 cars.

WarpedOne · 12-23-2008, 10:24 AM

Quote:

I have heard that 2.2 Ah cells had 1 in 10 million failure rate

In what timeframe?

If it is 5 years that this is not such a big problem, if it is one year than it is a big deal.
1 broken battery pack means 20.000 USD of warranty costs + labor cost + potentialy very bad publicity.

I can understand Tesla wishes for at most one broken ESS every few thousands cars. Their biggest and hardest task is to build the image of reliability. TG already said "too bad it doesn't work in real world". They don't want such a reputation.

JRP3 · 12-23-2008, 01:54 PM

A single cell failure wouldn't disable the whole pack as I understand it.
So the consumer would only be notified of a warning light or something needing service. Or maybe I'm wrong

AntronX · 12-23-2008, 01:58 PM

Quote:

Originally Posted by WarpedOne

If it is 5 years that this is not such a big problem, if it is one year than it is a big deal.

By failure i meant thermal runaway, or the cell catching fire. That's the worst thing that can happen to TM's reputation right now. Problem with 1 in 10M cell burn up rate, is that there is no way you can know when or which cell will fail. It can happen in the first batch of cells or in a batch in 5 years, does not matter. Going with lesser capacity, but proven and more reliable and safer cells makes a lot of sense. Plus, the added benefit of longer cycle life will pay off in the long run.

Quote:

I can understand Tesla wishes for at most one broken ESS every few thousands cars. Their biggest and hardest task is to build the image of reliability. TG already said "too bad it doesn't work in real world". They don't want such a reputation.

Nobody would want such reputation and I would not pay too much attention to Clarkson or Top Gear. He is a paid celebrity, a paid actor. He is the last person you want to listen to.

WarpedOne · 12-23-2008, 02:45 PM

JRP3:

Quote:

So the consumer would only be notified of a warning light or something needing service. Or maybe I'm wrong

No, you're right. Well, at least as long as it is a single cell failure. What exectly will happen if more than one cell fail, we'll I'm not sure I know. That water pump running 24/7 speaks to me that Tesla wants to be extra sure about safety.

Quote:

Nobody would want such reputation and I would not pay too much attention to Clarkson or Top Gear. He is a paid celebrity, a paid actor. He is the last person you want to listen to.

Ah, but we agree on this. It is just that, if some roadster somewhere somehow catches fire it will be "Li-Ion batteries explode, we told you so" all over the web. And that Clarkson's stupid remark will be repeated like a sacred mantra.

Tesla's playing safe for now may bring about big interests in the future.

GSP · 12-23-2008, 02:46 PM

It looks like the Roadster is using 2400 mAh cells. From Martin Eberhard's blog, in respone to a 17Dec2008 comment by dpeilow:

Martin sez:

Yes, I don’t know where this 2200 mAh number came from. True, in the early days of Tesla, this was state of the art. But my car is built with 2400 mAh cells, and 2600 mAh cells are the jellybean cells today. As you say, higher-capacity cells are already on the market.

Tesla Founders Blog

Also, R & T listed the capacity of the Roadster's battery at 59 kWh.

http://www.roadandtrack.com/assets/d..._datapanel.pdf

This is the latest data we have on Tesla's pack (more up-to-date than their website!). Do they seem reasonable? Lets do a quick check:

59,000 Wh = (6831 cells)(2.4 Ah/cell)(Y volts)

Y = 3.6 volts per cell. This seems to be a reasonable number.

GSP

AntronX · 12-23-2008, 03:31 PM

Ok, if Martin says they are 2.4 Ah cells, then I am definitely wrong.

doug · 01-02-2009, 10:17 PM

Quote:

Originally Posted by doug

The guy from Asahi Shimbun asked me for my best guess as which supplier Tesla uses...
I said likely Sanyo or Matsu [Matsushita, aka Pannasonic], but ultimately came down on the side of Panny for no particular reason.

Looks like their both the same company now.
Panasonic and Sanyo Aim to Merge by End of Year - NYTimes.com
Newsvine - Panasonic-Sanyo merger FINALLY goes through

Joseph · 01-03-2009, 11:28 AM

If the amph if the batteries they are using has risen, why hasn't range also?

Are they going to be using higher capacity batteries every year?

And on the Panasonic/Sanyo merge:

Scary. Honda and Ford and Toyota now source all their hybrid batteries from ONE company. Looks like Panasonic nearly has a monopoly. (GM is using cobasys or something.)