Amazing Core Tesla Battery IP - 18650 Cell

[ItsNotAboutTheMoney]

I thought this was no longer true. The original marketing for the Volt was a series hybrid to differentiate them, but when it actua came to market, the power train was a blended hybrid. It can run in series mode some of the time, but they couldn't put a large enough battery and motor to cover the upper end of the performance chart, so it also has to couple the gasoline engine output directly to the wheels. Which makes it even more of a compromised and complicated design.

Correct, the key differentiator between GM's design and theToyota/Ford design is that the major gearis on the eldctric motor instead of the engine, which is better for an "EREV" design.

I believe that GM's Chevy-first approach was right. As a Cadillac it would be underwhelming and inaccessible. The ELR will just grab a few sales and ZEV credits and bide time until the Volt Gen 2 platform switch and other Voltec models which, not coincidentally, will be using US-made cells as LG Chem have basically accepted taking a short-term loss and almost certainly US-made motors that GM is building in Baltimore, initially for the Spark EV.

High performance Voltec would require a larger, less efficient engine and either a larger battery or the engine coming on under load. Either way it ups the cost and lowers the overall efficiency rating. The aim was mass market, trying to get a jump on Toyota. They over estimated sales like everybody else, but now they have a product I want to buy.

 

[JohnQ]

I'd take anything Bob Lutz says with a large grain of salt. I've seen him on Bill Maher and the guy is dumb as a post. Not to say he didn't have a distinguished career compared to the other auto exec nimrods who ran the American car industry into the ground.

I agree with half your statement here ... grain of salt. Dumb as a post is a little extreme. He understands design and has been quite good at it. From an engineering standpoint I think he's a little light. That coupled with being a product of the 70s-90s automotive establishment gives him prejudices that are based on incomplete knowledge and an older paradigm. It's a weakness that results in inaccurate statements. He's not dumb, he just has blind spots.

 

[CapitalistOppressor]

I should have included this patent in the list of "core" patents at the cell level -

Battery Cap Assembly with High Efficiency Vent - Patent application

I had already identified that the caps were made of aluminum, based on the core design patent. In addition to facilitating ultrasonic welding, lower weight and the other benefits discussed in that patent, the aluminum cap also has a lower melting point than the sidewalls of the battery. So if the intumescent materials used in and around the battery don't manage to suck enough heat away from the battery to keep it from exploding, the aluminum cap will melt and allow the contents to be ejected out of the top of the battery.

The core design patent just mentioned scoring the caps to allow vents to open, but those methods are just defensive, while this patent is probably the core venting method used. BTW, as usual this patent claims to cover the broad concept, not just this specific implementation, just as the rest of the Tesla IP does.

So the sidewalls will be cool because of the cooling effect of the goo reaction, so shouldn't melt. If the battery goes critical anyways, it will melt the aluminum before the steel regardless, and spew out of the top. If the battery somehow overheats the supercooling first layer of goo and bursts anyways the, charring second layer of goo will turn into a hard char cylinder and direct thermal energy vertically. Again, seems bulletproof.

Here is the wikipedia link for the intumescent goo -

Intumescent - Wikipedia, the free encyclopedia

I wonder if every manufacturer out there has IP focused on containing/preventing thermal runaway within the battery, as opposed to preventing chain reactions between the batteries. Even Boeing failed at preventing a chain reaction, and they had only a few batteries. Their strategy seemed designed to contain the reaction, instead of stopping it, or redirecting it.

In contrast, Tesla removed all safety devices that are designed to contain/prevent thermal runaway, added what might be a novel use of intumescent's (which are widely used in fireproofing) in the battery internals, as well as the external environment, all of which is designed to stop a runaway reaction after it has started, and then channel it OUT of the battery (and away from other batteries) if that fails.

Boeing utterly failed at solving this. GM also failed, though not in as spectacular of a fashion. Looking at the Tesla system, I find it difficult to believe you could get anything other than small isolated failures unless the external controllers went nuts and ordered a large fraction of the batteries to self destruct (which might put enough heat into the pack to overcome the rest of the obstacles).

 

[Doug_G]

Patenting an otherwise non-functional design might be a literal bar for anyone attempting something similar, beyond the air cooled Roadster level

The Roadster pack is liquid cooled, yes?

 

[dsm363]

Yeah. The Roadster pack is liquid cooled.

 

[Mycroft]

I LOVE Tesla's "amazing, magical drive train"! LOVE IT!!!!

 

[rcc]

[...]

I wonder if every manufacturer out there has IP focused on containing/preventing thermal runaway within the battery, as opposed to preventing chain reactions between the batteries. Even Boeing failed at preventing a chain reaction, and they had only a few batteries. Their strategy seemed designed to contain the reaction, instead of stopping it, or redirecting it.

In contrast, Tesla removed all safety devices that are designed to contain/prevent thermal runaway, added what might be a novel use of intumescent's (which are widely used in fireproofing) in the battery internals, as well as the external environment, all of which is designed to stop a runaway reaction after it has started, and then channel it OUT of the battery (and away from other batteries) if that fails.

Boeing utterly failed at solving this. GM also failed, though not in as spectacular of a fashion. Looking at the Tesla system, I find it difficult to believe you could get anything other than small isolated failures unless the external controllers went nuts and ordered a large fraction of the batteries to self destruct (which might put enough heat into the pack to overcome the rest of the obstacles).

Yes, I think everyone else is going for "magic chemistry" that is well behaved and doesn't melt down. This shows in that every other major player uses a small number of big cells in their battery pack.

Whereas Tesla assumes that the batteries will fail and coping with blown cells is a fundamental part of their engineering design. If you go with that design philosophy, you want a battery pack with a large number of small cells. The large number of cells used raises the odds that a battery pack will see a blown battery but reduces the impact of losing a cell. And the smaller cells make it easier to contain thermal runaway because the amount of surface area for cooling is greater. The flip side of this is that it makes the pack bigger (more volume in the pack dedicated to cooling) and more complex.

So a Tesla battery pack is more complex to design and manufacture than everyone else's, uses a fundamentally different design philosophy and is part of an integrated system instead of being a module that meets certain straightforward specifications that can be replaced by another module that meets the same specs.

Bob Lutz has no idea of what he's talking about. The amount of design and validation effort that went into Tesla's core technology is huge.

 

[CapitalistOppressor]

Yeah. The Roadster pack is liquid cooled.

Ooops, my bad. :smile:

Let's rephrase that to mean stock batteries, or batteries which are not dramatically simplified the way Tesla has done. They can use internal fuses like stock batteries (or other gizmo's), but Tesla has IP on just about any way to armor batteries, either individually, or in the context of a pack, in an effort to keep thermal runaway from cascading through the pack.

This includes methods to armor interior zones of the battery, the external case, and the spaces between batteries. Not sure how the IP would hold up under a legal assault, but it exists.

 

[William13]

CO, I thank you for you interesting discussion. Many of your ideas seem to be correct. I wonder how many of the patents are disinformation or protective of IP but not actually used by Tesla?

 

[Dave EV]

I thought this was no longer true. The original marketing for the Volt was a series hybrid to differentiate them, but when it actually came to market, the power train was a blended hybrid. It can run in series mode some of the time, but they couldn't put a large enough battery and motor to cover the upper end of the performance chart, so it also has to couple the gasoline engine output directly to the wheels. Which makes it even more of a compromised and complicated design.

The Volt never runs the gas engine until the battery is either drained or the driver engages mountain mode or EV hold.

All other times the battery/motor generate 100% of the propulsion.

When running in hybrid mode (after the batteries are depleted) it can directly couple some amount of gas engine power to the wheels, but this is done to improve efficiency, not power.

 

[hcsharp]

Ooops, my bad. :smile:

Let's rephrase that to mean stock batteries, or batteries which are not dramatically simplified the way Tesla has done. They can use internal fuses like stock batteries (or other gizmo's), but Tesla has IP on just about any way to armor batteries, either individually, or in the context of a pack, in an effort to keep thermal runaway from cascading through the pack.

This includes methods to armor interior zones of the battery, the external case, and the spaces between batteries. Not sure how the IP would hold up under a legal assault, but it exists.

Not only does the Roadster use liquid cooling, but the ESS is very similar to the MS in the overall design of the pack. The Roadster cells are different chemistry and stock format as you call it but you haven't discussed much about the compartmentalization and electrical management using bricks and sheets. This part of the design is very important to the overall safety of the system, and very similar between the Roadster and MS. While developing the early brick design they spent a lot of time determining how to heat/cool all the cells evenly, which is critical for efficiency, longevity, and safety. That design has not changed much with the MS.

 

[Nickjhowe]

@CapitalistOpressor - great work! Well written and easy to follow. Appreciate the time you put into this.

 

[CapitalistOppressor]

CO, I thank you for you interesting discussion. Many of your ideas seem to be correct. I wonder how many of the patents are disinformation or protective of IP but not actually used by Tesla?

The vast majority. Even on patents they are likely using, they often include many alternate methods that might never be used.

 

[jerry33]

Whereas Tesla assumes that the batteries will fail and coping with blown cells is a fundamental part of their engineering design. If you go with that design philosophy, you want a battery pack with a large number of small cells. The large number of cells used raises the odds that a battery pack will see a blown battery but reduces the impact of losing a cell. And the smaller cells make it easier to contain thermal runaway because the amount of surface area for cooling is greater. The flip side of this is that it makes the pack bigger (more volume in the pack dedicated to cooling) and more complex.

In the early 1900s, Rolls-Royce showed that a large number of small bolts held the wheels on better than a small number of large bolts. It's not too surprising that batteries work the same way (although perhaps it's not obvious).

 

[CapitalistOppressor]

Not only does the Roadster use liquid cooling, but the ESS is very similar to the MS in the overall design of the pack. The Roadster cells are different chemistry and stock format as you call it but you haven't discussed much about the compartmentalization and electrical management using bricks and sheets. This part of the design is very important to the overall safety of the system, and very similar between the Roadster and MS. While developing the early brick design they spent a lot of time determining how to heat/cool all the cells evenly, which is critical for efficiency, longevity, and safety. That design has not changed much with the MS.

Yes, I didn't find any IP for the Roadster pack. I suspect it all came from the company that preceded Tesla (the previous company had a prototype vehicle and pack that was shopped to Elon and the other founders, and after a series of business moves, they all coalesced into Tesla). And the TMC discussions on the Roadster technology predate my join date in mid 2012, so I only have a passing familiarity with the Roadster pack.

As to the Model S, most of the IP I am finding for the pack itself seems to be defensive IP listing all kinds of ways to armor it against thermal runaway, with many of the ideas obviously having been replaced by the intumescent goo. There are a large number of patents to go through, but the only thing that strikes me as important so far is the central fusing and control system.

Tesla specifically claims that it is far simpler, less expensive to manufacture, and more reliable. I imagine they are comparing those features to the Roadster pack. I would not be surprised if much of the rest is very similar to the old pack. The Roadster pack has been very reliable and successful.

If so, then the big improvement is at the cell level and thermal management, with the addition of a central controller/fusing system to keep the batteries from exploding, after having removed all of the safety features from the individual cells. If you read the the patent on removing the protective casing, a fair part of the improvement they are listing (beyond the thermal effects I described) are the improved ability to properly anchor the cell (probably with an adhesive) inside of the pack. With a conventional cell, if you try to anchor it, you are actually anchoring the protective cover, since that's all you can really access.

But in mechanical terms, the cover is just a plastic film that is shrink wrapped around the cell. As a result, the cells would tend to shift around during thermal events, and make it more difficult to keep them away from other cells. Deleting the cover gives you access to the steel case, which conducts heat far better, and allows you to firmly anchor the cell.

 

[blakegallagher]

I keep hearing other manufactures talk about how easy it would be for them to get into EV's when they want to .... I did not believe the "idle" talk but once again I am impressed and grateful for this thread CO. Very intresting how Elon and company always seem to take things down to absolute basics and make them behave in ways that others have not thought about.

 

[pharma5]

And the TMC discussions on the Roadster technology predate my join date in mid 2012, so I only have a passing familiarity with the Roadster pack.

Tesla put out a white paper on the Roadster pack - just wanted to quickly share:
http://webarchive.teslamotors.com/display_data/TeslaRoadsterBatterySystem.pdf

 

[Kleist]

@CapitalistOpressor - exciting work - thanks. Make things much more clear.

If you know mfg any core is always cheap to make, cost goes up dramatically with the add-on. Smart move because now he controls the cost of the system and can make improvements in house. I don't think he pays more then 50 cents for the core.
Also interesting big companies would buy components and stop there ( whining about cost to high ), Elon takes an everyday item and strips it to the core what can we do less expensive. (I am working in a big company taking things apart and use only the essential stuff and do the rest ourselves would never fly with management).

Very exciting insight !

 

[hcsharp]

Yes, I didn't find any IP for the Roadster pack. I suspect it all came from the company that preceded Tesla (the previous company had a prototype vehicle and pack that was shopped to Elon and the other founders, and after a series of business moves, they all coalesced into Tesla). And the TMC discussions on the Roadster technology predate my join date in mid 2012, so I only have a passing familiarity with the Roadster pack.

That's not quite correct. Tesla has filed several patents on the Roadster ESS. By "the company that preceded Tesla" I assume you're talking about AC Propulsion? Tesla developed their own ESS early on and only briefly contracted to use ACP's inverter and drive system until they realized they could do a much better job themselves. Not sure why you're not finding their earlier patents.

I'm thoroughly enjoying the Model S research you're doing. Thanks for sharing it! What I would give to be able to talk freely with one of their top battery engineers...

 

[TEG]

I don't think the ACP pack was liquid cooled...

But Tesla engineered and patented a new way of liquid cooling 18650 packs:
BATTERY PACK THERMAL MANAGEMENT SYSTEM - Patent application