New Fascinating Details Emerge On Rivian Battery Pack Design


Is Someone Finally out Tesla’ing Tesla?

Teslarati’s Christian Prenzler was lucky enough to get a tour of the Rivian battery lab (ref). The tour uncovered some interesting additional technical details of Rivian’s battery pack design.

Here are the details:

As we reported in an earlier article the pack is indeed a 2170 double stack. A 7 mm flat cooling plate is sandwiched in between the 2 layers of cells.

Rivian’s solution to battery thermal management is the use of a cold plate that’s placed between two battery cells. A single cooling system chills both layers of cells at the same time. According to Rivian, this reduces the amount of energy needed to power the system, thereby allowing the car to have better range in all types of conditions. In addition to saving power, the cooling system’s design allows for tighter packaging of cells within the modules. According to Farquhar, Rivian’s unique packaging allows the module to be 25% denser than any other battery module on the market. 

 The different kWh pack sizes are made by including different numbers of modules. Each module has 864, 2170 cells (432 cells in each layer).

Although not verified by Rivian, customs import records indicate LG Chem is the manufacturer.

Assuming Rivian’s 2170 LG cells have the same energy density as Tesla’s, then each cell in Rivian’s 864, 2170 cells has a usable energy of .0177 kWh/cell (78.3 kwh/4416 cells). That pegs 1 Rivian Module at 15.3 kWh. Rivian’s packs are 135 kWh and 180 kWh, which would be equivalent to 3, 15 kWh modules (180-135), so the numbers make sense. LG makes a production cell INR 18650 MJ1 that has equivalent energy density to Panasonic’s NCR 18650B of 255 wh/kg. So, it is entirely possible that the LG cells Rivian is using are just as energy dense as Tesla/Panasonic’s.

Rivian will pre-cool the pack on the way to the charging station in order to optimize charging time.

This pre-cooling is very similar to what Tesla is doing with Track Mode. In Track Mode all the fans and cooling system come on and pre-cool the pack prior to track time. Tesla is using the thermal mass of the pack to store “cold” prior to track time. It looks like Rivian is doing the same thing with their pack for fast charging. I’ll bet money that Porsche is using the exact same technique in the Taycan prior to their high power charging. Also, Tesla may already be doing this according to Bjorn Nyland, but in order to make it work, you must have the navigation system set to take you to the next Supercharger so the car knows it will be Supercharging soon and it can prepare (ref).

The pack uses a carbon composite shell and a “ballistic shield” to protect the battery pack in case of accident or foreign object penetration.

If you are into pack design, these details are fascinating news indeed. They validate our earlier analysis showing that stacking batteries results in a more energy dense pack.

The flat plate design could result in a pack that is more energy dense than the Tesla Model 3 cooling snake design. Why? Elimination of Tesla’s cooling snakes should give tighter packing of the cells.

The flat plate kills two birds with one stone. It provides structural support for the cell layers AND cools the battery.

I know Keith and I have published an analysis that says Tesla’s cooling snake design transfers heat better than any flay plate design, but Porsche/Audi, Rivian, GM and Jaguar all use flat plates. These new LG cells must be able to handle the extra 8 degrees C higher operating temperature they will encounter with the flat plate design. Pre-cooling also helps the flat plate pack design survive by lowering temps during the charging session. In addition, the flat plat design is easier to assemble than Tesla’s Model 3 cooling snake set up and should be lower cost.

Definitely looks like a great design. I wouldn’t be surprised if Tesla uses this same approach on the semi truck pack.

Categories: Battery Tech, Rivian, Tesla

Tags: ,

Leave a Reply

67 Comments on "New Fascinating Details Emerge On Rivian Battery Pack Design"

newest oldest most voted

Of course, everything depends on the truck knowing it is going to be charged and the driver doesn’t forget to tell it. i’d think this would be of limited usefulness since the battery cannot be TOO cold to fast charge it.

@ Bill Howland,

“i’d think this would be of limited usefulness since the battery cannot be TOO cold to fast charge it.”

Yes we know the battery has an optimum temp band for fast charging and it can’t be too cold. But it looks like Tesla is using pre cooling for track mode, and for fast charging.

and now Rivian is using it as well. So it must be of some value.

You just prepare for the charging session by pulling the battery temp down to the bottom end of the band or perhaps a degree or 2 below.

The cooling system might not be able to make it too cold, but there is definitely an issue charging batteries that are too cold. Most l-ion cells should not be charged above 1C if below +5 degrees C (need to heat the pack before increasing charge rate)

More info here.

That sounds like LiFePo chemistry, but NMC can charge at lower temperatures. I don’t know the charging temperature range of NCA, which Tesla uses. Battery University did not distinguish between the cathode chemistries in that article you linked.

As I recall, no li-ion battery chemistry has any reduction in charging speed/efficiency charging unless it’s below 40°. Presumably no EV maker designs the battery pack cooling system to cool down the pack colder than that. At least, I can’t imagine why anyone would.

So… Rivian now uses the same cooling system as legacy carmaker EVs have for years, but for now it’s “fascinating”? Rather pathetic.
That said, we know that flat plates are cheaper but less effective for cooling, so Rivian won’t be able to compete with Tesla in terms of charging current over longer periods. It’s a trade-off and that’s fine for a utility vehicle, but def not “fascinating”.

I agree, there is quite a bit of speculation here that is presented as fact, when by a logical analysis it doesn’t even make sense.

Really, we’re supposed to believe that Tesla’s engineers couldn’t figure out that a flat plate was just as effective as a snaky ribbon, yet far easier and cheaper to put into the pack?

Nah, I don’t buy it at all. Tesla’s more complex, more difficult pack construction must give it better cooling, or they wouldn’t go to all that trouble and expense.

I also don’t buy that it’s “entirely possible that the LG cells Rivian is using are just as energy dense as Tesla/Panasonic’s.” If Rivian is using cells with lower energy density, then that would explain how they can get away with using what is almost certainly a less effective cooling system.

And if I’m correct, that also brings into serious question the assertion in this article that “Rivian’s unique packaging allows the module to be 25% denser than any other battery module on the market.” That assumption seems to take as a premise that the cells Rivian is using have the same high energy density as Panasonic/Tesla 2170 cells, which again appears rather unlikely.

In fact, as I think more about it, there’s another logical fallacy here: If Rivian’s design is even more energy dense than Tesla’s, that means they will have an even bigger problem with heat building up in the interior of the pack. The last thing they’d want to do is use a cooling system that’s less effective!

No, all the evidence that I see here points to Rivian’s battery pack using cells with significantly less energy density than the Tesla/Panasonic cells.

I talked to their battery guy at the LA auto show and he said that the cells were 2170 cells and claimed that they were available from multiple manufacturers. That sounded odd to me but perhaps they are generic equivalents of the Panasonic cells.

Clearly the cooling must be less good than the Tesla method since it requires heat in the cells to be conducted along the length of the cell rather than across the diameter. Perhaps it is good enough which is, after all, what it needs to be.

I don’t see how they do it better than Tesla and the outside cells are going to have far worse cooling than the interior ones, causing balancing, thermal runaway problems.
A cooling system must not just cool or heat, but do it equally to all cells, keeping them all the same temp.
Not that I like Tesla’s method either as they are designing new modules to solve assembly problems of the short range pack.
Metal plate heat exchangers used like this should be shorter and use in flat cells, not round ones with lots of cells.
Any EV company not making their own cells need to be able to switch to various producers if they want the best price and insurance if a cell turns out bad, can easily switch.
But using 2170s you can even switch to flat cells in the same space if needed. I believe flat cells will win out in the end as easier to produce, cool, pack and many fewer cells needed.

Oh come on PP…

Energy density and heat buildup has no correlation. Power density and heat buildup has. But that’s why the plate is in between the two layers, where the temperature is the highest.

I also don’t know why you‘d think LG and is worse at building cells, than Panasonic. I guess Rivian could have also gone to Panasonic, if they were much better.

I‘d say the reason why Tesla and Panasonic work so close is that those two have history and some mutual trust. But I don’t think Panasonic can build much better cells, than LG. Of course energy density is just one side of the coin, but if LG wants, they can probably match Panasonic. And if the packaging is better, then the pack will be more dense.

Power is merely the rate of energy transfer. “More power” simply means more energy in a given amount of time. It’s true that power density and energy density in batteries are two different (altho related) characteristics, and it’s possible to increase/decrease one to some degree without increasing/decreasing the other. But that’s rather irrelevant to the ability of a cooling system to transfer waste heat out of the battery pack, which is what we’re discussing here. Nobody makes Panasonic’s “2170” cells but Panasonic. They are made to Tesla’s exact specification at Gigafactory 1, and Panasonic doesn’t sell them to anybody else. Other battery makers do make form factor 21700 cells — that form factor already existed before Panasonic started using that form factor for cells made at Gigafactory 1 — but other manufacturers don’t call them “2170” cells. It’s not that I think Panasonic is better at making cells than LG Chem; it’s that Tesla uses cells with a higher energy density than other EV makers. That creates some problems which other EV makers aren’t willing to deal with, and one of those problems is needing a cooling system which can transfer heat energy out of the battery pack faster. But… Read more »

“Oh come on PP…

Energy density and heat buildup has no correlation. Power density and heat buildup has. ”

This is why I suggested in the article that flat plate would be good in the Tesla truck. Low power density. The truck pack only runs at a C rate of about 1.

Makes sense with the Rivian’s 105kWh, 135kWh and 180kWh battery options. Assuming the same voltage as the Tesla battery, my understanding is that higher AH would mean a higher discharge power at a lower burst C, so lower heat. The cooling wouldn’t need to be as efficient if less heat is being produced. Same goes for charging. Now… if they wanted faster charging, which a huge battery like this might want…then it may become an issue.

No, wrong. Higher energy density means less thermal mass for a given amount of energy stored. A higher energy density battery pack with the same capacity is smaller. With the same amount of heat generated in a smaller volume plus a smaller thermal mass, there will be a need for more rapid heat transfer to prevent overheating, given the same amount of power. A higher energy density pack is also smaller, meaning less surface area and thus less passive heat radiating away.

Since you’re an engineer, George, I’m quite surprised you need to have any of this explained to you.

As I said, the evidence points to Rivian’s battery pack using cells with a lower energy density than the Tesla Model 3… not a higher energy density.

Down-vote away, that doesn’t alter the facts nor the physics involved.

George plainly says he is talking about power density. You make the sleight-of-hand that the energy density is higher – but he’s not talking about that. You probably don’t understand the difference.

You should really stay out of these discussions since its obvious to the rest of us that they are over your head. I wonder if you still confuse MPG with MPGe, as a for instance.

Panasonic might not have much of an advantage in energy density anymore, but I bet they retain a big advantage in cost for a long time. First mover advantage blah blah.

If you read George’s article you are replying to above he states in there that the Rivian design is more energy dense than Tesla, but will run hotter as the cooling of the flat pack is not as good. Engineering tradeoffs. So yes, that is exactly what they are doing.

It’s only an engineering tradeoff if it’s actually true. Read the article a bit more carefully. The speculation belatedly inserted — that Rivian is using a battery chemistry more resistant to heat — is merely an attempt to escape the logical implications of a highly improbable (if not outright impossible) conclusion, which isn’t surprising since it’s a result of unfounded premises and assumptions. As we computer programmers say: Garbage in, garbage out.

It certainly violates the heck out of Occam’s Razor! The simplest explanation is that Rivian — just like pretty much every other EV maker — is using cells with lower energy density than Tesla’s. That gives the battery pack larger thermal mass and more radiating surface, enabling it to use a less efficient cooling system. The only assumption we need make for that scenario is that the Rivian rep was wrong when he claimed that Rivian’s battery pack has a higher pack-level energy density than anybody else’s.

You are clueless here, – no offense – the simplest explanation is that the battery is bigger.. Why can’t you get that through your head? Oh, I forgot: Pushi’s garbage-in leads to Pushi’s Garbage- out.

Higher cobalt content yields higher temp tolerance.

the cell might be less dense which allows for an pack wich is denser because of less cooling

The bigger the battery capacity, the lower the C rate for a given power level. And the lower the C rate, the lower the need for cooling. It might bite them in hot climates, so the charge rate sometimes slows down for people living in the stinking desert? But by the time this thing hits the streets in quantity, there might not be anyone left living in the stinking desert. But I hope they get it to market before the whole planet is dead. I am serious, although this comment took an unintentional turn into dark humor.

The bigger the battery the easier it is got increase overall density but I agree that the only way they can achieve 25% over Model 3 Long Range is having less cooling in the module.

I think some assumptions in the article are a bit doubtful, but my experience with liion batteries tells me all major makers are in the same level of energy density, there’s nothing groundbreaking in Panasonic cells.
Some small companies are able to do battery packs with very high energy density, easily beating any mass produced EV, the problem is price, some more than others but Nissan, Tesla, BMW, … all are constrained by costs.

From a Tesla press release:

“Cells used in Model 3 are the highest energy density cells used in any electric vehicle. We have achieved this by significantly reducing cobalt content per battery pack while increasing nickel content and still maintaining superior thermal stability.”

If Tesla’s marketeers say it, there can be no debate!

A 2017 battery cell teardown and chemical analysis found no difference between the 18650s from a Model S and Panasonic’s off-the-shelf 18650. They may have made slight tweaks since then, but the 2170s seem to have the same density and performance specs as the 18650s. There was no “30% improvement” or any of that crap we keep hearing. Tesla has no super-secret formula.

Tesla is alone in using NCA. Almost everyone uses NMC, even the upcoming 60 kWh Leaf. NCA used to have much higher specific energy, but NMC has caught up.

Point me to this testing please.

It’s possible that Tesla’s claim is not true. But that is indeed what they claim.

The point is that different battery chemistries used in different EVs have different EDs (energy densities). While we may rationally argue whether or not the 2170 cell has higher ED than Panasonic’s 18650 cells, what cannot be rationally disputed is that Tesla uses higher ED in their cells than other EV makers do.

Of course, that quote I cited doesn’t prove that the 2170 cells Tesla is using have a higher ED than the ones Rivian is using. But if you ignore the groundless premises and unsupported conclusions in the article above, and focus instead on the actual facts reported in the article, it seems almost certain that Rivian is using lower ED cells and that their battery pack has a lower pack-level ED than the TM3.

Tesla does not use panasonic cells. Tesla owns the cell chemistry. Panasonic makes the cells on tesla improved machines with Tesla chemistry.

I see I made one error here: the claim “Rivian’s unique packaging allows the module to be 25% denser than any other battery module on the market” comes from Rivian’s Richard Farquhar, Rivian’s vice president of propulsion, and isn’t an assertion by the authors of this article.

“I see I made one error here: the claim “Rivian’s unique packaging allows the module to be 25% denser than any other battery module on the market” comes from Rivian’s Richard Farquhar, Rivian’s vice president of propulsion, and isn’t an assertion by the authors of this article.

right. Farquar made the claim it is more energy dense and we did an analysis on the Tesla truck that showed the math as to why it is more energy dense. There definitely seems to be a benefit to stacking batteries in the energy density department.

@PMPU @M01
“Really, we’re supposed to believe that Tesla’s engineers couldn’t figure out that a flat plate was just as effective as a snaky ribbon, yet far easier and cheaper to put into the pack?”

Nah. I stated very clearly in the article that we know by our analysis that the plate plate design doesn’t cool as well….and of course Tesla engineers know the same thing.

“I know Keith and I have published an analysis that says Tesla’s cooling snake design transfers heat better than any flay plate design, but Porsche/Audi, Rivian, GM and Jaguar all use flat plates. These new LG cells must be able to handle the extra 8 degrees C higher operating temperature they will encounter with the flat plate design. Pre-cooling also helps the flat plate pack design survive by lowering temps during the charging session. In addition, the flat plat design is easier to assemble than Tesla’s Model 3 cooling snake set up and should be lower cost.”

Yes, after having painted yourself into a logical corner, you tried to escape it by piling another groundless assumption on top of the ones made previously in the article.

May I point out, respectfully, that Occam’s Razor shaves in a very different direction than the one you’re pointing to.

Let me introduce an illuminating concept – the concept of different sized batteries.

A VERY LARGE battery needs relatively little cooling PER UNIT AREA.

The battery in my Lawn Mower needs more cooling PER UNIT AREA than does the battery in my BOLT for 2 BIG reasons:

1). The battery must charge much faster than the battery in my Bolt.

2). Since it must complete charging sooner, the battery charging rate PER Unit AREA is high.

Now does that mean my lawn mower is more powerful than my 200 hp BOLT ev? Overall, no, but per unit area it is.

Does my lawnmower require greater charging facilities OVER THE ENTIRE BATTERY than my Bolt ev – No it doesn’t.

But then you don’t understand the concept of per unit area since you’ve already proved you have no idea how Atmospheric Pressure works in the real world since you think someone can seal a hole in a space ship by placing his FAT CAN over the hole. You learned this, you said, by reading LUNA SPACE NOVELS.

hint: Fiction tends to mean that they are not true.

Just to prove that LUNA crap is nonsense, a 6″ round hole in a space ship, would place a force on your cement ARSE of 416 pounds. That pressure will cause the skin to rip open.

It’s really hard for you to conceive that possibly Tesla isn’t the best at everything, isn’t it? IIRC my Li-ion battery history, it was the Koreans that were ahead in this technology and the Japanese that had to catch up, so why couldn’t LG’s batteries be equal to, or even better than Panasonic’s? Why is this impossible in your world?

It’s really hard for a serial Tesla basher like you to admit that Tesla really is the best at anything, isn’t it?

* * * * *

I’m happy to stand on the arguments I’ve made in my comments. I haven’t jumped to any conclusions, and I’ve tried to follow good scientific principles by observing Occam’s Razor — that is, the simplest explanation is the one most likely to be true. The same can’t be said for the arguments or conclusions presented in the article above.

Please don’t insult Tesla by thinking anything they do is in any relation to PUSHI.

Wouldn’t it be ironic if the new Model 3 battery pack design
For the SR battery, which is “lighter and easier to make” used a flat plate cooling design?

It certainly would be surprising! If Tesla is able to switch to flat plate cooling, then simple logic suggests they would have found another way to improve cooling efficiency; an advantage to make up for the disadvantages of using flat plate cooling.

If Porsche uses it for their 350 kW charging, I guess the flat plate design might be sufficient.

If Rivian uses similar cells, then the problem will rather be finding a charger that can handle enough power. With their biggest pack accepting more than 500 kWh (assuming Porsche can do 300 kW at 100 kWh in a production ready car).

So adjectives are “pathetic” now? Seems a bit fussy, no? Mr Spock thought that just about everything was “fascinating”. But maybe you are an English major? (In which case, Mr Spock was a science fiction character.)

Also theres likely one ends temp is different then the other.

This seems like it won’t cool the cells very well compared to the Tesla snake design. Having a cooling plate with the smallest surface of each cell in contact with the cooling plate is not good.

It’s also where one of the electrical connections are, so doubly confusing.

Agree Joe. This was my first thought. All the weight of the upper cells on an aluminum plate, even with a thin thermal conductive/electrically insulating film, pressing down on the lower cells, doesn’t seem right. I’ll be looking forward to a Weber Auto teardown to see the design.

Actually you can connect cylindrical cells on one side, so does Tesla.
Nevertheless, cooling is not good with Rivian‘s design.

No, the cells are oriented butt to butt as it were. Electrical connections are at the top and bottom of the pack.

Yes. Furthermore, having cooling only at one end puts the cooling effect as far as possible from the center of the cell’s mass. Tesla’s design puts the cooling effect significantly closer to the center, where heat transfer (and thus cooling) will be the most effective.

I know from reading about the engineering of the Volt’s battery pack that it’s important to limit the gradient of temperature across the pack. Presumably (that is, logically) that must also apply to the temperature gradient across the cell. Having the cooling effect only at the end of a cylindrical cell which is much taller than it is wide is, again, a substantially inferior design as compared to Tesla’s design.

Are you sure it is just batteries and there are no fins or other forms of cooling coming from the cooling plate? Also the batteries have a metal shell that if soldered to the cooling plate for good heat conduction that if the batteries have no plastic covering then the metal shell acts as cooling along the sides of the cells.

Just placing the cells against the cooling plate will not work well, but if the cells are thermally bonded to the cooling plate the plate will cool more than just the bottom of the cell.

Well of course I’m not sure. I just pointed out that some of the reasoning in this article appears to be faulty, and in some cases appears to be contrary to the facts presented.

This whole article is mostly just speculation. We’re most definitely arguing in the absence of facts.

But the claimed advantage of the flat plate design is that it’s very simple to construct. Once you start adding fins or sleeves that fit around each individual cell, you can no longer reasonably claim it’s a simple and easy-to-build design.

True, I have been trying to figure how to improve cooling for a flat plate design myself for my own power pack. I think I have a solution that is not expensive to make … BUT it is very precise in the placement of the copper fins that improve cooling.

On a DIY design I think it will work great, but for mass-production I don’t think it will work cheap/fast enough. So far the only simple cheap solutions I see is:

1) Use a low temperature melting soldering of the cells directly to the cooling plate, and depend of the metal case of the cells to transfer the heat.

2) Use a heat-sink compound instead between the cooling plate and the bottom of the cells and again depend on the metal wall of the cell to help with the cooling, ground is taken from the side of the cell instead of the bottom.

3) Find a liquid that will not interfere with the operation on the cells and just submerge in the cooling liquid. (PS. I have not found one yet that I would trust for years of operation.

I think my next build will use #2 to test that idea.

The Volt’s cells are flat plates, about 10″ x 11″ in size; two sandwiched together with the cooling plate in between. Those cells have a substantially higher C rate and likely get pretty hot. My guess is the more compact form factor of the 2170 and lower discharge rates means they don’t get as hot. Cooling the base of the 2170s, with ventilation at the top for heat to escape may allow the lower temps to filter up through the battery.

The Volt pack weight is much heavier than the Tesla pack weight on a per kWh basis.

It wouldn’t at all surprise me if the Volt’s battery pack has a higher gravimetric energy density (that is, more kWh per pound) than a Tesla pack. After all, all Tesla packs use cylindrical cells, which by their basic geometry forces there to be some space between the cells. As you say, the Volt (and Bolt EV) uses flat (pouch) cells, glued into a rigid frame, with a cooling plate between each pair.

One of George Bower’s first articles here at InsideEVs, if I recall correctly, was a comparison of the Volt and the Model S battery pack cooling systems. Unlike this article, that was based on pretty well established facts, and I presume it was mostly or entirely correct.

Rather different than the current article, which is mostly speculation and has some assumptions and conclusions which are, to say the least, questionable.

Keep in mind the snake design is also not ideal. It’s a series cooling loop that will largely only have a point contact between the ribbon and cell (limited surface contact). The Rivian design will have a higher surface contact and with a parallel cooling design will result in more even cooling from one cell to another. It would be interesting to see how well the casing of the cell vehiculates heat. If the casing moves heat well then having cooling at the bottom/top might be inconsequential.

No, not a point or even a line contact. The “snake ribbon” cooling loop in a Tesla battery pack winds partially around every individual cell, giving a three-dimensional area of contact. Your assertion that the Rivian design will have higher surface area contact with each cell is at best jumping to a conclusion, and is likely wrong.

comment image

What you’ve just illustrated is that Tesla’s system can adsorb heat into the Snaking Ribbon at only a small minority of the Surface Area of the individual cell by Conduction. At the point of inflection of the Snake, the contact pressure will be the highest and the heat transfer at this point will be, by Conduction, the most efficient. But the rest of the cell can only be cooled by Radiation, or Convection into the ribbon, whereas GM’s flat plate system covers the vast majority of the pouch area. In
Tesla’s system therefore, to get to the Ribbon by CONDUCTION requires that the heat at the most distant part of the cell must travel through other heat generating material of the cell, or to a lessor extent, through the cell casing by this admittedly round-about method. (point 203, as a for instance).

Now, is that any disadvantage for Tesla? No, since as I keep repeating, the heat generated per cell is quite low, hence the cooling requirement per cell is low, and Tesla’s system seems to work just fine, Tesla’s warnings not to continually fast-charge not withstanding.

Bring it to mass production before Tesla and I will be impressed. Bildung a prototype with a coming new design you have red about, doesn’t make you a first mover. Mass production it does.

But I cheer to any progress in the EV space, so Cheers!

Not fascinating.
Imaging a cell that has a real problem and goes 🥵 , the problem might occur at any location within that cell. Imagining all possible locations I would take a snake design anytime over the one sheet between the 2 layers anytime.

Fascinating? Interesting maybe but not fascinating

Spin said:
“Fascinating? Interesting maybe but not fascinating”

LOL. In the original copy it was merely interesting. However during editing it must have gotten upgraded from interesting to fascinating


I will be really curious to see if these other companies battery packs can stand up in the long run
Or if the battery packs over heat at some point because they don’t have as good of a heat management system as Tesla’s

Do you know the temperature of the cooling plate? If it is really low, it can on average maintain a good thermal factor for the batteries.

You are right, except the plate can not be too cool. Li-ion cells have problems if they are too cold. It would be a bad thing if you try to operate with too cold a plate as it could damage the cells. On the other hand, how much coolant they push thru that cooling plate can make a big difference to without lowering the plate’s temperature.

Isn’t that just a variation of what Chevy does for the Bolt battery pack?

GM’s battery pack cooling system, used in the Volt and the Bolt EV, puts a flat cooling plate between each pair of flat-sided pouch cells. That’s a much, much greater surface area used to cool each cell, and it uses many small cooling plates instead of one large one.

But one could argue the engineering principle is the same, so to a certain extent, you’re right.