GM Versus Tesla: Bolt EV And Model 3 Battery Packs Compared



It turns out the Chevrolet Bolt EV battery pack is simpler to assemble than that of the Tesla Model 3, while the Tesla pack is more energy dense by weight.

We had an excellent video here recently on InsideEVs: “Watch Chevy Bolt Battery Reassembly.” I was impressed with the ease that the Bolt EV battery pack went together. John Kelly was very thorough, even measuring the weight of two of the Bolt EV modules so we can compare those with Tesla’s battery weight as well.



GM’s pack is simpler to assemble in this author’s opinion. However, as expected, Tesla has the edge in the energy density department.


First, a little background on how these two packs cool their cells. Tesla’s method involves running a ribbon-shaped cooling tube between the cells. They have kept this technique since the Model S and are still using it in Model 3.

Tesla has attempted to automate the assembly of the Model 3 battery pack and, unlike Model S, they are now attaching the individual cells to the cooling ribbon. This procedure is described in more detail in our article: Tesla battery pack modules made like a machine gun. It also appears likely that Tesla is attaching the cells to the cooling ribbon with glue.

Obviously, the whole process is difficult as Tesla has twice delayed the promised “5,000 Model 3’s per week” production bogey.


The Bolt EV pack is much simpler. The individual prismatic cells simply sit on a cooling plate with a thermal mat between the cells and the plate.

Below is a great screenshot from the video showing the plate.

Next is a screenshot of Professor Kelly lowering section 5 modules in place.

Pretty simple and easy to assemble.

.…and another screenshot showing the cooling plate.

Energy Density

At the 41:57 mark in the video we have a screenshot (below) of Professor Kelly weighing modules 1 and 10 just prior to setting them on the cooling plate. We must subtract the weight of the holding fixture to get the total weight of modules 1 and 10.

The Bolt EV has 10 modules: 8 with 30 cells each and 2 with 24 cells. Module 1 and 10 are the bigger modules with 30 cells. We also find 57 kWh of usable energy in the Bolt EV pack thanks to some sleuthing by fellow engineer Jeff Nisewanger in “ Jaguar and Chevy have LG in common.” This gives us 5.95 kWh for one big Bolt EV module, which results in a module energy density of 11.9 #/kWh.

Tesla battery module weights are based on weights published in HSR Motors online catalog (aka Jason Hughes). Usable energy per Jason Hughes is 98.4 kWh for the P100D pack. There are 16 modules in that pack, so we can get the energy density of 1 module of 10.7#/kWh. Energy density of Tesla Model 3 cells assumed equal to P100D cells. See “Tesla Model 3 2170 cells=same energy desity as P100D 18650 cells”

Therefore, Tesla’s module is more energy dense by weight than the Bolt EV module. Not a surprising finding.

However, there is one caveat to all this. We have not included the weight of the case in our calculations and Tesla’s case may be heavier than the Bolt EV case. Why? Because Tesla’s battery chemistry is much more flammable than GM’s and it requires more armor to deflect foreign objects.

So, Tesla may not have as big an edge as these numbers indicate if we were to include the cases in the calculations.

Categories: Chevrolet, Tesla

Tags: , ,

Leave a Reply

124 Comments on "GM Versus Tesla: Bolt EV And Model 3 Battery Packs Compared"

newest oldest most voted

Tesla’s cooling solution is also higher performance as it’s wrapped around the center of the cells, not the top or bottom. Therefore the individual cells are less likely to achieve a dangerous temperature.

This. It would have been good to see an analysis on the effectiveness of each cooling system design.

There is some evidence of real world testing. 2014 SparkEV cooling is pretty much what Bolt is using, and it’s able to get 2.4C charging to 80% even in middle of 100F+ summer. Bolt charging is only 0.8C (or 0.4C at 80%), so their cooling is more than adequate.

By contrast, there’s no Tesla that maintain charge power to 80%. There’s also evidence that Tesla slows down from high speed driving due to heating. This suggests Tesla approach is not as effective in removing heat even at ~2C rate of discharge.

“There’s also evidence that Tesla slows down from high speed driving due to heating.”

I thought I read this had to do with the power electronics overheating, not due to the battery overheating. 2C isn’t very fast of a discharge so I seriously doubt that it would heat the battery that much.

Your statement that the Tesla system will have higher performance may not be true.

The metal sheath that goes around the cell triplets in the Bolt EV is an aluminium heat sink. That heat sink is then in contact with the cooling plate through the thermal pad.

One advantage of the GM design is that the cooling plate will maintain a fairly uniform temperature and therefore maintain a fairly uniform temperature across the pack.

With the Tesla series configuration heat from the cells near the coolant inlet will heat the cooling loop affecting coolant temperature of the cells downstream closer to the outlet. Therefore not maintaining as uniform of battery pack temperature and possibly causing uneven charge – discharge properties and possibly uneven degradation. Tesla likely mitigates this with high coolant flow and a strong chiller but that consumes more energy.

You’re theory of cells closer to the start of the Tesla cooling inlet would be better cooled, would be true if the speed of the cooling in the circuit was very slow. A high flow rate in the cooling circuit would completely cancel that effect.

Oops, missed your last sentence.

“…A high flow rate in the cooling circuit would completely cancel that effect…”

Only if the fluid were traveling faster than the speed of light. A neat trick and would make Tesla worth more than half the GDP of the entire planet.

Look again at the size of those packs.
With high acceleration or fast charging, there is going to be high temperatures in the center. With the cells, the centers of each is much closer to the heat ribbon, and has more exposed area.

You have to be careful comparing the effectiveness of the cooling. Pouch cells like LG uses don’t have metal cases on the like Tesla cells, so the frames LG’s cells are in are effectively like the cases Tesla wraps each of their individual cells in. There might not be a huge difference in cooling/heating capability. Also GM’s packs has hundreds of cells versus thousands in Tesla’s case.

Not to be the case in one situation where the Tesla caught on fire. I think your facts are mixed up because that was proven already.

Agreed, Tesla’s method should allow much better heat transfer. Not only does Tesla’s method give a greater surface area of contact between cell and cooling loop, it also puts the cooling loop closer to the center of mass of the cell.

The comparison in this article is rather inadequate. “Simpler” does not equal better.

Better does not equal production ready. There is a reason why concept cars don’t always look like their production counterparts.

Higher performance is BS. I’m sure the GM is as high as the Tesla battery.

Well lets just wait until the Model 3 battery teardown shall we?

One thing that will be the same, is both M3 and MS use the battery tray for torsional rigidity. To penalize something structural, like the case weight, might be considered in light of its other function. That’s what stood out to me. I’d bet GM does same, to some extent, but may yield less contribution to stiffness.

Can someone confirm “module weight” meaning both cells and BMS?

Yes, according to a video interview with a GM engineer that was posted of an early Bolt unveiling, yes, the Bolt’’s high voltage battey box is an integral strength member of the cst body.

The deeper the box the stronger it will be. I’d hazard a guess that the Bolt battery provides more strength to the car than the Model 3 battery does.
As for the cooling system, the Tesla is using an aluminium microchannel extrusion which is in contact with about 1/4 to 1/3 of the radius of the cell. I understand that the Model 3 has glued the cells to it using a special electrically isolated, thermally conductive adhesive. Really, so long as the battery is kept as close to 25’C as possible, for as long as possible, both batteries will have long service lives.

This is precisely Tesla’s advantage. They are willing to things that are hard.

You don’t get points for doing things the hard way. Engineering is about finding the best solution for a problem. Extra points for simplicity.

+1 for this. The best engineering solutions are often the most simple.

The energy density at beginning of lfe is about 0.5 Kg per KWH. What I am wondering is what happens as the battery ages. For the sake of argument, if the Tesla and Chevy both start with 60 KWH batteries, the Tesla should have a 30 Kg mass advsntage, minus the presumed extra mass of Tesla’s moremarmored battery box.

What happens after a few years as the Batteries age? I ran across a university lecture video regarding battery chemistry. It is hard to judge if there were bias of the lecturer, but the upshot seemed to be that GM and LG Chem are ahead of Tesla and Panasonic in developing the secret sauce chemistry that prevents formation of metal as cells near full charge. If the Chevy battey retains more capacity over time, then the initial Tesla mass advantage could be nullified.

I’ve also wondered if charge balancing thousands versus hundreds of cells makes any difference over time.

What battery is only .5 kg for a kWh? That off by a factor of at least 20.

You are correct, but Jack is saying something very different.

When you have manufactured anything in volume, you quickly discover that complexity kills agility in the build process and reliability of the finished product. Given a trade off of a few % of density vs a vastly simpler design, most engineers would pick the latter. We know from the falcon door fiasco that Tesla prefers the science project route for most of its designs. So, this is not terribly surprising. In the EV world, we all have a vested interest in Tesla succeeding in the long run. I truly hope that their board revamps the process by which design decisions get made and grant authority to someone more experienced in this than an eccentric dot com billionaire.

Therein lies the issue… When you’re trying to get EV’s to the masses, a simpler, less expensive option would be better IMHO.

I mean, the complex design is ok for a $100k car, but I’d think the compromise is ok for a $30k one.


Dear InsideEVs,

Despite the numerous improvements made to the website, from usability on a mobile phone screen, it’s gotten worse. It’d be great if you could work on that.

Yes, worked better before the change on my cell phone.

Also – for me, and an LG-G5 is not exactly a wimp of a Smart Phone, either! Barely usable now!

On Adroid using chrome you can use the simplified view/mobile view option.
Sorted all viewing and sizing issues out on my phone.

Awesome to know!

Yes, it’s has too many intrusive pop ups for the “Motorsports network sliding up every time you scroll down. Including the social media links for twitter constantly getting in the way. By the time the keyboard on my Samsung Galaxy Note slides up,I can’t even type a message in landscape as the whole message box is covered. And in portrait. The sides off the message box are cut off so you can’t even completely see the words you are typing. It is definitely becoming frustrating. Any other Samsung mobile users experiencin the same difficulties?

Exactly right. The social media buttons are completely useless, they can be removed outright. The motorsports popup should be a static footer on each page instead of a popup.

Yes, please kill the social media buttons.

Those issues are being addressed. We are working to make it function properly. We are excited about the better look, increased readability, and new options. But the comments system, and some mobile issues, as well as inconsistency among different platforms is problematic. We anticipated this and it is top priority. Thank you for the input.

We have been dealing with some traffic reporting issues and other inside issues over the last two days. The developers have a long list of bugs, thanks to our awesome community sharing such problems and insight. They are tackling it promptly. Please continue to let us know and even send screenshots to the site email. In a few days, after they’ve tackled all these initial problems, we’ll send out an update post and ask for all of you to inform us what is still not up to par. We’re all in this together and we greatly appreciate you bearing with us and keeping us apprised! Thanks so much to everyone!

Steven, Per “We are excited about the better look, increased readability, and new options.” – as a user for over a year, maybe over 5 years, I don’t find this new website interface one that passes the user experience test.

As all users have mentioned above, it tries to do everything on a Mobile format, with less screen real estate, that it does on a large monitor, thereby crippling its usability on the mobile platform.

At least – in the transition – there should have been some posting of how the new looking interface should be looking – before the switch – giving users a chance to go to a test page in the new format, for a while, while you get feedback!

I use a number of Google Tools, and sometimes find that their updated sites are more difficult than their ‘Classic’ ones, so this is not just your team struggling with these types of issues!

Yes, we are working on the mobile side for this reason. There was a test site, but it wasn’t open to all. The mobile side obviously didn’t get as much attention as the desktop side on the redesign. This is because our last redesign was specifically for mobile only. We didn’t even have a mobile site before the previous redesign. So, the mobile side of things is undergoing tweaks now. Thank you for sharing and understanding.

The featured section, the first article on the main page which is blown up to cover almost everything are two bugs that needs to be squashed.
(yes, I know they aren not real bugs, but just because you actually want them there doesn’t make them good features).

Also the inflated pictures that are first on every story you open. It can be removed for almost every article since it’s pretty useless. If I want a huge picture I’ll click on it, now it’s just in the way.

And the “get updates” can be removed or deleted. If you really want it on every page then move it to the right side where you can see it if you want to., like it is on the front page.

We are working on sizing as it’s different on each browser, device, and operating system. The large pictures at the top of every post came directly from the old design. They are the same size (750 x 563). We didn’t change that. There may be two pictures at the top of the posts that were published during the redesign because the system used both, but going forward, that will not be the case. It will remain like the old site with one picture and then the story. The one new interface design that many like is that every story gives related stories on the right. I’m not seeing the get updates, so I don’t understand that comment. We value everyone’s opinion, and that’s obviously why we’ve been sitting in front of the computer reading every comment and responding for nearly 48 hours. Changes will come and it will be great! Thanks again!

I made it a lot better:

Removed the first article. Moved the newsletter sign up. Made things more compact. Increased the size of how old the article is (this is one of the most important things, make it bigger and with a better font and color to make it readable easily, it’s the only thing I look at when updating the page to see if something new is happening).

The only really interesting thing I want to see on the first page is the articles and how old they are.

I know you are doing your best, and working extremely hard. I will appreciate it once I’m out of my “why make it worse”-sulking period longing for the great that was.

I’m trying to moving into an accepting and constructive criticism phase but I’m sure I will still relapse a bit.

What I meant about the “get updates” on the article pages is this:

I definitely like the newsletter moved down. Thank you! I never saw the get updates. I will send both of these to the developers!

Thanks for the jpgs. I sent both to the devs. Nice work.

This is what it looks like now. Need to scroll down through 3 pages to see the first 4 articles.

I want to be able to read the first 4 articles directly on the front page without scrolling.

I can zoom out to make it fit, but then how old the article is (and the author 😉 ) gets too small. There is plenty of room to make how old the article is a lot bigger and with better color and font no matter if zoomed out or at original size.

Went to (I am trying to figure out if and how annoying the motorsport network bar at the bottom of insideevs when you have scrolled up is.) and wanted to kill myself. It was a sobering page to look at since it reminded me how awesome this sites design still is compared to most.

How about an auto-hide of the “motorsport network”-bar at the bottom of the screen? When you scroll up it shows but then it stays until you scroll down again. I’d like it to disappear after 3-4 seconds (or disappear directly if you scroll up or click anywhere).

I agree with that. The mobile version is definitely worse. I’m not very fond of the desktop version either but it’s workable

Multiple fixes and improvements are being addressed, as with any redesign. Your input is greatly valued. We want it fixed ASAP as well. Thank you much!

Agree 100% with these comments. I spent more time than I should have composing a reply earlier on my phone and the web site ate it. Never again.

IEVs used to be my go to site on my phone when I had a few spare minutes. Not with the new design.

We just did the redesign just prior to the weekend. There are surely some initial bugs on different platforms. We are addressing them as promptly as possible. I wouldn’t say never again, since it will be fixed and work as it’s supposed to soon. Please bear with us, as a redesign like this is never initially trouble-free. We are happy that the site is running and working and looks wonderful. We’re not happy about the bugs either. Thanks for the input!

The big question is still why? Why ruin something that people love? Why break something that is working great?

The best you can hope to achieve is to get to the point that you started at and almost all scenarios ends up being worse than when you started.

Obviously, the goal was not to ruin the site. That would be truly silly. The goal was to update the site. It is a work in progress and we are making great strides. We have received a huge amount of positive comments and are happy with many aspects. Bear with us and keep us informed as we fix those that we aren’t happy with. That’s the goal and that’s our job. We would never make a concerted attempt to make something worse on purpose or hinder something. Our goal is completely the opposite. Thanks for your comments.

I know, I know. Try to ignore the unnecessary and non-constructive comments. Love the work you guys are doing. 🙂

Thanks for the additional pics and ideas. We’ve shared them all with the developers. Kudos Mikael!

It’s a top priority. We plan to have it working as normal as quickly as possible. Thank you!

Yes. The new interface is problematic on mobile. iOS with Chrome.

Get rid of the social media banner at the bottom that is just taking up screen real estate.

Yes, iOS with Chrome is the main issue at the moment.

Nice article. I’d love to see the Tesla Model 3 battery assembly robots on a video someday.

Here’s a direct link to my article that is mentioned:

The battery engineering of Tesla reflects their company’s philosophy – high tech & advanced but overcomplicated for no real reason.

While having a more energy dense battery is definitely an advantage, the complex assembly method, high energy cooling requirements & heavier protection materials to safe guard the more volatile cells feels counter-productive. If the end result is equal weight for the battery packs & not too different cooling performance, then Tesla only wasted its money + effort in achieving what GM did at a lower cost & less complex method.

The reason I say this is from my programming professor. If one can write a 10 line code to do the same thing as an elaborate code that does the same in 15, the guy who wrote the first is the one that’s getting hired. The engineers at Tesla (or perhaps Elon) needs to learn what KISS principle means.

The biggest mistake when talking about complexity is understanding one major difference. Complexity of assembling for humans and complexity of assembling for machines. While GM’s approach is easier for humans to assemble, for machines it makes no difference.

In terms of costs and energy density, Tesla wins hands down. And as for volatility, the less volatile nature was sacrificed on larger cell sizes. So there should be no advantage there. In terms of better cooling, Tesla cars will see lower degradation of battery cells and ability to charge faster due to better cooling.

Complexity matters to robots as well. If a robot has a 10 step process versus a 5 step process that means one robot can be twice as efficient. Also don’t forget the GM’s pack has approximately 300 cells versus Tesla’s 1000’s for the same capacity.

I would respectfully disagree. GM’s prior EV like the Spark EV & Volt both have incredibly slow battery degradation rates that mirrors Tesla. This shows GM’s cooling solution is just as effective as Tesla, but done at an ingeniously more simple manner that is less energy intensive (Tesla’s snake like cooling ribbon requires high fluid velocity to be effective). Hence KISS principle.

The same reason applies to pouch cells – more industries use it outside of electric cars. Take a look at the batteries use on passenger jet liners or on commercial locomotives. All pouch style cells. So what if hybrid cars use them? Sharing components & design saves cost, a idea Tesla seems completely foreign to. KISS!

Also building complexity applies to both humans & machines, as engineers/programmers are the same ones who are teaching machines how to mimic the actions after watching how people assemble it. The fact Tesla rely so heavily on the machinery while discounting the human factor is why the company is in “production hell” despite the promise the Model 3 line is highly automated.

Spark EV and Volt aren’t fast charged.

Cylindrical cells are the majority used cells. And as for sharing with hybrids, Tesla doesn’t make hybrids.

What I meant was what is complex to a human is not always complex to a machine and what is complex to a machine is not always complex to a human. The Tesla battery design may be more difficult for humans, but it is easier for machines.

Tesla hasn’t the team/capital to develop pouch cells, would be my guess. Using form-factor commodity cells, and ending up with more energy dense results is quite and accomplishment. It also balances the risks of a variety of sales results, without having to amortize down R&D over potentially too few cars.

They also put them in something compelling. Batteries are expensive, and heavy. Though they are getting lighter, they never belonged in “econo” looking cars, first. There’s no business case for that, unless you are into labeling electric cars something they’re not and continuing to make money on something else?

It’s not about if Tesla has or hasn’t the capital to develop pouch cells or not. The question is do they need to? I think a lot of people are missing where each automakers priorities lies.

Tesla’s solution is cheaper and more energy dense. Probably enough to secure Tesla a few years of advantage. In the end, the pouch cells will probably win in both cost and energy density, but that might take 10 years or so. By then, you will most likely have a new technology take over be it solid state or metal-air or maybe even super capacitors.

So why are all these automakers bothering with pouch cells? The answer is simple, Hybrids! The automakers see the majority of their lineups with hybrids and pure electric on the side.

Pouch cells are better for hybrids, and even if a new technology comes out like solid state, metal-air, or super capacitor, they don’t particularly offer much of an advantage for hybrids.

This is why VW is spending 25 billion on electric and 100 billion on diesel. All these automakers are thinking hybrids.

I honestly don’t see that Tesla has any real advantage at all compared to GM. They are both basically comparable. Maybe a few differences here and there, but functionally they are the same.

Other than a 3 year advantage on energy density?

And of course functionality is the same, they are both batteries! But Tesla has a superior battery for EV usage over GM.

All else being equal, yes pouch cells are cheaper to make than cylindrical cells with a metal case. But all else is not equal. Pouch cells have to be supported in an EV battery pack by putting them into a frame. Would the assembled pouch cells battery pack be less expensive than using cylindrical cells which can be supported by merely a plastic sheet with dimples at each end? Would the pouch cells be as resistant to unwanted shifting around inside the frames as a metal-cased cylindrical cell?

With the Model 3, Tesla has moved away from using 18650 form factor cells to its own 2170 cell design, while still maintaining the cylindrical cell shape. I think it’s pretty clear that Tesla sees some advantage to the cylindrical cells, or they would have moved to pouch cells with the Model 3.

I’m going to say the reason for staying with the cylindrical cells is down to cost. Tesla when developing the Model 3 was already alarm bells tolling about the financial state in the company. Instead of spending R&D + time on an something Tesla has no experience with (pouch cells) they revert to a similar design as the prior Model S & X.

GM doesn’t have any experience with either, it is all LG. If Tesla thought there was a benefit to pouch cells they would have just asked Panasonic who has more experience with all kinds of batteries than anyone.

Again, it is all a matter of focus. For Tesla, they care about what is best for BEVs right now and short term then plans to jump to next technology long term.

For other automakers, they care about what is best for hybrids in the long term. Because most makers see hybrids, not BEVs as the way to go. This is why Toyota teamed up with Panasonic for prismatic cells as well. Because they care about hybrids.

Cylindrical cells are not that great for hybrids.

Let’s see where degradation falls in to know , Tesla takes great care in their product from the driving force behind their inception, complexity is not a negative and the delays have not been a specific result of Tesla straining from just the cooling strip , they had supplier chosen to perform totally disappear, let’s not forget gm is a complicit company with oil in the damage we all live in on our home and bolt would not exist with out Tesla .

I watched him put the Bolt back together the other day.

I look forward to his tear down of a Model 3.
Showing pictures of patent drawings is not a comparison.

Informative. The Bolt cooling design is defective. They’ll have poor lifetime, I predict.

Just like the Volt right? Lol

Volt cooling is different than Bolt.

If one’s used to Leaf that doesn’t have cooling of any kind or “passive cooling” of eGolf, they’d be paranoid as to think cooling may affect battery. But the control system in active cooled battery will ensure there’s adequate cooling by limiting the power.

When Bolt charges at 0.8C (or 0.5C at 80%), there’s no way temperature will affect it. For charging, there’s zero problem with Bolt. Even in normal driving, I doubt Bolt will see much over 50 kW (0.8C) even on hills.

The Bolt EV battery cooling design is very similar to that of the Spark EV utilizing the bottom cooling tray.

Only 2014 SparkEV uses bottom tray like Bolt. 2015+ use plates between cells like Volt.

Below is a link to pretty well reasoned and detailed presentation regarding battey chemistry. The prsenter suggests more than slightly that GM and LG chem are fairly out font in the trace additives that are needed to prenvent reduction of metallic ions into solid metal – shortening battery life.

I am a Tesla fan because I own some stock shares, however I have to give my manufacturing practicality nod to Chevy Bolt. In this case, I believe GM did the math correctly that ease of manfacture beats out the arguably higher tech Tesla battery. If the LG Chem / GM battery happens to have less degradation over time, GM’s will have been a superior choice. We’ll have to watch for the next few years to see which choice, if either was better.

“…the trace additives that are needed to prenvent reduction of metallic ions into solid metal – shortening battery life.”

Aging does not reduce metallic ions in li-ion batteries to solid metal. Some of the positive and negative ions do combine to form inert salts as the batteries age.

If this error is an indication of the accuracy of the “facts” in the source you point to, then it must be a very poor source of information.

Well, I’m a little over a year in and 30,000kms on the car and the pack in my Bolt is reporting 100% battery health (via FleetCarma module on the OBDii port).

I think Tesla’s insistence on Cylindrical cells has to do with cost. When you can shoot out millions of the same cells from an automated line that can be used in different kind of packs for different applications it has got to be cheaper. It should also make it easier to direct the capacity towards different applications depending on the market conditions and other factors

Given the Bolt has 288 cells and the Model 3 has closer to 3000 cells they would need to be making them at a much higher rate.

After reading all of the comments, I believe most people are missing a critical design point. What is the longest heat diffusion path to the assumed isobar of the cooling loop? To state that more obviously: how far does the heat generated internal to a battery have to travel before it reaches the cold plate? We can assume the thermal design of the cooling loop is properly done so that the entire loop is approximately the same temperature. That is, it’s thermal uptake capacity exceeds the heat generated by the individual batteries. But what of the heat path internal to each battery to get to the cooling loop? We know this is non-zero and that battery internal construction materials are similar: similar chemical coatings on either side of a similar separator material. So we can reasonably assume the internal battery thermal resistance per unit length is identical. Therefore, a longer heat transfer path will allow higher internal battery temperatures even with the same temperature at the cooling loop. The thermal path of a Tesla Model 3 2170 battery is about the diameter of a cell: 21mm or 0.827 inch. The thermal path of a pouch cell looks to be 5x… Read more »

The thermal path of the flat pouch cells is the thin aluminum wall of the battery cassette. Per GM’s battery engineer, each cassette holds two cells. Each cell has its own aluminum wall. The heat is carried down the wall to the heat transfer plate.

Yeah seems where ever you look at the GM design, there is adequate surface area and, as others have said, the glycol coolant is probably universally cool on either system, but it is paralleled in the GM so they all start out pretty cold; the satisfying thing here is there is much surface area on the plate, and plenty of contact with the cooling water.

On the other end of the system (the chiller), I know that on a hot day in august the system removes HUGE amounts of heat when you get out and look quickly after driving 160 miles non-stop. AS my Roadster handled the problem, lhe Bolt EV will favor cooling the battery over passenger comfort should the control system deam it necessary.

Good source! Two packs per aluminum case means each pack large flat side is in close contact with the aluminum case, so the thermal path length should be about half the case thickness. This looks to be on the order of the thickness of his index finger, call it about 0.5″.

An error in my analysis above: I forgot that the 2170 is an aluminum case, so the path length should be half the diameter, or 10.5mm = 0.413″.

“his looks to be on the order of the thickness of his index finger, call it about 0.5″.

An error in my analysis above: I forgot that the 2170 is an aluminum case, so the path length should be half the diameter, or 10.5mm = 0.413″.”

“Length 11.417″ (290 mm)
Width 8.504″ (216 mm)
Thickness 0.2795″ (7.1mm)”

Bolt cell thickness is 0.2795″.

Nevermind. Wrong spec.

It was LEAF’s cell.

Dang! I can’t seem to find the Jaguar iPace video where the battery engineer is showing us the 34 modules in the pack. I seem to remember that there were 8 flat cells in each module and wondered about the thermal path length for the cells in the center. Can anyone find that video? Help!

This is not a comparison. This is a fluff piece. Not enough data. Try again next year.

I don’t know that I would call this a “fluff piece”, but at best it’s a case of rather lazy armchair engineering, one which compares only a few factors on paper to each other while ignoring other, equally important factors.

A proper, meaningful comparison would be to do bench tests of both battery packs under high discharge rates (thus generating a lot of heat), and measure the actual temperature at the core of the packs to see which was better at maintaining a temperature within safe operating parameters.

No offense to the writers of the article, but other than an academic exercise it won’t matter much to either ‘3’ buyers or ‘BOLT ev’ drivers. This article is actually much beyond ‘splitting hairs’.

Pick a car and say its batteries have the higher thermal resistance to the chilled glycol. All the control system has to do is take into account the calculated temperature of the battery (determined by the amount of heat currently being removed – and also by the present current going through the batteries and their SOC), and DECREASE the set temperature point of the chilled glycol to set the battery temperature EXACTLY where it is wanted to be.

The only downside is a very minor decrease in refrigeration efficiency. No Biggie.

Your new web format for mobile is cutting off the images. Or images are not resizing properly.

Yes it’s a programming issue on some devices. We are fixing it. Thank you!

The cooling system on the Model 3 seems to be an improvement over the S/X as the Model 3 was able to tackle laguna seca raceway for several laps without overheating and going into limp mode.

That may have more to do with the motors than the battery. I’m fairly certain the performance bottleneck of the Model S/X was cooling of the induction motors. Model 3 switched to PM motors which should be easier to keep cool.

“I’m fairly certain the performance bottleneck of the Model S/X was cooling of the induction motors.”

I’m more than “fairly” certain this is incorrect. The Model S which was run in the Pike’s Peak race was seen with bags of ice stuffed under the car just before the race, to cool the battery pack down. If the problem was the motor overheating, then cooling the battery pack wouldn’t help much!

But if you have any facts or authoritative citations to back up your assertion here, then I’d be interested in seeing them.

See Rocky H’s post here:
The first link has a thread that shows the P85D has significantly higher limits than the P85 did. If the battery was the bottleneck this would not be the case.

Note: the Roadster bottleneck was the PEM and motor:
The main issue is that it’s difficult to cool the rotor in a induction motor. There was speculation that Tesla would use a liquid cooled rotor in the Model S since they got a patent for it, but in the end they didn’t.

See this thread for in depth discussion about the cooling of Model S motors:

Journalists (and probably the people who used the ICE pack idea) all assume by default the issue is the battery because that is the intuitive thing. That said, the battery pack idea wasn’t completely useless. By coldsoaking the battery it frees up some extra cooling capacity in the cooling system for use for cooling the inverters and motor.

Even the simple cooling plate that GM has employed would be great to see in the Leaf. That’s one battery pack that is crying out for some active cooling.

Not sure about the relative plus/minuses of the various schemes (Tesla vs. GM), but GM does seem to have optimized the cost-effectiveness of their system. None of the parts seem really expensive to make, and of course they didn’t get into the other end of the Chilled water system other than to show the physically dinky ‘hot water heater’ for wintertime operation. But Tesla’s system seems to work – haven’t heard anything as of yet on TMC as to any growing pains with them as of yet.

Even in the roadster – Battery cooling was never an issue. Waiting for the battery to warm up in the cold weather was another thing.

It’s hard to tell what if the cooling system or the chemistry is the bottleneck, but the Bolt’s DC charging capabilities seem fairly weak. There was talk about the 80kW spec, but that’s only the spec for the port/chargers. The latest test shows it maxes out at 55kW.

The Model 3 LR maxes out at 116 kW. Even if you scale down by capacity to estimate SR’s speed, it’ll be 82kW.

Point well taken. I wonder if the battery and the ability to get rid of heat limits the Bolt DCFC rate, or is it that the contractor(s) in the Bolt can’t handle the current?

There are a lot of unknowns related to the charging capabilities. It could be the Bolt is capable of faster charging but that GM simply chose a very conservative taper to maximize battery longevity. You can see with the entire design of the car that GM puts less emphasis on DCQC. Hell, the car doesn’t even come with the capability by default, it’s only available as an additional option.

Yup, and those of us who purchased a model without fast charging like the $750 plus tax we saved since we don’t need it/can’t use it.

People don’t seem to understand that if it is an OPTION and you want it, then get it for yourself. I don’t like being forced to pay for something I’ll never use.

At the pack level, the Model 3 seems to be about 25% more energy dense than the Bolt.

With Bolt pack at 57kWh/960lbs, that’s 16.84lbs/kWh at the pack level.

Tesla’s EPA application has the 3’s pack at 1058lbs, and it used 78+kWh on the J1634, which is 13.56lbs/kWh, or a little less using the 76kWh figure in the diagnostic screen.

There is something strange about the claimed size of the Bolt pack. The GM battery engineer insists it is 60 kWh. Every time we charge the car, the app shows the percent charge in 1% increments. Every time we drive we see the kWh used for the miles driven, and the percent of charge remaining. The numbers align with it being 60 kWh. But LGs label says 57 kWh…definitely strange.

Perhaps you could compare the two EPA documents, and tell us how they compare?

Warren: This is easily explained.

Per D.O.E. constructs the battery is ‘Rated’ at its 95% level to take into account ‘Average Degredation’.

57kwh is 95% of 60.0 kwh. My 22,000 mile BOLT ev’s capacity is far closer to its ‘new’ capacity than its official ‘rated’ capacity.

Perhaps in a few years my battery will have degraded down to its ‘official rating’. In the meantime I’m enjoying the 60 kwh battery.

Supposedly, battery life decreases ‘rapidly’ to 95%, then stabilizes for a LONG time. Hasn’t happened to me yet, so therefore, as far as the BOLT ev is concerned, I’m assuming the stable time is going to be VERY LONG.

I don’t think looking at Tesla pack weight will be that fair. This is because Tesla’s new pack has the DC-DC converter, the battery charger, as well as contactors inside. These parts are outside the pack in practically all EVs. You would have to include the weight and volume of those in the comparison.

As for ease of assembly, I don’t think we can say yet until disassembly is done and we see how it’s made. If I remember correctly, the talk about gluing the cooling ribbon was complete speculation.

Having those parts inside the battery isn’t the best from a maintenance standpoint.

Time will tell what’s best for EV’s. GM has already made press releases that their next battery platform will be cylindrical based. They don’t come out and say it directly, but it reeks of clues. Low cost, low height, high energy density, scalable.

Wow you totally read that differently. There are few benefits to cylindrical cells. GM like Jaguar, BMW, Nissan, Renault, Mercedes, etc… are all using pouch cells. No one outside of Tesla that is planning higher volumes wants to deal with thousands of cells in a pack.

No one cares about how many cells there are in a pack. The reason you see a lot of focus on pouch cells is due to hybrids.

We all know the Tesla power packs barely lose any juice over time. I wonder what it is going to be like for the bolt in a four-year period,!

I’d guess it will probably be pretty similar. The only car that is absolutely notorious for suffering battery degredation is the passively cooled Nissan Leaf.

Look at the Volt’s battery for hints. GM is one of the most conservative manufacturers when it comes to battery preservation.

22,000 miles on my Bolt ev so far. If LI batteries supposedly lose capacity quickly and then stabilize, I have yet to see it. Everyone here seems hung up on cooling – If there is an issue it probably only starts to become a problem with the DCFC option, and the extreme tapering the system does which irks the 55kw DCFC optioned buyers. But since I charge at a 6-7.5 kw rate most of the time, I’m unaffected by it and most of the heating I see is after a long, long drive at expressway speeds on a hot day. The refrigeration system is adequate for that, and this hang up on ‘thermal resistance’ between the pouches and the chilled glycol seems overblown, since when immediately lifting the hood after an extended drive – the heat removed from the radiator is mammoth – with the refrigeration system drawing initially all the public charger (6 kw) charging power – cooling the battery from just getting off the expressway and now cooling it for charging, admittedly a dinky amount at this rate. The unit seems to run at full tilt for only a few minutes, but that is due to chilling the… Read more »

One think I think we can all agree on: the Leaf (any gen) has the worst thermal management system of any mainstream EV on the market, new Leaf included.

Glad you posted the prof’s video, Bro…..

The body language here is key: He seems almost DISGUSTED with the 2018 Leaf, which basically says you can’t fast charge it, or, if you can, its only after a long wait NOT CHARGING AT ALL… So why fast charge, or pay for the option if you don’t dare use it?

Meanwhile, the Bolt ev, and all Volts (and ELR’s) just work.

Coolant needs to be changed at 5 years/150,000 miles like most GM vehicles. That is one thing lately about GM vehicles. Not fancy, but they perform the bread and butter functions of a vehicle flawlessly. The ev driver doesn’t have to worry about much. You just drive when you want and plug-in when you want.

Driving OR plugging in extends the battery life – just the opposite of the Leaf.

I think so. Plus experience has shown that to be correct.

I wonder which one is more cost effective.

If you really want to compare modules, this study is more of a Bolt vs. P100D module comparison but I understand they it was labeled M3 to get clicks.

To do a real apples to apples comparison, you need to add the mass of the cold plate and the hold down brackets to the Bolt module in order to have the equivalent functionality of the model s module. Once you do that then you’ll have an ever bigger difference in energy densities. Its not just “a few pounds.”