Battery Cooling System in the Bolt EV

[picasso]

Great article from InsideEVs!

GM’s Volt, and Tesla’s Model S both use an active liquid battery cooling system. Tesla snakes a flattened cooling tube thru their cylindrical cells resulting in a very simple cooling scheme with very few points for leakage.

GM Volt and Spark EV use thin prismatic shaped cooling plates in between the cells with the liquid coolant circulating thru the plate.

The BMW i3 cools the bottom of the battery case with refrigerant eliminating the liquid coolant entirely.

The Volt cooling scheme is very effective from a cooling point of view but it is complicated. The cells are encased in multiple plastic frames.

These frames repeat and are then stacked longitudinally to form the whole pack. The main feed line for the liquid coolant runs along the bottom edges of the pack. This main coolant passage is cast into each plastic frame and as the frames are stacked lengthwise the coolant passage is formed. Each inter cell cooling plate is fed off this main feed line.

The problem with this scheme is there are multiple potential points where leaks can develop since there needs to be a seal between each plate but we must point out that there doesn’t seem to be a lot of problems reported in production Volts. Tesla’s system is simpler and less prone to leaks since each battery module has one continuous cooling tube.

This “repeating frame” cooling system seems to have been abandoned in the Bolt. Here is an excellent video animation of the Bolt EV battery pack and power train:

At 1:04 minutes into the video we can see one three cell group being removed from the pack. The active inter cell cooling plates that were used in the Volt are totally absent . Instead we a see a passive plate which is wrapped around each cell. Keep it simple.

Where does the liquid coolant go? It does not appear to be between the cells as used in both the Volt and the Tesla.

Consider this high resolution slide of the Bolt battery pack.

Now look at a close up view .

In this photo we can see what appears to be liquid coolant connection fitting on the front of the pack. Inside the pack we can see liquid coolant tubes. We know they are liquid and not refrigerant tubes because GM has told us so in the early spec release.

The following “bottom cold plate description” is not directly from GM but is based on the author’s inferences from GM’s high resolution photo of the Bolt’s battery pack.

We can also see that the liquid coolant tube drops down to the bottom of the pack into a flat black plate. The authors believe this is a bottom cooling plate. Bottom cooling plates for battery cooling are not unprecedented. The BMW i3 uses it and GM had just such a system in the Spark EV when A123 was the supplier of the Sparks battery. This Spark bottom cooling plate was abandoned however when LG Chem was chosen as the battery supplier for the 2015 Spark in favor of the cooling scheme used in the Volt.

Searching the web we find that the same supplier of the Volt’s inter cell active cooling plates also makes bottom cooling plates. These bottom cooling plates can be dimpled or channeled to take the liquid coolant. The ingenious part is that the cooling bottom plate can also be used as a structural member of the battery pack. The cooling plate could also double as the structural battery tray.

Simpler, lower cost and less prone to leaks.

 

[columbo]

Maybe GM has started to think like Tesla - this drivetrain and battery system is so simple the whole thing could be dropped and swapped to get the customer back on the road, and the techs can figure out what went wrong later...

 

[mark111]

Very interesting article. Thanks for posting it on Chevy Bolt forum.

 

[lukem5]

This seems like a flawed design for battery life longevity. We all know heat rises.... so why would the cooling plate be put on the bottom of the cells? This would create an uneven warmer area on top of the cells and cooler on the bottom part of the cells- a recipe for battery degredation disaster.

Not to mention the large pack of cells under the back passenger seats located on top of the main bottom arrangement of cells... do those have a cooling plate under then? I think not. These cells will undoubtedly get hotter and be more prone to failure.

I do not have confidence in the bolt's battery life, it will be interesting to see what kind of range owners are getting after 50k miles or if the packs will need to be replaced due to uneven wear in the cells (due to insufficient cooling) causing bigger electrical problems.

The Volt seems to have a solid battery pack, but they use the inter cell cooling system.

 

[gbobman]

They are also using LG's higher nickle batteries that are supposed to have much greater temperature ranges. I doubt they are taxing the life of the cells, especially since they put the 8yr/100,000 on it. They don't need more liability.

 

[EldRick]

Heat only rises when you are talking about convection in a fluid medium. the Bolt coolant appears to be mainly flat and two-dimensional, so convection should not be an issue.
Why would you ASSUME that there is no cooling for the under-seat battery back?

I do not have confidence in the bolt's battery life...

That's funny, apparently Chevy does, as they warrantee it for eight years.

 

[DucRider]

Heat only rises when you are talking about convection in a fluid medium. the Bolt coolant appears to be mainly flat and two-dimensional, so convection should not be an issue.
Why would you ASSUME that there is no cooling for the under-seat battery back?

I do not have confidence in the bolt's battery life...

That's funny, apparently Chevy does, as they warrantee it for eight years.

Remember that 8yr/100K warranty is federally mandated (it falls under emission controls) and applies to any BEV sold in the US. There is no definition of failure provided, so manufacturers are free to put there own terms on it.

With the Bolt, you can lose 39.99% capacity at any time (from degradation) in the first 8 years/100K miles and not be covered by the warranty.

Tesla (and others) specifically exclude ALL degradation from the warranty. With them, it is possible that if the car can drive at all, the battery will not have "failed".

That being said, active liquid cooling (of any type) on a battery is likely control temps much better than air cooled batteries. The bottom of desirable methods is the passive battery cooling of the LEAF, e-Golf, and i-MiEV.

 

[SeanNelson]

With the Bolt, you can lose 39.99% capacity at any time (from degradation) in the first 8 years/100K miles and not be covered by the warranty.

A lot of people take that warranty spec and assume that it means that GM expects all Bolts to loose 40% of their capacity in 8 years. But the reality is that GM does not want to be replacing battery packs in 8 years, so they've doubtless engineered the battery for the typical usage case such that only a very small percentage of them actually reach that limit.

 

[lukem5]

Heat only rises when you are talking about convection in a fluid medium. the Bolt coolant appears to be mainly flat and two-dimensional, so convection should not be an issue.
Why would you ASSUME that there is no cooling for the under-seat battery back?

I do not have confidence in the bolt's battery life...

That's funny, apparently Chevy does, as they warrantee it for eight years.

I assume mainly because it's an american engineered car.... Also because the detail photo of the battery pack shows a single large cooling plate underneath the battery, I don't think they would add a separate cooling plate to the smaller pack on top and I can't see one in the photo.

True about heat only rising in a fluid medium. however, the cooling pack is still only contacting the bottom part of the battery pack... unevenly cooling it. I am pretty sure uneven temperature fluctuation across the cells would cause degradation/other problems.

It's obvious the best way to cool batteries is using teslas patented design where coolant flows across the entire cell. Why didn't chevy do this?

The more I read this forum about all the other little problems this car has is really turning me off. Never trust american engineered cars, especially with new technologies.

 

[michael]

With the Bolt, you can lose 39.99% capacity at any time (from degradation) in the first 8 years/100K miles and not be covered by the warranty.

A lot of people take that warranty spec and assume that it means that GM expects all Bolts to loose 40% of their capacity in 8 years. But the reality is that GM does not want to be replacing battery packs in 8 years, so they've doubtless engineered the battery for the typical usage case such that only a very small percentage of them actually reach that limit.

That is true, however what we don't know for sure is what is the typical degradation at 8 years/100K miles. I think it is likely that the typical case will be substantial if the owner does not exercise good battery management practice.

If you assume typical 200 mile range for a full depth discharge, this is about 500 cycles. While we don't know the specific characteristics of the LG cells used in Bolt, typical LG data sheets state degradation < 40% after 500 full cycles. So this is what I would expect to see.

The way to avoid this is to use good battery management practices over the 100 K miles...avoid full charges and full discharges whenever practical, and charge "just in time" for departure whenever practical.

Notice I said "whenever practical"...not "always"

 

[SeanNelson]

The more I read this forum about all the other little problems this car has is really turning me off. Never trust american engineered cars, especially with new technologies.

You mean like... Teslas?

Seriously, everything you've written is pure wild-eyed speculation, which I why I'm ignoring it.

 

[SeanNelson]

With the Bolt, you can lose 39.99% capacity at any time (from degradation) in the first 8 years/100K miles and not be covered by the warranty.

A lot of people take that warranty spec and assume that it means that GM expects all Bolts to loose 40% of their capacity in 8 years. But the reality is that GM does not want to be replacing battery packs in 8 years, so they've doubtless engineered the battery for the typical usage case such that only a very small percentage of them actually reach that limit.

That is true, however what we don't know for sure is what is the typical degradation at 8 years/100K miles. I think it is likely that the typical case will be substantial if the owner does not exercise good battery management practice.

I certainly wouldn't argue against any advice to treat your battery well. But with regard to the warranty, GM would not be so foolish as to assume that everyone will do that. And again, they most certainly do not want to be replacing any substantial number of batteries after 8 years. So I think we can assume that unless you are particularly hard on your battery you won't have to worry about loosing that much capacity over the 8 year period.

Some capacity loss? Certainly. Less capacity loss if you treat your battery well? Absolutely. But I'm pretty certain that 40% loss is not going to be anywhere near typical.

 

[michael]

It's impossible to say. I agree, they feel confident few will exceed the fade limit. What I don't know, but I"m sure GM and LG do know, is how tight is the sample distribution of the fade.

For example, it's possible -- NOT SAYING IT'S SO-- that they expect the batteries to have faded 30% mean with a standard distribution of a few percent at warranty expiration. Virtually none would have exceeded the warranty limit, and they wouldn't have used batteries excessively good (read: expensive) to do the job. But most of us would be driving 170 mile cars, not 238 mile cars at that time.

In three to eight years we will know. For now we can only assume and speculate. My assumption is that the batteries will fade noticiably over that span of time and miles.

 

[SparkE]

All EV batteries fade "noticeably". 10% over 8 years is "noticeable".

 

[cyaopec]

All EV batteries fade "noticeably". 10% over 8 years is "noticeable".

Yes, if but if you're already beating the EPA numbers by 10-20% then you're simply declining back to what you expected when you bought the car. My Volt has 90 thousand miles on it, 50k if which are electric. It is four years old. I'm getting 10.4 to 10.6 kWh out of the battery each day, which is what I was getting when it was new. Also, yesterday I got approximately 62 miles out of the battery on a 100.5 mile commute in Los Angeles, of which 32 miles each way is on the freeway. Declining from my numbers would still put me way ahead of the EPA 38 miles electric rating.

 

[boltage]

My Volt has 90 thousand miles on it, 50k if which are electric. It is four years old. I'm getting 10.4 to 10.6 kWh out of the battery each day, which is what I was getting when it was new.

The first generation Volt has a 16 kWh battery. So if it uses only about 10.6 kWh out of it when new, then it is presumably being managed extend battery life (e.g. not charging all the way up to the full 16 kWh).

 

[michael]

My Volt has 90 thousand miles on it, 50k if which are electric. It is four years old. I'm getting 10.4 to 10.6 kWh out of the battery each day, which is what I was getting when it was new.

The first generation Volt has a 16 kWh battery. So if it uses only about 10.6 kWh out of it when new, then it is presumably being managed extend battery life (e.g. not charging all the way up to the full 16 kWh).

Not only that....For all we know the charge window may be slowly widened so that the same 10.5 or so kWh are available even though the battery's full capacity may have reduced from 16 kWh to something less.

I know some may reply "what do I care if I still get the same useful amount" and I do agree. But that should not be used as an argument to prove the battery doesn't fade. It simply means the designers used good practice to avoid range reduction despite actual fade.

 

[cyaopec]

My Volt has 90 thousand miles on it, 50k if which are electric. It is four years old. I'm getting 10.4 to 10.6 kWh out of the battery each day, which is what I was getting when it was new.

The first generation Volt has a 16 kWh battery. So if it uses only about 10.6 kWh out of it when new, then it is presumably being managed extend battery life (e.g. not charging all the way up to the full 16 kWh).

Not only that....For all we know the charge window may be slowly widened so that the same 10.5 or so kWh are available even though the battery's full capacity may have reduced from 16 kWh to something less.

I know some may reply "what do I care if I still get the same useful amount" and I do agree. But that should not be used as an argument to prove the battery doesn't fade. It simply means the designers used good practice to avoid range reduction despite actual fade.

The charge window is not widened. This is a myth that has been discussed for years on the Volt forums and GM engineers have always said so. Batteries in the Volt do fade. Some people get 10.2 or 9.9, etc on 2011 Volts that are old. But it's no where near 10% let alone 40%. Great thermal management and a "relaxed" state of charge of 65-70% counts a lot for the longevity. You can simulate this in the Bolt by only charging to 65-70% mon-fri if you don't need to drive 200+miles in a day. GM counts end of life of the Volt battery as when the capacity drops >20%, which their engineers have said unofficially should occur in 11 years in a hot, unprotected climate like Arizona or in 15 years in a cold environment like Michigan. A small amount of battery fade is way better than a failed oil pump, fuel pump, etc on a regular car. With an ICE it's either working or it's not. NOT leaves you stranded. Battery fade only means a bit more planning on your part.

 

[SeanNelson]

A small amount of battery fade is way better than a failed oil pump, fuel pump, etc on a regular car.

To be fair, the Bolt has coolant pumps which, if they fail, may have just as catastrophic effect.

The Bolt has a lot less to go wrong, but there are still some points of vulnerability as we've seen from some of the reports on this forum.

 

[phxsmiley]

I currently have a Leaf, and I'm considering a Bolt, but I'm concerned about keeping the pack from degrading in the Phoenix, AZ heat. I'm certainly well schooled in how the Leaf batteries handle the heat.

I understand that when the Bolt is plugged in and the temperature is above approximately 90 degrees, the cooling system will run to keep the battery cool. Is that correct, and does anyone have numbers on how much this would draw, on average, when the outside temperature is around 105-110? I'm considering the cost with our time-of-day electric rates.