Chevy Bolt EV Forum banner
  • Hey Guest, welcome to ChevyBolt.org. We encourage you to register to engage in conversations about your Bolt.
1 - 20 of 79 Posts

Registered
2017 Premier
Joined
1,282 Posts
Discussion Starter · #1 ·
So I've been gathering data and observing when and how my car has been running it's battery cooling processes under different scenarios. I've gathered quite a bit of preliminary data, I haven't poured over the Torque logs yet (I'm not a data analysis guy) so there are some gaps but I feel like I have enough of an insight into when and how the car does battery conditioning to be of use to fellow owners (especially since we're still knee deep in summer.

The first thing I've observed is that the car REALLY cares about whether it's "plugged in" and will definitely cool sooner and to a lower temperature if it's plugged in vs not. FTR I've only observed being "plugged in" to a L2 charger, I don't know whether connecting to L1 or DCFC will produce the same results.

The second major criteria is whether or not the car is "on", this changes the conditioning behavior both "plugged in" and "unplugged".

I have yet to observe any evidence of battery conditioning when the car is both "off" and "unplugged". I've seen battery temps as high as 37C/99F with ambient temps as high as 42C/108F and the car has not decided to cool the battery even with high states of charge (above 80%). I have heard that other owners have witnessed off/unplugged battery conditioning but I've never seen it so I have no idea what temp might be required to trigger it other than above 37C.

One of the things I don't know is what specific temperature sensor(s) the car is looking at when it decides to cool the battery. There are several OBDII PIDs for battery temperature and several for electronics temperature. Most of the time these sensors all have different temperature measurements within 1C-3C of each other. I'm sure this is something that could be nailed down looking at the data.

Also note that all the numbers I'm using are Celsius meaurements as observed via Torque because that's what the car uses. I've converted many of these numbers to Farenheight for the benefit of my fellow countrymen but I've used standard rounding so 31C becomes 88F not 87.8F.

There are two main modes of battery cooling that I've observed, I'll call these "aggressive" and "weak" modes. I have only observed aggressive mode cooling when the car was plugged in.

Under aggressive mode cooling the car will start cooling at about 33C/91F and stop cooling at about 27C/81F (battery temperature)

Under weak mode cooling the car will start cooling at about 35C/95F and stop cooling at about 31C/88F.

OK with all that out of the way here is the list of different scenarios I have seen.
off/unplugged - n/a
on/unplugged - weak mode
off/plugged/charging - none
off/plugged/not charging (charge complete or charge not needed) - aggressive mode
on/plugged/charging - aggressive mode

Some other notes:

If the vent fan for the passenger compartment is on when the car is performing battery cooling you will get cool air into the cabin whether or not you have "heat and AC" on in the climate settings. Interestingly this will also cause the displayed percentage of your power used for climate settings to increase and the car will attribute more power to climate settings than it will to battery conditioning by about a 2:1 margin.

The converse is not true, if you are running cabin air conditioning and the battery is as warm as 32C/90F the car will make no effort to cool the battery.

If anybody wants to pour over the data themselves I've got all my Torque logs uploaded to a shared Dropbox folder. Feel free to peruse.

Conclusions:
I'm actually quite disappointed in the decisions that GM made when designing the battery conditioning profiles for the Bolt. Really the only time the car decides to prioritize cooling the battery is when the car is plugged in. I'd expect that most people who live in hot climates are like me and do not usually have ready access to an L2 charger during the hottest part of the day. Now I understand that many people would not want to see their range being decreased due to battery conditioning but IMO they have gone too far and are allowing the battery to get and stay much too hot which will negatively effect the long term capacity retention of the battery.

Between this frankly pathetic approach to battery conditioning and the lack of options for limiting the maximum SoC it's abundantly clear to me that GM does not remotely prioritize the long-term health of the battery. I expect that a Bolt battery will still significantly outlast a Leaf battery that completely lacks active battery cooling but I don't expect it to perform nearly as well over the long term than a Tesla Model S battery. Being that the average amount of time that new car buyers keep their cars is only about 6 years combined with the reality that many Bolt drivers are lessees instead of purchasers I suspect that GM (correctly) figures that most of the people who will have to deal with the repercussions of this approach will be 2nd or 3rd owners.
 

Registered
Joined
8,885 Posts
Love how data driven you are.

It's unclear to me that GM has not prioritized battery longevity enough, as there are way too many variables at play. Perhaps the Bolt is more temperature tolerant than Tesla. Perhaps the max and minimum voltages are more conservative than Tesla.

Time will tell, but I'm sure the Bolt batteries will hold up better than the Leaf. I've heard of people complaining about battery degradation in their Gen I Leaf that is only 3 years old.
 

Registered
Joined
8,885 Posts
It could be that not being as aggressive about battery conditioning when on battery power alone is the better way to preserve battery life. While temperature is a component of battery longevity, so is battery use. The less use it gets, the healthier it is. There is some balance to be had between maintaining a healthy temperature range, and not excessively using the battery. Given that, it makes complete sense to maintain tighter temperature control when wall power is available, and be a little more loose when it is not.
 

Registered
Joined
771 Posts
Great data.

I also have concerns about battery performance and potential degradation in the heat of the Mojave desert here. The last 30 days has been averaging 109掳F during the day and about 90掳F at night. The garage stays above 94掳F most of the late afternoon and through the early morning. I plan to buy a evaporative cooler this weekend. This is my data below. Note the time is "PM", and the Bolt is inside the garage:


So I had been doing a little research as well. Perhaps you have seen this: What is the Best Electric Vehicle Battery Cooling System? , and 2017 Chevy Bolt Battery Cooling Details. These give some overview regarding the types of EV battery cooling system designs, and also as important it seems is the type of batter chemistry and format. An example is the 7000-odd Tesla 18650 battery cylinders require different cooling engineering than the Bolt pouch. Also, the Volt's battery was by far the best cooling system, but complex and potential risk of coolant getting to the cells.

I was able to speak briefly with a guy at Battery University who suggested that the higher ambient temperatures have little affect on the longevity of a modern, actively cooled EV battery. He emphasized that by far the best way to prematurely kill any EV battery was:
A.) Frequent Fast charging (DCFC - Where a Fast charging session could raise a pack temperature to 140掳F)
B.) Frequent Level2 charging
C.) Charging above 80% SoC

Although it was free advice, it may be helpful?
 

Registered
Joined
1,989 Posts
{...}

I was able to speak briefly with a guy at Battery University who suggested that the higher ambient temperatures have little affect on the longevity of a modern, actively cooled EV battery. He emphasized that by far the best way to prematurely kill any EV battery was:
A.) Frequent Fast charging (DCFC - Where a Fast charging session could raise a pack temperature to 140掳F)
{...}
I think that the temp of the battery pack during charging (especially during DCFC) depends quite a lot on both the rate of charge, and how much you cram into it at once (and the SoC when you stop).

I rarely charged my Spark EV up to over 90% at any charge rate (generally 80-85% when I charged, which wasn't every night), When I DCFC'd (which I did every couple of weeks, to add a quick 40 miles), I rarely charged past 65% or so - thus keeping the battery temp lower. Also (mainly because I had found a free 24 kW charger that I used for about a year) I generally charged at lower kW rates.

I saw very, very low degradation of the main battery over a 2 year period with my Spark EV - less than 5%.
 

Registered
Joined
8,885 Posts
So I guess this means that it is better to plug in the car when not in use during the day.
I would say generally yes, especially if Hilltop Reserve setting is used. Then there is the added benefit of the car checking the 12v battery more frequently, I believe.

I don't see much benefit in going multiple days without plugging in, especially when hilltop can be used to limit the charge. It just makes it more difficult to remember plugging in when needed since you get out of the habit of always plugging in.

My wife used to plug in the Prius anytime we went somewhere together, which was all the time. On the few occasions she wasn't with me, I'd forget to plug in.
 

Registered
2017 Premier
Joined
1,282 Posts
Discussion Starter · #11 ·
I was able to speak briefly with a guy at Battery University who suggested that the higher ambient temperatures have little affect on the longevity of a modern, actively cooled EV battery.
But I think this assumes that an "active cooling" system is actually being used when it's needed. Just having a system doesn't help if you don't use it.


I wonder if this aggressive/weak conditioninig also depends on the state of charge. I would think that if the SOC is high, the conditionining should be more agressive than if the SOC is 50%.
I have not noticed a difference in behavior based on the SoC, I have tested above 80% and below 45% in the same situation and the car has behaved the same.
 

Registered
Joined
8,885 Posts
There was that Canadian guy with the hockey avatar who ran his battery SoC very low and tried to DCFC in very cold temps. It refused to charge saying he had to connect to L2 first.

I'm guessing there is some state of charge, perhaps around 40%, that prevents even the weak conditioning when the car is off.

My WAG is there is "aggressive" conditioning when plugged in, "weak" conditioning when unplugged and on, and "even weaker" conditioning when unplugged and off when the SoC is above some threshold, and no conditioning below that.

Hasn't this been thoroughly discussed in the cold discussion? I would think similar behavior exists with hot temperatures.
 

Registered
2017 Premier
Joined
1,282 Posts
Discussion Starter · #13 ·
Hasn't this been thoroughly discussed in the cold discussion? I would think similar behavior exists with hot temperatures.
I will admit that I haven't seen that thread, it's not something I would have looked for because the coldest temperatures we will ever see here are in the high 20's (F).

Also, as I understand it cold temps may prevent or slow your charge and temporarily reduce your capacity but they won't permanently reduce your capacity like high temps can.

I think eventually we need to compile all the data we owners can gather into a wiki or at least a thread, GM sure isn't going to tell us how the car behaves in certain scenarios.

I can tell you I've already changed my behavior based on what I've learned during this process. I'm now much more likely to plug in during periods of high temperatures knowing that the car will keep the battery significantly cooler when plugged in I believe it's better to keep the battery cool than to avoid sustained SoC above 80% (hilltop reserve seems to be fairly consistent at about 85% raw SoC for my car).
 

Premium Member
Joined
2,576 Posts
I have driven over 25k miles in my Bolt, in a 4 season environment (0 to 100F) and fast charged liberally over the last 18 months. I fully charged the other day, and the raw SOC% was 96.5 % according to Torque Pro, same as it was new.
I have 0 worries about battery longevity of my Bolt. While small degradation is inevitable, I actually am more confiden in GM's battery longevity than Tesla's. 25k miles and 0 discernable degradation.
 

Registered
12/16 build, 2017, white LT
Joined
14,880 Posts
Between this frankly pathetic approach to battery conditioning and the lack of options for limiting the maximum SoC it's abundantly clear to me that GM does not remotely prioritize the long-term health of the battery. I expect that a Bolt battery will still significantly outlast a Leaf battery that completely lacks active battery cooling but I don't expect it to perform nearly as well over the long term than a Tesla Model S battery. Being that the average amount of time that new car buyers keep their cars is only about 6 years combined with the reality that many Bolt drivers are lessees instead of purchasers I suspect that GM (correctly) figures that most of the people who will have to deal with the repercussions of this approach will be 2nd or 3rd owners.
If you want to know battery thermal limits for the Model 3, I believe they are revealed in the CAN code in this video.

https://youtu.be/CLOEGFtFIPA
 

Registered
12/16 build, 2017, white LT
Joined
14,880 Posts
I have driven over 25k miles in my Bolt, in a 4 season environment (0 to 100F) and fast charged liberally over the last 18 months. I fully charged the other day, and the raw SOC% was 96.5 % according to Torque Pro, same as it was new.
I have 0 worries about battery longevity of my Bolt. While small degradation is inevitable, I actually am more confiden in GM's battery longevity than Tesla's. 25k miles and 0 discernable degradation.
I believe the % charge is based primarily on the pack voltage. I have a seven year old bike battery that still charges to 100% at the same voltage as new, but the usable capacity went from 24 Ah to 20 Ah.

You need to go by the total capacity number, which presumably takes into account voltage sag, under load, from increased impedance.
 

Premium Member
Joined
2,576 Posts
I have driven over 25k miles in my Bolt, in a 4 season environment (0 to 100F) and fast charged liberally over the last 18 months. I fully charged the other day, and the raw SOC% was 96.5 % according to Torque Pro, same as it was new.
I have 0 worries about battery longevity of my Bolt. While small degradation is inevitable, I actually am more confiden in GM's battery longevity than Tesla's. 25k miles and 0 discernable degradation.
I believe the % charge is based primarily on the pack voltage. I have a seven year old bike battery that still charges to 100% at the same voltage as new, but the usable capacity went from 24 Ah to 20 Ah.

You need to go by the total capacity number, which presumably takes into account voltage sag, under load, from increased impedance.
If degradation was occurring and the SOC window was changing, the raw SOC% would also be changing with time. But so far it's still 96.5% when fully charged.
 

Registered
12/16 build, 2017, white LT
Joined
14,880 Posts
If degradation was occurring and the SOC window was changing, the raw SOC% would also be changing with time. But so far it's still 96.5% when fully charged.
My point is that state of charge simply means that the pack has reached a specific voltage, which is not the same as capacity. Capacity is the average wattage (voltage X amperage) over time. Unless a cell has actually failed, as opposed to the normal course of all cells losing capacity, the maximum voltage will remain the same. Even a single cell, with reduced capacity, will allow the battery to reach maximum voltage at full charge, as was the case with some recalled batteries. Since the cells are top balanced, it is only near minimum SOC (voltage) that the odd reduced capacity cell will reveal itself.
 

Registered
Joined
182 Posts
OK with all that out of the way here is the list of different scenarios I have seen.
off/unplugged - n/a
on/unplugged - weak mode
off/plugged/charging - none
off/plugged/not charging (charge complete or charge not needed) - aggressive mode
on/plugged/charging - aggressive mode
Great write up. One data point - I noticed that my car conditioned while DC fast charging on a road trip yesterday. The battery pack was already at 30C when I pulled in, and by DCFC for over an hour I suspect it hit a temperature threshold that triggered conditioning. The car was definitely off and the conditioning happened mid-cycle.
 

Registered
Joined
189 Posts
I fully charged the other day, and the raw SOC% was 96.5 % according to Torque Pro, same as it was new.
As someone who has given up waiting for the new Bolt I ordered 6 months ago to arrive in Ontario, Canada (now that our incentives are gone) and will instead be flying out to BC and driving a used one 4500 km across the country home, I hope you're correct.

But my understanding, from my research of the Leaf when I was on the fence, is that the Leaf has a SOC visible in car, a true (or raw?) SOC visible through a PID in Torque pro and a SOH visible through a PID in Torque Pro (or leaf spy). And the entire point of the SOH is that even the true SOC degrades as the battery does. The true SOC on the Leaf is the true % of the degraded capacity.

How do you know, where are you getting your info from that this "raw SOC" can be used to track degradation as the Leafs SOH can be and isn't just the true SOC (that just reports the true % of the now degraded capacity)?
 
1 - 20 of 79 Posts
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top