Welcome to Tesla Motors Club
Discuss Tesla's Model S, Model 3, Model X, Model Y, Cybertruck, Roadster and More.
Register
This site may earn commission on affiliate links.
I had always assumed that the battery pack required some heating before high-speed DC charging, but that once the charging got underway most of the noise the car makes was due to cooling. I had also assumed that ~25°C was best for the batteries. See this (admittedly old) post which mentions staying <35°C: A Bit About Batteries

I normally charge to 75% at home from a 240V 15A socket, but yesterday I charged from 50% to 99% at a (non-Tesla) 50kW CCS charger. Outside temperature was ~15°C, and the charge rate was ~35kW for most of the charge.

I have recently started using TM-Spy to monitor a lot more real-time info about the car. This is the first time I have used it during a fast-charge session, and I noticed some very unexpected thermal behaviour. The image below shows a situation near the end of the charge with SoC ~97%. The charge power into the battery was ~11kw and the cell temperatures were ~53°C (which seems pretty hot to me), but the car was still trying to HEAT the battery. It was pouring 3.5kW into each of the front and rear motors, with a reported stator temperature of 130°C, which was heating the coolant up to 60°C at the battery inlet! I would have expected the car to be trying to cool the batteries down, not heat them up. It also seems like an awful waste to be using 7kW of the "charge" power just to make things hot. Does anyone have any thoughts on this??

hot-battery.png
 
The target temperature for On Route Battery Warmup is 40°C, based on a comment by a Tesla engineer at the V3 Supercharger test station when it first opened in Fremont. I think the battery will charge at max rate when that temp has been reached, which is not trivial in cooler weather and with only a 7 kW of heating from both motors. Based on Tesla Bjorn videos, the charge rate will throttle down below the optimum taper when the batteries heat to 60°C.

Like you experienced, I've also seen in his videos where the car is heating the battery very late in the charge cycle but I don't understand the rationale, unless it's as simple as it always tries to heat the battery to just under 60°C. You may have noticed, but after the charge session stops, the car then tries to vigorously cool the batteries. Based on this, it's clear the Supecharger protocol is optimized for charging speed rather than charging efficiency.
 
Last edited:
Ahhh, my favourite topic.

tl;dr version: Battery get hot. Very hot. Battery stay hot. Much waste. Confusion. Anger. Acceptance.

Many, including my past self, have preconceived notions of how the battery temperature management must work. Many, including my past self, are apparently wrong.

I too assumed all these noises and fuss were from cooling, and I was very wrong (see: Time Remaining vs. kW rate and also Battery Cooling: How Important?). I ain't got no high horse to ride here, but I do have new knowledge to share!

The first thing to know is Tesla seems to treat all DC charging the same, whether it be from a Supercharger or a CHAdeMO station. I'll explain why this is bad later.

Charging Basics
Lithium ion batteries, like those in our car, can be safely charged faster at lower states of charge (SoC). At higher SoC, they must taper off. In general, too much heat is bad, but Tesla has found it is beneficial for charge speed (not necessarily health).

Preconditioning Prior to Supercharger Arrival
Precondition temperature targets are variable. The goal is to raise the battery temperature by wasting heat in the drive motors, which heats the battery, which allows for a faster charge rate. The AWD is potentially better than the RWD at this, since it can fully use the front motor for heating.

Generally, at low SoC, the target heating temperature is also relatively low, perhaps 30°C. The battery may very well be above this already and preconditioning might not be needed, especially if you have leftover heat from the last Supercharger (uhh, spoiler warning).

At high SoC, it gets hotter prior to charger arrival. I don't recall an exact number here (it's not often I stop to charge at high SoC since it's so slow), but it's definitely significantly warmer. Perhaps 43°C? Maybe a bit higher? Completely depends on SoC though.

Conclusion: My theory for this behaviour is that it needs more heat to pump the power in at higher SoC. Why? No idea, really. But at low SoC the battery can already accept much, much more power than high SoC and maybe needs less "help" from active heating to achieve high charge rates. Additionally, low SoC implies it has more time during charging to continue to heat the battery anyways.

During Charging
This sucker's gonna get even hotter now!

Charging induces quite a lot of heat. Most people think you want to get rid of this heat. Hah! Tesla laughs in your face.

100% confirmed, the battery will continue to be actively heated with the drive motors while DC fast charging, in addition to the heat added via charging at such high powers. And it will continue for a LONG TIME.

The drive unit internals get super hot. Apparently like 100°C hot. This is to heat the battery, again. Which gets super hot. Like hurts-to-touch hot. I want to say it can definitely get as high as 55°C. While heating, the motors will be hotter than the coolant, and the coolant will always be hotter than the battery. Yes, the "coolant" is actually being used to heat the battery. Terminology is confusing.

The technicals: 3.5kW/motor is what we've seen. This seems about right but I have some more precise data coming soon (I expect it's a bit higher or I have some unaccounted-for electrons). The "most efficient" Model 3 would probably then be the Long Range RWD (not AWD), since it can waste less heat during the charge compared to its charge rate.

Cooling may or may not occur during your charge (especially based on my SoC theory above). I'd almost bet money that if you stop before 90%, it probably will still be heating.

Conclusion: Your motors and battery will be rippin' hot during charge, and cooling may not occur until the very end of your charge, or...

After your charge
Cooling begins! The radiator fan kicks in (yes, a Model 3 has a radiator, it's even in the usual spot for cars), and the AC may kick in depending on parameters I appear to have no understanding of. Anywho, it doesn't cool that much. This takes a few minutes, much less time than the heating takes.

Many minutes down the road, you will notice your battery is still at something like 35°C. This is apparently normal. Both in the cool of Colorado in March and in the hot summer of my home area, this seems true. It's like a happy temperature for the battery to stay at; it could easily cool it further in Winter with the radiator alone, but it just... doesn't.

35°C sounds too warm. Everything we know about phone batteries says this is too warm, and the Model 3 battery pack is different from those in size more than anything else. Many praise Tesla for their thermal management, but few seem to know just how hot the battery pack usually is.

Conclusion
: Your battery will stay quite warm after a charge, and battery cooling is fairly minimal.

Extra benefits of a hot-off-the-charger battery
Well, you can coast into the next Supercharging without it needing to waste as much to heat. For Winter road trips this is kind of a good feature? I'm not sold.

But where this is sweet for Winter is actually with the Model Y and cabin heating. It's just using the 1000lb battery as a massive thermal tank with a heat pump, unlike the Model 3 (which uses a resistive eletrical heater for the cabin). Having a hot battery means it can draw heat from it for quite a while for heating the cabin before resorting to less efficient methods (which are nearly equivalent to the Model 3 heater).

Conclusion: Hot-after-charge batteries are actually useful in the context of Winter road trips (where you're likely using fast charging the most). Summer? Not so much...

Implications for CHAdeMO Charging
This sucks for CHAdeMO charging. I've sat at CHAdeMO stations for an hour and it heats the battery the whole time. Keep in mind it's not preconditioned prior to arrival for CHAdeMO, there's just no way to do this.

25kW CHAdeMO chargers exist around me. 50kW ones are more common, but sometimes 25kW is what you have to use. In another case, I had access to only a 10kW one or a 6kW Level 2 charger. The 6kW Level 2 charger is faster for Model 3.

Why is that? The car decides to take a decent chunk of the available power for heating the battery because it's DC charging. Basically, they have the assumption that all DC charging will require maximum throughput. This is great for the Supercharger network, but absolutely not necessary for CHAdeMO charging, especially lower-powered ones.

For Level 2 AC charging, it behaves entirely different. It only gets the battery up to about 15°C (I can't remember, somewhere in my post history) even for an 11.5kW setup. So Tesla is basing the target temperature on the type of charging, not the communicated available charge power.

Even at a 50kW station, something near 18% of the station's power output is getting completely wasted. The battery is almost certainly hot enough (I've charged at Superchargers when the battery is cold, so I have an idea of what it can pull at lower temps). If anything, only brief heating might be necessary to increase throughput. Don't get me started on the 25kW station math. It makes Model 3 one of the worst rural BC EVs - it has the range, but the charging sucks. Yes, other EVs handle this better.

To make matters worse, the car lies to you in the unit most people understand: Time. The 10kW station made this the most obvious. Tesla does the time calculation for DC fast charging based on powered received from the station, not power that will actually make it to the batteries. Normally this is close to rounding error for a Supercharger, but it makes a significant difference when nearly 20% at minimum is being wasted at a CHAdeMO stall.

Conclusion: Tesla Model 3 is the most wasteful and slowest-to-charge EV if you must use CHAdeMO stations (like in BC, Canada). I don't see them fixing this uncommon case because frankly, I don't think they care about the consequences.

"camalaio's take"

I do think it's an excessive waste in some cases as well. In Winter I definitely appreciate the preconditioning, because it is very slow otherwise. But I'd really like the option to, say, add 20% to my charge time if it means wasting less energy and/or having a cooler battery.

Thinking about this for California makes me a little sad. So many Teslas there using Supercharging as their primary charge method. It's a lot of waste and a lot of thermal cycling.

As the industry moves towards faster charging standards, bowing to consumer desires, I don't see this getting "better". Tesla probably doesn't give you the option to slow down Supercharging, because that implies you effectively tell the next person in line they need to wait longer because you care about your precious battery and/or waste like it's more important than where that person needs to go in a timely manner. Early adopters adapted to slowing down in general since they need to stop and charge. Common adoption will require the fastest charges possible.

The singular exception to all this hot battery stuff is Track Mode. I've yet to get my hands on one to measure what it does, but Track mode will actively and preemptively engage battery/motor cooling to my understanding. This is an interesting contrast to the Model S/X Ludicrous mode, which apparently heats the battery so that they can achieve higher discharge rates (just like heat is used to raise charge rates). The implied difference is launch (short duration) vs. laps (long duration).

Actual tl;dr: Yes, the battery is heated quite a lot and this is normal for a Model 3. Cooling is quite rare and very brief, and still leaves the battery objectively warm. Model 3s run a lot hotter than many people think, and the temperature is more passively managed than many people expect. It took a lot of measuring, reading, and being wrong to determine this.
 
Adding in another bit on L2 charging, just because you also mentioned it in your post...

L2 charging does potentially result in battery heating as well, the target temps are just a lot lower. At some threshold I still can't figure out, it will start heating until the coolant reaches some variable temperature (this also varies by SoC I think?), about 20°C. Room temperature-ish. The end result is the battery is kept around 15°C or so (59°F) when in cold temps. If it somehow dips below about 10°C (50°F), it would probably start heating again (I've never witnessed this). The heating is on/off rather than variable.

The way it works is that the motors will suck ~3.5kW each. If your available power exceeds what the motors are drawing, it will go to the battery (and yes, the drive unit internals are getting very very hot just like for Supercharging). The AWD sort of delays charging because of this: it can suck back ~7kW for motor heating, which exceeds the 6kW typically available at public stations in North America. A RWD, however, will be able to use about 2kW for charging right away, and eventually the full amount once it has warmed. Which is faster might depend on how long you'll be charging; obviously 2kW for 10 minutes is better than 0kW and heating the battery for 10 minutes, but there may be a break-even point since the faster-heated battery can also reach maximum charge rate sooner.

For L1 charging, this is not good (but there isn't really a better alternative). The behaviour is the same as far as I can tell, but you have much less power available. On a standard 120V setup, you have 1440W available. 300W is going to the pumps and computers. You have just about 1kW to heat the battery with - this takes a very long time, and if outside in very cold temps apparently has a hard time getting to the point of being warm enough to charge at all.

L2 charging will never result in so much heating that the battery will need to be cooled, unlike DC fast charging. The heat here is genuinely useful just to enable charging at any reasonable rate at all, and is not excessive.
 
Wow, this is all really interesting info. Clearly Tesla are okay with their batteries being a lot hotter than I expected. There is a 350kW CCS charger not too far away, so I'll go and find out how much extra heating happens there, and also see if it will charge while track mode is active, and what difference that makes to the heating/cooling.

Thanks for the replies - makes me feel a bit more comfortable that this is normal behaviour. But I'm glad most of my charging is at home at L1.
 
  • Like
Reactions: wildguess
and also see if it will charge while track mode is active, and what difference that makes to the heating/cooling.

Oh man, some software developer out there probably does not like you. Eager to hear the results :p

Didn't the latest update add preconditioning to 3rd party DC fast chargers?

Yes, but also no. To my understanding, at least for North America, these are the 3rd party stations that have Tesla connectors. There are incredibly few of these. For Europe where the Model 3 comes with CCS, my understanding is that they are actually listing many third party stations since they are compatible without an adapter. No CHAdeMO stations show up on my map in Canada nor the US for the places I've been.

L2 charging goes to 80 amps. In older cars (Model S) that can accept this cooling will occur.

This is a thread in the Model 3 section about a Model 3's behaviour, which is limited to 48A (and only that high on the Long Range version). There are many more problems with my post than the charge current if you try to apply it to older or even current Model S/X, like the battery heating method. I assume they behave different in many other ways actually, but I'm unfamiliar with them. Apologies if I did not make it clear that this information only applied to Model 3 (it's probably safe to assume it applies to Model Y though).
 
  • Informative
Reactions: Arctic_White
Didn't the latest update add preconditioning to 3rd party DC fast chargers?

There are a lot of 50kW CCS chargers in Australia, with some 350kW types now as well. All of these can plug straight in to the AU Model 3, but none of them show up in the car's navigation, and there is no way to precondition enroute to them. The first time I charged at the 350kW station (from 50% SoC) I only got ~30kW from it. On my second test I arrived at the charger with 1% SoC, and it only took a few minutes for the charge rate to ramp to 183kW (~1,200km/h), which is close to the limit imposed by the Model 3's ~400V battery and the 500A rating of the liquid-cooled CCS plug. From what camalaio has said above, it seems preconditioning is much more important at higher SoC, which aligns with my experience here. But yeah, no preconditioning for any CCS chargers around here. Nor any NoA compatible roads, either! :-(
charging.png
 
I had always assumed that the battery pack required some heating before high-speed DC charging, but that once the charging got underway most of the noise the car makes was due to cooling. I had also assumed that ~25°C was best for the batteries. See this (admittedly old) post which mentions staying <35°C: A Bit About Batteries

I normally charge to 75% at home from a 240V 15A socket, but yesterday I charged from 50% to 99% at a (non-Tesla) 50kW CCS charger. Outside temperature was ~15°C, and the charge rate was ~35kW for most of the charge.

I have recently started using TM-Spy to monitor a lot more real-time info about the car. This is the first time I have used it during a fast-charge session, and I noticed some very unexpected thermal behaviour. The image below shows a situation near the end of the charge with SoC ~97%. The charge power into the battery was ~11kw and the cell temperatures were ~53°C (which seems pretty hot to me), but the car was still trying to HEAT the battery. It was pouring 3.5kW into each of the front and rear motors, with a reported stator temperature of 130°C, which was heating the coolant up to 60°C at the battery inlet! I would have expected the car to be trying to cool the batteries down, not heat them up. It also seems like an awful waste to be using 7kW of the "charge" power just to make things hot. Does anyone have any thoughts on this??

View attachment 558874

I have watch all of Bjorn's videos on Youtube and you should watch all his Tesla charging videos. You learn A LOT about Teslas and charging.

Anyway, the car wants the batteries to be close to 60C. That's normal. 30C is considered cold. You might notice that even if outside was 80F, and you navigate to a supercharger from home, it will precondition the batteries.
 
Extremely interesting discussion, I learned a lot from carmalaio's explanation and Push2Eject's data points.

I've no reason to doubt any of this, but why do the cooling fans start screaming midway through a supercharging session? If it's trying heat the battery, why would it need to push keep out of the radiator/condenser? It's done this with and without A/C on.
 
Extremely interesting discussion, I learned a lot from carmalaio's explanation and Push2Eject's data points.

I've no reason to doubt any of this, but why do the cooling fans start screaming midway through a supercharging session? If it's trying heat the battery, why would it need to push keep out of the radiator/condenser? It's done this with and without A/C on.

There's multiple reasons the fan can run, but I think you have those covered. If you started at really low SoC and between active heating and power going in you get a very hot battery, I could see it turning on the fans sooner than I've implied.

When they spin up like that, my observation is they're not on for very long (and also end up not cooling the battery that much). Same for you or not really?

I guess I could pay more attention to what's happening. They might also just be trying to cool the drive units rapidly (remember they're much hotter) and get rid of the heat overhead in the coolant (just so it no longer heats either the battery or motors). This would explain it being a brief fan spinup since the thermal mass would be smaller than if their intent was to significantly cool the battery.
 
That’s a good question, I don’t recall how long the fans were on. I have an SR+as well, so the smaller pack might heat a little faster. I have an adapter and Scan My Tesla app, but don’t use it often… I’ll have to break it out next time a Supercharge.
 
You know it was a fascinating read and a long answer to the question of 60c is just fine. I think we can say that in terms of battery and thermal management we get the best in class resulting in fast charge speeds and long cell life. Nothing to worry about here.
 
You know it was a fascinating read and a long answer to the question of 60c is just fine. I think we can say that in terms of battery and thermal management we get the best in class resulting in fast charge speeds and long cell life. Nothing to worry about here.

I want to make a small correction that Tesla themselves will tell you (though they have a hard time saying it directly).

Having the battery this hot is not good for it's health. 60C is not "just fine". It's a trade-off Tesla has deemed acceptable to achieve faster charge rates, which is not the same thing as being good for the battery. I'm highlighting the difference between "normal" and "beneficial".

Charging at high such high temperatures (especially when not useful, e.g. how it works for CHAdeMO/CCD stalls <=50kW) does bad things to your battery longevity. Nearly every other manufacturer (EVs, phones, tool batteries, etc.) prevents you from charging at high temperatures. This very likely isn't because Tesla's chemistry is better or unique, just that they're accepting a risk for a different gain.