Welcome to Tesla Motors Club
Discuss Tesla's Model S, Model 3, Model X, Model Y, Cybertruck, Roadster and More.
Register

Understanding the battery heater

This site may earn commission on affiliate links.
So why doesn't it warm the battery more aggressively? Why is there still a regen limit after preconditioning?

Because it is a waste of energy and doesn't help the battery. The battery works just fine at cold temperatures. Only at temps below freezing will it start to show limit at driving and charging. So as long as the battery is just above those temps, there is very little benefit in keeping it warmer but a big hit in energy used. You don't get much extra range out of the battery by heating it up from 10 Celsius (50 F) to 25 Celsius (77 F). But it would require about 8.5 kWh to heat the battery from 10 to 25 degree Celsius. That's almost one full hour of full power charging with a single charger. Since pre-heating only goes on for 20-30 min and a good amount of energy is needed to heat the cabin, you get an idea how much is left to heat the battery. The battery heater itself has a max power of 6 kW. So even if it would run at full power for one hour, it would not be able to heat the battery to a level where it has zero regen limit.

Excellent explanation, very helpful (as were the clarifications and discussion by others).
 
Last edited:
Another bug: when regen is "disabled" it's really really disabled. You will see power draw for the heater(s) even under deceleration. errr.......

Ya, this one bugs me too,... when there is a call for power by cabin heater or battery pack warming, and regen is limited, you'd think it would be better to generate and divert (dump) that power to where it is needed. The battery charger would be the first priority, but not the only thing. Next would be battery heater, and if that is being hit with all it can take, then the excess beyond that would spill over to cabin heater (if on). Then and only then any excess that can't be soaked up by something with the immediate need would cause freewheeling (less capture). And this would all happen seamlessly and in a balanced way per speed reduction cycle.. because we have computers that can figure this out on the fly. Not rocket science. Just car science.
 
Another bug: when regen is "disabled" it's really really disabled. You will see power draw for the heater(s) even under deceleration. errr.......
I've noticed that, I've always said, at least use the regen electricity to feed the heaters, no point drawing electricity from the batteries while giving up free electricity at the wheels. I understand you can't put it back in the pack, but there's no reason you can't feed all the heaters with it.
 
Also, for people who like the idea of using shore power to heat the battery.. have you considered just laying down a heater under the car where you park it every night, that is thermostatically controlled. Imagining a couple of wall mount ceramic flat slab heaters laying on top of a styrofoam cushion off the cement floor. Probably 2 inches of air gap between heater and battery. Place it so you just drive over it wheels on either side, the thing would be narrow enough to fit between inside tire edges. I have actually thought about doing this.. instead of heating whole garage just put it where it counts. Heat rises.
 
Also, for people who like the idea of using shore power to heat the battery.. have you considered just laying down a heater under the car where you park it every night, that is thermostatically controlled. Imagining a couple of wall mount ceramic flat slab heaters laying on top of a styrofoam cushion off the cement floor. Probably 2 inches of air gap between heater and battery. Place it so you just drive over it wheels on either side, the thing would be narrow enough to fit between inside tire edges. I have actually thought about doing this.. instead of heating whole garage just put it where it counts. Heat rises.

Hmmm.....Slab heaters are usually hydronic or heat trace cable in the concrete. It heats the concrete and not much else. It would have to be a slim, narrow and weatherproof radiant heater (think of the salt/slush on the undercarriage and wheel wells). Possibly not electrically safe. Also - where would the water go? it would probably freeze up *around* the Tesla.
 
I've noticed that, I've always said, at least use the regen electricity to feed the heaters, no point drawing electricity from the batteries while giving up free electricity at the wheels. I understand you can't put it back in the pack, but there's no reason you can't feed all the heaters with it.

It has to feed the pack regardless. There is no direct line from the DU to the auxiliary systems.
 
  • Like
Reactions: brainhouston
Here is an example: Let's say the battery is at freezing (0 C / 32 F) in the morning. You want to heat it up to be perfectly warm to have no regen limitation (25 C / 77 F). It would require 14 kWh. That's the equivalent of 47 miles of energy. That's more than most people's daily commute! It would be wasted every morning, every day! So Tesla decided to only heat the battery to a point where it's fine to drive with a reasonable limit on regen and save a lot of energy. Using the unavoidable losses in the drive train to further heat the battery as you drive is a much smarter and more efficient way to use energy. Using these losses is free energy to heat the battery.

The battery gets reasonably warm while charging, right? And I didn't do all that well in high school physics, but isn't it correct that once the battery has achieved a temperature it takes less energy to keep it there?

What I'm getting at is that another reason to have the option to heat the pack manually would be to keep the pack warm from the time the charge has completed until the time the car is to be driven. Yes, we can try to time our charges to end exactly when we are leaving, but sometimes stuff happens. We leave later than we planned to, we estimated badly, or we just left a little too much wiggle room. Please correct me if I'm wrong, but I would think that a pack that has finished charging in a cold garage and then needs to sit for 30-45 minutes could be kept warm reasonably efficiently with some pack heating, and that the energy used keeping the pack warm could very possibly be more than made up for via regenerative braking that would otherwise not have been present. While that may sound like a "border case" I think it's pretty much what many of us in cold regions experience almost every day if we're making the effort to try to plan our charges such that they end shortly before we depart.
 
The battery gets reasonably warm while charging, right? And I didn't do all that well in high school physics, but isn't it correct that once the battery has achieved a temperature it takes less energy to keep it there?

I will be interested to hear others chime in here, but I don't think this is the case. It should always be more efficient to have the system turned off and then heat it back up. It's like the (incorrect) argument that you should leave your thermostat heat on while you're gone because it "takes more energy to heat the house back up from cold." This is not the case.
 
The battery gets reasonably warm while charging, right? And I didn't do all that well in high school physics, but isn't it correct that once the battery has achieved a temperature it takes less energy to keep it there?

What I'm getting at is that another reason to have the option to heat the pack manually would be to keep the pack warm from the time the charge has completed until the time the car is to be driven. Yes, we can try to time our charges to end exactly when we are leaving, but sometimes stuff happens. We leave later than we planned to, we estimated badly, or we just left a little too much wiggle room. Please correct me if I'm wrong, but I would think that a pack that has finished charging in a cold garage and then needs to sit for 30-45 minutes could be kept warm reasonably efficiently with some pack heating, and that the energy used keeping the pack warm could very possibly be more than made up for via regenerative braking that would otherwise not have been present. While that may sound like a "border case" I think it's pretty much what many of us in cold regions experience almost every day if we're making the effort to try to plan our charges such that they end shortly before we depart.

I've been measuring 92-93% efficiency charging. That's not a whole lot of waste heat left. From 55-60F max battery power estimates 55 minutes at 5kW draw, so you'd have to charge 60+kW to get the battery to max power temp assuming you could do that in 55 minutes.

I think the main problem is they only have one power bus. I don't think it's physically possible to not charge the battery if regen produces power, therefore, no regen.

- - - Updated - - -

I will be interested to hear others chime in here, but I don't think this is the case. It should always be more efficient to have the system turned off and then heat it back up. It's like the (incorrect) argument that you should leave your thermostat heat on while you're gone because it "takes more energy to heat the house back up from cold." This is not the case.

If it's resistive heat this is true, but not for heat pumps (variable efficiency curve).
 
The battery gets reasonably warm while charging, right? And I didn't do all that well in high school physics, but isn't it correct that once the battery has achieved a temperature it takes less energy to keep it there?

Charging definitely creates heat losses right inside the battery so yes charging does warm the battery. Not a lot, though, unless you are Supercharging. In winter, it's a good idea to use the charge timer and set it to a time that charging finishes just before you leave. But as you say, sometimes the timing doesn't work.

The energy needed to heat something up depends mostly on the mass of an object. How much energy is needed to keep it at a certain temperature depends on a lot of things. Mostly difference between temperature of the battery vs outside temperature. How much airflow there is, how much surface area it has in relation to the mass and so on. Kind of hard to calculate. I think as long as we are talking about 1-3 hours it should all be fine. You shouldn't have to worry.

Honestly, I think the car should have an option where you can tell it, 'I want to leave at 8 am'. The car then does whatever it needs to do to get ready. Charging, preheating the battery, heating the cabin. How much it needs of everything depends on many things, like state of charge, desired state of charge, outside temperature, battery temperature. It would be a pain if we had to do all of that manually. The car has all the info, just let it handle it. Someone suggested that to Elon 2 years ago at a public meeting. He nodded and said I agree. We are still waiting for it :)
 
Charging definitely creates heat losses right inside the battery so yes charging does warm the battery. Not a lot, though, unless you are Supercharging. In winter, it's a good idea to use the charge timer and set it to a time that charging finishes just before you leave. But as you say, sometimes the timing doesn't work.

The energy needed to heat something up depends mostly on the mass of an object. How much energy is needed to keep it at a certain temperature depends on a lot of things. Mostly difference between temperature of the battery vs outside temperature. How much airflow there is, how much surface area it has in relation to the mass and so on. Kind of hard to calculate. I think as long as we are talking about 1-3 hours it should all be fine. You shouldn't have to worry.

Honestly, I think the car should have an option where you can tell it, 'I want to leave at 8 am'. The car then does whatever it needs to do to get ready. Charging, preheating the battery, heating the cabin. How much it needs of everything depends on many things, like state of charge, desired state of charge, outside temperature, battery temperature. It would be a pain if we had to do all of that manually. The car has all the info, just let it handle it. Someone suggested that to Elon 2 years ago at a public meeting. He nodded and said I agree. We are still waiting for it :)
Does "smart preconditioning" take care of one part of that equation? I do think it's kind of lame you can't program preconditioning or set up start and stop times for charging. Even the volt has a "depart at xx time" charging setting. Even without doing everything you suggested it seems like a rather small software update to allow a "complete charging by xx time" option. At least that could solve some of these issues.
 
I think the main problem is they only have one power bus. I don't think it's physically possible to not charge the battery if regen produces power, therefore, no regen.

It is possible - they could do the same as they do for cabin preheat from AC power: control the charging current (regen level in this case) to be exactly the same as the load being drawn by the HVAC, such that although the battery is connected to the same bus the net current in or out of the battery is zero.

They may well be doing this already - it wouldn't be immediately obvious from the power gauge on the dash, since that shows net power at the battery including HVAC load (ie. at zero regen with the heater on, the gauge is showing a sliver of orange; if regen was occurring as above, there would be no green on the gauge it would just be showing zero).

Trouble is, a couple of kW of regen doesn't actually give much deceleration.

Because you can't get more efficient than resistive heat. Why some suggest MS should have a heat pump is beyond me.

That's not correct. Resistive heat is 100% efficient, but a heat pump can be 300% or more - you get more heat out of it than the electrical energy put in (the extra energy coming from the 'cold' side of the pump).

However, heat pumps don't work so well when the difference between hot and cold sides is too large - so they'd work less well in really cold weather, just when you really care about it, and you'd probably have to have resistive heat as well. So while I originally hoped that Model S would have a heat pump, I can accept that Tesla have probably got it right in deciding the extra cost/weight/complexity isn't worth it for the minor benefit it would bring.
 
  • Like
Reactions: Big Earl
Because you can't get more efficient than resistive heat. Why some suggest MS should have a heat pump is beyond me.

That's a common misconception. You might want to do some reading on how heat pumps works. In short, instead of producing 1kW of heat using 1kW of power (resistive) it moves heat from one place to another, effectively heating (or cooling). And the process of moving energy uses less energy that generating heat.

Simple google it, here's a well worded quote (source):

Coefficient of Performance (COP) is the most common measurement used to rate heat pump efficiency. COP is the ratio of the heat pump's BTU heat output to the BTU electrical input. Conventional electric resistance heaters have a COP of 1.0. meaning it takes one watt of electricity to deliver the heat equivalent of one watt. Air-source heat pumps generally have COPs of 2-to-4, meaning they deliver 2-to-4 times more energy than they consume. Water and ground-source heat pumps normally have even higher COPs of 3-to-5.
 
  • Informative
  • Like
Reactions: MTOman and Big Earl
Charging definitely creates heat losses right inside the battery so yes charging does warm the battery. Not a lot, though, unless you are Supercharging. In winter, it's a good idea to use the charge timer and set it to a time that charging finishes just before you leave. But as you say, sometimes the timing doesn't work.

The energy needed to heat something up depends mostly on the mass of an object. How much energy is needed to keep it at a certain temperature depends on a lot of things. Mostly difference between temperature of the battery vs outside temperature. How much airflow there is, how much surface area it has in relation to the mass and so on. Kind of hard to calculate. I think as long as we are talking about 1-3 hours it should all be fine. You shouldn't have to worry.

Honestly, I think the car should have an option where you can tell it, 'I want to leave at 8 am'. The car then does whatever it needs to do to get ready. Charging, preheating the battery, heating the cabin. How much it needs of everything depends on many things, like state of charge, desired state of charge, outside temperature, battery temperature. It would be a pain if we had to do all of that manually. The car has all the info, just let it handle it. Someone suggested that to Elon 2 years ago at a public meeting. He nodded and said I agree. We are still waiting for it :)

It goes without question that there should be a way to at a minimum, schedule the end-time for a charge. I don't think anyone really disagrees on that. I've written to Tesla about that a couple of times in my letters requesting a battery preheating option.

As for the affects of charging on battery heating, and the affects of a warm battery on efficiency, I have some interesting data to share, from the last three mornings.

As some of you may know, since last February I have been recording the regular trips between Ithaca and Syracuse my wife and I make in our P85D, and comparing our efficiency results against what EV Trip Planner would predict. I started doing this because at the time I was concerned that we were not seeing the efficiency gains from Torque Sleep that others were seeing.

In addition to the metrics available through the car, I also attempt to record outside temperature, by looking at it regularly during the trip, and then estimating an average, and wind speed, using the tool available in the thread here: Tool to estimate the head wind while driving !. I ask my wife to do the same thing. I also record weather conditions like rain, snow, etc.

The last three mornings are interesting for the following reason. The temperature my wife reported was 10 F each morning. She reported wind speed of a headwind of 4MPH one morning, and 5 MPH the other two. She drives the same speed on the highway each day, and I also use Visible Tesla with geofencing to record when she leaves and arrives, and will include that data, but the bottom line is that these three trips, for all intents and purposes, were about as similar in conditions as we could hope for. The one variable was battery temperature based on charge time. On Monday morning the pack was cold, as it had barely charged at all overnight, since it had charged to 90% the night before. On Tuesday morning the pack was somewhat warm, with the charge having completed about 90 minutes before my wife left for work. This morning the pack was apparently warmer, as my wife left just 30 minutes after the charge completed. The data is below, and I think it tells quite a story about the impact of charging on battery temperature and efficiency.

DistanceRMTotal AverageCabin OutsideWeather Duration
Date(Miles)UsedEnergy (kWh)Energy (Wh/mi)TemperatureTemperatureWindConditions(Minutes)
1/4/1653.96919.937068104HWFlurries65
1/5/1653.96218.233868105HWClear64
1/6/1653.95917.131870105HWClear60


Note that this morning's trip, 1/6, was the fastest, had the cabin heater set two degrees higher, and in spite of that was still the most efficient by far.
 
  • Like
Reactions: Big Earl
How are you controlling the battery heater? I have never found a way to turn it on or off.

Do you mean you preheat the cabin?

I left the mountains last week after skiing. It was 25F. I left with 41% and got home with 16%. Regen was 100% disabled for the entire 6000 foot decent. I would have normally netted 11.3 KW of regen on the same decent trip from that location. Just the day before, I used the battery heater before leaving for a short drive under the same temperature conditions and it only took 2% to get the battery warm enough for full regen.

I realize that's a bit of a corner case.

The other scenario is in the morning using shore power to heat the battery so I have full regen (90% charge) when taking off. The benefit is that I use less brake pad and have more power. I don't use to the point where it says "ready" for max battery power, but it's enough that I'll make 430KW rather than 395KW when taking off in the morning. In this case, 10 minutes of battery heater prior to leaving while on shore power does it.