Cold weather range.. Cell warming

[TikiSteve]

I've read a lot of the cold weather range threads and still feel unfulfilled.

Prior to being an owner, I read that the majority of energy loss in cold weather was not poor cell performance, but cabin heating. Now I have a MYP with heat pump and have watched the 75 minute Weberauto youtube on the heat pump (very impressive!), I am convinced that at highway speeds, there is enough heat recovered (1KW) to give the heat pump a COP of >2, I am thinking this would mean ~750 watts for cabin heating, compared to 20KW being burned up by the drive train to cabin loss is small. The part that makes my head explode is how much heating do the batteries need at highway speeds (and say 10F)? If they are self sustaining due to internal dissipation, there should not be much heating required from the cell energy. Yet I still see many comments that cold weather loss will be 20-30%+. Are there any charts on cell capacity (range) vs temp (with heatpump). Certainly cell pre conditioning for departure will be a factor, at least at the start.

Background. I work at a ski resort. On sunday morning, bright and early I drive 15 miles 2000' uphill. The car is cold soaked in mid vermont, for 10 hours. When finished work, I descend the 15 miles, 2000' to my 2nd home. I then need to drive 180 miles to my primary residence (little to no time to top off). I have been able to do this with 85% charge in warm weather, I am not sure I will make it in the cold. If I can't make it, I have to divert 30 minutes out of the way to hit a supercharger.

FWIW, I am a retired electronics designer. My last project was a lithium ion pack, it used Samsung INR1650. This chemistry at 30F had a fully charged capacity of.....ZERO!

 

[GtiMart]

There shouldn't be active heating during your drive to heat the battery. Waste heat from motors / inverters will be circulated in the battery if it's still cold, instead of possibly being used for your cabin (in a heatpump model).

As you drive, the air circulating around your car will cool your car quite fast. You can't just calculate the energy needed to keep your car at temp while sitting idle. I'm not an expert at calculating that energy loss but it's important. Another thing is that the air you're driving through is denser in winter, resulting in a higher rolling resistance. Finally, you might be using winter tires which also might have a higher rolling resistance. If you don't check your tire pressures regularly you might end up with lower than 42psi as your tires cool and pressure thus reduces.

EDIT: and a cold battery will give back less energy than a warm one so while it warms up it will be "less efficient". You can bypass this is you warm at home before leaving, but you will have spent energy warming the battery.

 

[Dr. J]

I've read a lot of the cold weather range threads and still feel unfulfilled.

For your fulfillment, see: https://www.tesla.com/support/winter-driving-tips?redirect=no. That showed up in my inbox today.

 

[chdlmc]

For your fulfillment, see: https://www.tesla.com/support/winter-driving-tips?redirect=no. That showed up in my inbox today.

One item I saw within your link today came up for me today with a pop up warning on my screen.

“Leaving your vehicle plugged in whenever possible and keeping the charge level above 20% when not plugged in will reduce the impact of cold temperatures.“

What is the difference if I have a cold night at home, car parked outside, and the SOC is 15% / 25% / 35%? The above message from Tesla is making me think the battery will behave differently at 15% Vs either 25% or 35%.

The overnight low I’m expecting is 30F.

 

[ucmndd]

What is the difference if I have a cold night at home, car parked outside, and the SOC is 15% / 25% / 35%? The above message from Tesla is making me think the battery will behave differently at 15% Vs either 25% or 35%.

The overnight low I’m expecting is 30F.

The lower the state of charge, the more energy the battery “locks out” to protect itself. You’ll typically see this manifest as a section of blue on the charge slider in the app - energy in the battery that it won’t necessarily let you use for propulsion.

In my Model S, if I parked on a cold night with a low state of charge, say around 20%, the car would “lose” another 5-8% overnight. In general, don’t park on cold nights like that at low states of charge without charging.

 

[chdlmc]

Ok. I plan to frequently park outside without being plugged in. If I get level 2 charging at home later, that may change. (I have level 2 charging available now at work.)

That being the case, I’ll plan to have 30% plus SOC during winter months if I’m parking overnight unplugged.

 

[AAKEE]

The lower the state of charge, the more energy the battery “locks out” to protect itself. You’ll typically see this manifest as a section of blue on the charge slider in the app - energy in the battery that it won’t necessarily let you use for propulsion.

In my Model S, if I parked on a cold night with a low state of charge, say around 20%, the car would “lose” another 5-8% overnight. In general, don’t park on cold nights like that at low states of charge without charging.

There is no energy locked out.

The car measures the real SOC even in very cold but the car displays a ”artificial” lower SOC to compensate for the cold losses.
In short, the displayed range (if displaying range instead of percent) is shortened to compensate for the losses from cold.

I have never seen more than 4-5% lower SOC than the real one on neither my MSP ir the M3P.

Using SMT (scan my tesla) the SOC displayed is the ”real SOC” so looking at the display and SMT we can see both values at the same time. Its also shown in teslafi as the ”normal SOC” with the ”cold SOC” in blue.
This morning, cell temp -16C (-24C outside).
Charged to 80% one day before this photo and the car was parked outside not used:

(”SOC expected” seem to be the displayed SOC at this time).

Real SOC 80%, and cold displayed SOC is 76%. These can also be seen in the Tesla app as well but its not easy to distinguish the exact numbers except the same number as the car displays.

 

[ucmndd]

There is no energy locked out.

…

Real SOC 80%, and cold displayed SOC is 76%.

I guess I don’t understand the difference? Are you saying it’s just a display modification and the car will let you use that ~4% if you need it?

 

[AAKEE]

I've read a lot of the cold weather range threads and still feel unfulfilled.

I had a M3P with heat pump and I logged a lot of data. Also drove with a screen showing data in front of the steering wheel.
I live in a quite cold climate…

The car newer heat the battery spontaneous. Do not heat it when just driving and do not heat it when when parked. Except when doing scheduled departure or starting the cabin heat in the app. In these cases it do not heat the battery if it isn’t more or less freezing cold, and when it heats the battery it only heats it to a few degrees above freezing (like 5C or so).

The heat pump is only allowed to use the battery heat down to about 11-12C (at least the in the M3P 202*1) so a “simple precondition” is not enough to allow the heat pump to use battery heat.

When driving (in cold WX) the heat loss in the motors and other heat warms the battery and when reaching about 17-18 C the heat pump starts using the heat from the battery which will continue until the cell temp is 11-12C, then this will repeat.

In colder temps than about -10C, the heat pump do not use ambient air to heat the cabin but works in a inefficient mode “Lossy mode” ( about like the electrical heating before the heat pump was added).

When the heat pump can use the battery heat in extreme cold the consumption is much lower than when it have no heat source to use and have to use the “Lossy mode”
I had a opportunity to test this many times, heres one example:
Supercharging and driving in about -20C one way with the battery wam enough to supply the heat pump with heat energy to cabin all the way.
191Wh/km with the warm battery.

On the way back I started with the battery at 10C ( it was connected to a slow charger so it was not very cold but not warm enough for the heat pump to use the heat.
260Wh/km when I passed the supercharger on the way home, 251Wh/km arriving at home.
I had the excact same max and average speed and the ambient putside temperature was the exact same. It was also 100% no wind (inversion in the cold and the neareat airport had 00 kt in the metar all the time).
Both the supercharger and the end of the trip is at sea level.

Today i made a 250 km drive in -24C to -20C with my 2023 MSP
First 195 km with a cold battery (276Wh/km before the preconditioning for supercharging set in). This part of the drive is actually slightly downhill (130m ish).
The last part of the drive (55km) is not that downhill.
Same average speed and same max speed.
Sliiightly warmer (or less cold) by a few degrees, this part of the drive the consumption was 214Wh/km according to the screen (teslafi is not realky correct here).
Summarized the consumption is about 30% lower in really cold WX, both with my M3P and MSP (both use heat pump and octovalve).

The things to take away:
- A (really) warm battery redices consumption and increase the range by far.
- keep the car in a warm garage before the drive if possible.
- Charge as late as possible with as high power as possible to mane the battery as warm as possible.

I’ll ad a few datalogs/graphs from my computer later, that shows the cell temp and heat pump behaviour.


*) Your MYP batt type is probably the same as my M3P had (2021 and newer, 82kWh panasonic ) so the temp limits is probably the same as my M3P had.

 

[AAKEE]

…

I guess I don’t understand the difference? Are you saying it’s just a display modification and the car will let you use that ~4% if you need it?

Yes.

The SOC the car measure is the same as a warm day.
Its very common to believe that the car actually measure a lower SOC due to the cold. But the OCV of a lithium battery is very stable and do not change much with the temperature. Also, depending om the SOC, the OCV can be both lower and higher when the battery is really cold.

I have a couple of years old post here on TMC where I took a panasonic 2170 outside from +24C, measured the cell voltage and had it at -26 or 28C for about three hours and then measured the cell voltage again. My voltage meter only shows hundred, and did show exactly the same. I did it again recently with a viltage meter showing mV.
From +24C to -7C there was a few mV lover on two of the three cells but one showed the same (they had different SOC).
(There also is several research reports testing this so its nothing new or undocumented).

Well. The car measure (virtually) the same SOC number independently of the cell temperature.
But as the car can display range this range is adjusted to (for example) the need to preheat the batterry, and the more heat losses the battery have at low temperatures. My guess is that it was the will of Tesla to show a more correct range when the range is displayed that made them adjust the range, and because that range and SOC is coupled, the SOC displays the equivalent difference less as well.
And, yes. You are not locked out from that energy but for example being parked with 24%, and showing 20% means that when driving and reaching 0% displayed you have also used the 4% difference (to heat the battery and due to cold losses like higher internal resistance etc.)

 

[AAKEE]

To see examples of the difference, we can look at the EPA applications and the charge depleting tests.
For my Plaid, it delivered 99.4kWh on the regular “summer test” (charge depleting highway) but only 93.5 kWh on the 20F charge depleting. Thats a 6% less energy delivered.
(But that loss helps heating the battery so cabin heating can use part of that loss = less than 6% real loss in range numbers).

As a side note, I recently saw a BMW EV test with huge loss in the 20F test.
I actually look at research reports to find if there is a difference between NCA and NMC that made this abnormous difference, but I have not found my data for this yet.

 

[AAKEE]

I am convinced that at highway speeds, there is enough heat recovered (1KW) to give the heat pump a COP of >2, I am thinking this would mean ~750 watts for cabin heating, compared to 20KW being burned up by the drive train to cabin loss is small.

The heat loss in the motors heats the battery.

The part that makes my head explode is how much heating do the batteries need at highway speeds (and say 10F)? If they are self sustaining due to internal dissipation, there should not be much heating required from the cell energy.

All heat energy dissipated is stored in the battery. The cooling circuits from the motors heats the battery and when the battery is warm enough, the heat pump use the battery heat to heat the cabin.

The battery is (passively) heated by the motors heat loss, which is used at a heat energy storage. This increases the energy output/ reduces the heat losses in of the battery.
The heat generated in the battery also contribute to the stored energy.

There is no active heating of the battery if you step in to the car and go driving with a coald soaked battery. But the battery will slowly reach acceptable temperstures.

I guess we need more than 1.5kW for csnin heating at 10F when driving at higway speeds.
A wall connected electric cabin heater with 1.5kW to heat a parked car might be able to barely heat the cabin in the parked car at 10F, but that is heating the same air (vs when driving and exchanging the air).

Also, the cooling effect on the windows and the big glass roof will increase the energy needed.

At temps not colder than about 10C ( ~15F) the heat pump is able to use the outside air to heat the cabin, but probably not with the best coop.

This is a picture from Teslas patent:

Yet I still see many comments that cold weather loss will be 20-30%+. Are there any charts on cell capacity (range) vs temp (with heatpump). Certainly cell pre conditioning for departure will be a factor, at least at the start.

As I wrote earlier preconditioning the cabin only heats the battery if needed, and only to a about 5C above freezing (~ 40F).
It takes the battery up from the ineffective es being very cold, but on a longer drive you still would get the battery warm enough.
Starting the drive with a battery as warm as you poosibly can get it, will reduce the consumption, specially below 15F where the heat pump can not use the ambient air yo any good.

 

[AAKEE]

Here's a log from the yesterdays drive.

Cell temp -16C ( ~2F) after a cold night. It was 24C in the morning and I was not able to precondition completely, due to not knowing when we would leave, and that it was 200km to the SuC (and, not being able to charge full due to only using UMC with 3kW power and another EV needing to be charged as well.)

I started the cabin heat 18 minutes before we drove awy so the cell temp was able to rise from -16.8C to -1C when we drove away. After 1.5hrs the cell temp had rised from -1 to 12C. After this the preconditioning started due to the planned Supercharrting session.

The heat pump most probably have a fixed power, so it can not run on the heat losses produced by the directly, as these most probably is to small to make the heat pump run efficient. This is the most probable reason for the heat pump to let the battery go up in temp and then take the energy in a shorter time period than using it all the time.