Battery Heating explained

Battery Heating explained

So most of my reference data is and Bjorn's videos. I will work to get reference data that will show the specific scenario as written below(FACT paragraph). I am also working to attempt to get data showing heating power vs speed but that is not an easy or quick test.

Ok so here is an explanation of "preconditioning"/"battery preconditioning"/"On route battery warm up".

So all the above terms I am going to simplify as they all mean the action of putting non-motive power to the motor to create heat which then is transferred to the battery through the coolant system. Now the different scenarios as to when this function is activated create differences in how much non-motive power can be applied to the motor(s), and/or what goal temperature the system applies for the battery.

Battery Preconditioning(BC) general description: The Model 3 has the ability to put a non-motive power waveform to the motors in order to generate heat in the Stator(motor). The maximum amount of power applied to each motor is approx 3.5kW. The heat generated is then passed via conduction to the coolant flowing through the motor. This hot coolant then flows to the Superbottle(coolant reservoir). The hot coolant also flows from the superbottle to the battery, and then back to the superbottle.

Lets go through a few scenarios where BC is automatically engaged:

Routing to a supercharger:
When you route to a supercharger and the battery is less than approximately 30°C, BC will turn on and you will get a note on screen that says it is preconditioning. This BC is intended to warm up the battery in order to allow for maximum Supercharging speeds since charging a cold Li-ION battery can damage the battery. When routed to a Supercharger, BC will come on automatically whenever the car feels like it. I haven't found a good correlation on time or distance or variations thereof in reference to battery temperature. I have seen it come on automatically while being over 50 miles away from the supercharger. Now, just because the car "says" it is BC, doesn't mean that it is actually applying the whole 3.5kW(RWD), or 7kW(AWD) to the motor to generate heat. The heating power is greatly reduced when the motor(s) are being used for motion/regen. In a RWD vehicle, the BC power is greatly reduced when going over 40mph. In an AWD vehicle, the BC power to the rear motor is the same as on a RWD vehicle, but the front motor can still get the full 3.5kW as long as it is not being used to move the car. In general cruising on dry roads, the front motor in an AWD vehicle is barely used up to ~90mph.

Preheating Cabin using phone APP:
When the car/battery is cold and you engage climate control from within the Tesla app on your phone, whether you set a temperature, or hit defrost and it goes to max temp, BC will also be engaged(depending on battery temperature of course). The reason for this BC is presumably to help gain some regeneration capability at the beginning of your drive. It would also of course help supercharging speeds if you have to hit a supercharger at the beginning of your drive. I don't know what the target battery temperature for this is yet. On a RWD vehicle, it will apply the 3.5kW non-motive power to the motor to generate heat just like the first scenario above. I would presume that it would apply the 3.5kW to the front motor as well in an AWD vehicle.

So general questions relating to this have been, what is the fastest way to get rid of the regen dots. My general answer to that is, get BC turned on. Heating the cabin in and of itself is going to do practically nothing(this would be getting into the car and turning on climate control and just sitting there.) Turning climate control on via the app should do it, sitting in the car an routing to a supercharger should do it, getting in and setting climate controls and then turning on climate keeping and getting out will do it as well. Now the time to heat the battery up has not been determined yet but that is also going to depend on how cold the battery is in the first place.

FACT 1: In an RWD vehicle, during preconditioning at a standstill, the stator temperature can get from ~24C to over 55C in 10 minutes at ~0C outside temperature. At this point if you start driving at highway speeds, with preconditioning still activated you will see the Stator temp and battery inlet temperatures go DOWN. This indicates that even though the car is still showing a preconditioning condition, it is not actually working.

FACT 2: When you go to a supercharger and your battery is still cold, the car will continue to use BC in order to heat the battery. This is shown by the CANbus messages showing power going to the motor, the stator heating up, and an increased coolant flow rate through the powertrain and battery.

FACT 3: When setting Keep Climate ON in cold weather, after leaving the vehicle, the car will set the battery target temperature to 30C and battery heating will commence as needed.

So I just captured some data on 1/13/2020 in my LR RWD vehicle. Graph is here The graph starts as I am leaving my driveway and ends approximately 5 minutes after I leave the supercharger. I did not stay long at the supercharger, and the 2 little spikes while I was there were me switching stalls in order to ensure the power input was stable no matter what stall I was in. I engaged BC near the beginning as I was going 60 miles per hour. The whole point of this exercise is to show that as you speed up, the preconditioning goes away. This graph shows that as speed increases, the stator temperature levels off and/or goes down. I will also note, that if I had done this trip without BC on, my stator temperature would probably not have gone above ~25°C. Questions welcome. Ohh and yes the radiator was bypassed 100% the entire time.

Ron.Olsberg | 8 January 2020

The Scan My Tesla App should be able to verify all of your assertions in this post.

jallred | 8 January 2020

Thanks derotam,

I noticed that in the summer, my car would start preconditioning when about 10-15 highway miles away from SC. Last week with temps around 40F, I saw preconditioning come on when about 50 miles away.

I've never noticed any range impact with preconditioning on. I'm RWD, so I think that supports your thesis that the RWD doesn't generate enough heat to sufficiently raise battery temp for SC during highway driving.

I think another factor in this is that the front motor on AWD 3 is an induction motor, but the rear (and only motor in RWD) is a Permanent Magnet Switched Reluctance Motor. I'm guessing that it is generally harder to generate excess heat in the PMSRM.

Again, thanks for your insight.

derotam | 8 January 2020

@Ron.Olsberg: The Scan My Tesla App DID verify all that I posted. :) I have it and an OBDLink MX+ and the appropriate wiring harness.

derotam | 8 January 2020

@Jallred: Well any differences in the motor type would be a TBD. As far as from a standstill they still pull the same amount of power. Now I guess you could look at the temperatures between the motors and look at the difference. I don't have an AWD to look at though.

Bighorn | 8 January 2020

From distant recall, optimal temp where heating turned off was in the 30s C, so your example of no heat in a 55C situation obviously makes sense. And Preconditioning messaging may refer to cooling in that circumstance.
As you’ll recall, my early empiric data showed approximately 4 kWs of draw while preconditioning my P3D.

I also thought the process was based on the PM motor, so am curious about the role of the older motor in heating. Bjorn is not someone worthy of being a primary source, so I would not predicate any conclusions on his thoughts.

derotam | 8 January 2020

Heating may turn off at 30C for the BATTERY but I never got even close to that. The stator continued to heat past 55C in the 10 minute test. The battery temp and battery inlet temp barely moved in that time period.

Bjorn has shown a total of ~7kW for battery heating in his AWD vehicle. 3.5kW on each motor at standstill. Nothing is predicated on Bjorn's thoughts, it is based on ScanMyTesla CANbus data shown live in his videos. I don't take his word for anything, I look at the data he is showing from the CANbus live.

jallred | 8 January 2020

It's all a big differential equation for heat in and heat out. A differential equation that never has to be solved. Instead it is all done empirically.

It looks like when it is really cold outside that the heat loss of the battery to ambient is greater than the heat in from the RWD motor excess heat.

It is my understanding that the new PMSRM for Model 3 is more efficient. Same thing as saying it is harder to suck excess heat off of it.

It is a bit perplexing that the battery inlet temp doesn't show a big jump. You would think that if you are moving the heat from the motor to the battery that the inlet would show a high temperature. Guess it would be good to look at radiator temps as well to see if for some reason they would heat up the motor but not push all the heat to the battery.

Also, we are assuming the battery temperature set point is a constant number. It most likely is different based on either SOC or predicted SOC upon arrival at the SC.

Bighorn | 8 January 2020

Gotcha about the battery vs stator temp.
7 kW is also what the reddit test found while still, but immediately dropped to around 4 kW when D was engaged. Still nobody has corroborated the 45 kW "data" seen in Portland.

derotam | 8 January 2020

Yes, 45kW is still elusive.

jfaubl | 8 January 2020

I appreciate all the work you are doing but for us non engineers can you give us a simple sentence like precondition your battery for 10 minutes below 40 degrees. I don't know what a stator is or if 7 kW is good or bad.

Thanks for your hard work data gathering.

cmh95628 | 8 January 2020

@Bighorn, re: Still nobody has corroborated the 45 kW "data" seen in Portland.

Me thinks those are really air quotes around the word data... ;-)

Bighorn | 8 January 2020

Is there such a thing as hot air quotes?

derotam | 8 January 2020

@Jfaubl: there is no simple statement about it.

cmh95628 | 8 January 2020

@Bighorn. Not that I know of...

FISHEV | 8 January 2020

"Still nobody has corroborated the 45 kW "data" seen in Portland."

That was actually a number you fabricated from the empirical data provided and proofed via pix of the Energy Graph and Projected Range. I only showed 20 miles of range loss from Precondiitioning on 38F day, you then got all twisted up, claiming a dedicated 7kW heater and then finding out there was no dedicated battery heater at all...most amusing as you then went on a rampage deleting all the threads where you put out the two clangers...the fictional "45kW" and "7kW resistive heater". It's always the cover up that gets you.

Always good to keep that in mind for some perspective on your pronouncements.

No back to the blind men and the "elephant", the Model 3.

FISHEV | 8 January 2020

Ah...found the original post of "45kW"...Shocker!


Left the house with 106 miles of range for an 11 mile run to the Vancouver WA SC. Plugged in the SC in Nav to activate the Battery Preconditioning with 73 miles showing on range.
Travel 11 miles to SC and as you see from the arrival pic of the Energy Graph, it took 33 miles of range to go 11 miles with Preconditioning on. Heat was on but range is calculated with use of heat/AC so that should not have that big an effect.
Because the graph only runs when the car is moving because it is Wh/mile, the graph may be understating the power usage. As you see on the pic, the projected range with Battery Preconditioning on is 41 miles of range vs. the Rated Range of 73 miles so it i suggesting even higher power usage.
You can also see the energy usage is off the charts (900 Wh/mi max)as PC peaks.
Car will be sitting 10-12 hours at work in cold weather and the Woodburn SC is 18 miles from work. I’ve often run into Woodburn with 2-20 miles left and battery in the “red”. I’ll have to be more careful in winter and make sure I’ve got at least 50 miles left on battery as PC could use 18-36 miles per these numbers. It looks like PC peaks early and runs for about 5 miles so the draw on an 18 mile run might be the same. News at 11 on that.

andy.connor.e | 8 January 2020

the non-human response generator is back. it was probably supercharging its battery

WW_spb | 8 January 2020

Lol Seagle is back. Welcome back AI

FISHEV | 8 January 2020

"it was probably supercharging its battery"

Nope...driving the Q7 last couple days while FISHEV is getting front pumper fixed. You would not believe how cool the Blind Spot Indicators are...some day Tesla might have tech as advanced. Virtual Pilot is the bomb.

Bighorn | 8 January 2020

Funny he’ll never find posts corroborating the lies he perpetuates about me.

WW_spb | 8 January 2020

Bighorn. Don't worry we all saw original pre conditioning thread that Fish deleted bc he looked not very smart.

Bighorn | 8 January 2020

Obviously not the original because that said it was a ten minute drive, before it became a 30 minute drive. Curiously, this version has no time mentioned which precludes the ability to calculate a power draw. It’s amazing how someone can be so clueless about how to interpret data. Ripe for a flatbed.

FISHEV | 8 January 2020

"Obviously not the original because that said it was a ten minute drive, before it became a 30 minute drive."

Obviously was the and the boy's deletion has caught up to poetic. Look Ma' "43kW".

Bighorn | 8 January 2020
FISHEV | 8 January 2020

Glad you corrected yourself after the clangers.

Bighorn | 8 January 2020

11 miles in ten minutes. Where have I heard that before?

WW_spb | 8 January 2020

Fish did you even look at the dates? This was beginning of last year. Bighorns statement.

FISHEV | 8 January 2020

"11 miles in ten minutes. Where have I heard that before?"

In the three "Preconditioning" threads you started and deleted because of multiple clangers you can't man up to?

FISHEV | 8 January 2020

"This was beginning of last year. Bighorns statement."

No wonder he deleted all his recent was funny to watch and educational.

WW_spb | 8 January 2020

Don't we all love how he is trying to convince us that this new fixed post is his original one. When quick Google shows otherwise

Bighorn | 8 January 2020

Keep digging. It’s rather entertaining.

WW_spb | 8 January 2020

My Google search shows your thread buddy

Bighorn | 8 January 2020

And I must have had a stroke and changed my mind about the Model 3 having a battery heater when I derived the 4 kW draw empirically on mine.

WW_spb | 8 January 2020

"My Google search shows your thread buddy"
That was for Seagle

FISHEV | 8 January 2020

"My Google search shows your thread buddy"

Nobody's Google shows @Bighorns deleted "Preconditioning" threads. That he felt forced to delete them was precious.

WW_spb | 8 January 2020

Based on your logic you and your sock gang deleted the thread bc you look like a fool in them.

jallred | 8 January 2020

Fish, thanks for hijacking this thread. Go away.

FISHEV | 8 January 2020

"Fish, thanks for hijacking this thread."

All the credit goes to @Bighorn.

"Still nobody has corroborated the 45 kW "data" seen in Portland."@Bighorn

jallred | 8 January 2020

Sure. I give him credit. But you are continuing it.
You sound like a child “he did it first”.

EaglesPDX | 8 January 2020

Who is Seagle?

kevin | 8 January 2020

@FISHEV I do like the blind spot blinkers on my mirrors on my 2018 Audi Q5 for sure. I was surprised when my wife's new Model 3 didn't have that. Hopefully they add video to the screen for it before I order my P3D :)

FISHEV | 8 January 2020

" I do like the blind spot blinkers on my mirrors on my 2018 Audi Q5 for sure."

I like how Audi puts them on the inner molding of the mirror so it is more in the driver's peripheral vision than the ones mounted at the outer edge of the mirror.

The Rydeen BSI/RCTA system I installed on the Model 3 is similar with the indicator lights mounted right in front of the speaker on the upper door.

FISHEV | 8 January 2020

"Sure. I give him credit. But you are continuing it.

So are you.

FISHEV | 8 January 2020

But I'll stop...point was made and I do try and stick to just car talk.

FISHEV | 8 January 2020

But it does answer the "who is EaglesPDX" question....not FISHEV.

M-A-B-MCMLXXX | 8 January 2020

:: troll warning ::

Even fishev doesn’t believe its own lies. And before anyone else thinks to take anything at face value it spouts which is even hintingly technical, fish believes tesla cars can use the full specified power output of the motors the heat the battery.

ReD eXiLe ms us | 8 January 2020

PSA: FEESHICE is a Troll of long standing. He even denies his own posts when they are presented. He then starts his own threads, then complains of being 'censored' often when he himself deletes an evidenciary thread that fills with proof he is wrong. Not yo be trusted, he is best ignored and never responded to.

derotam | 9 January 2020

FISHEV Oct 20, 2019 in his OP "Battery Pre-conditioning uses how much power?"
"The Preconditioning message was on for the 11 mile, 10 minute trip. Taking out the 11
miles for the trip, the Pre-Conditioning looks to have used 34 miles of range, 8kWh of

8kWh in 10 minutes = 48kW power draw.

jfaubl | 9 January 2020

"@Jfaubl: there is no simple statement about it."
Thank for the reply. I have a 4 mile commute I don't worry about preconditioning the battery. I just turn on the heater for 5 minutes before getting in the car. I do like skiing and last year as a new owner I had some range anxiety and did the same thing at the ski resort. I didn't want to run the heater for 30 minutes thinking it would eat up too much battery and I wouldn't get to the supercharger. I wonder if there is a point where it is better to precondition vs not much preconditioning for energy usage. This winter we do have preconditioning before supercharging so maybe it doesn't matter. It's been warm so I haven't gone yet. I'll have to see what the car does at different temperatures.