MASTER THREAD: 2021 Model 3 - Charge data, battery discussion etc

[AAKEE]

Too late to edit my post?( 1hr limit?)

From my experience with lithium batterys in other applications the balancing always is done on top of the SOC, thats also the aim of the balancing, to have similar SOC and voltage( = performance) from all cells.
If you dont go to the absolute zero SOC, of close to it, the only real gain is to balamce it on the top of the range, to reach as close as 100% on all cells. Charging must be stopped when the highest cell reach 4.20V. If The imbalance causes the average Voltage to be 0.1V below the sibgle highest cell at 4.20V, we loose more than 10% capacity.
For cycles between below some 90% and below there might be less advantage of balancing the cells, a balancing session at low SOC might even cause more imbalance at higher SOC’s, because of different cell characteristics.

Well, knowing this, I arrived from a two day trip, about 600km.
My battery has shown 4.00mv imbalance for more than two months with quite shallow charged. I never have seen less than 4mv, so I leaning to the guess that 4 mv is the target when the BMS stop balancing.
When I arrived at home yesterday, I had 20% SOC, 3.47v / cell 6mV imbalance and 16.75C batt temp. I put the car in the garage and waited 5hours.
Still 20% SOC, 3.48v/cell and still 6mV. Cell temp 15.75C.

No balancing seem to have taken place during these 5 hours.

I put the car on charge to 62% with the WC at 5.5kW and by calculation from both charge time remaining and also the Kwh needed/ charging power I did get a time early this morning, I added 5hrs and that time just passed so I checked the imbalance. Still 6mV.

As it looks, no cell balancing seem to have been occuring during the charging session or after during these 5 hours.

So, the car is now on charge for a 90% SOC, I will check the imbalance when the charge is complete and also leave it for 5 hours and check.

Cell imbalance should not be a BMS estimate but actual values, so I think we can rely on these voltages as ’true’.

Charging just finished at 90% SOC.
6mV balance all the way up to last check point 73% SOC.
I saw 4mV at 87% SOC.
I sat in the car for the last 10 minutes of the charge and SMT showed alternating 4 and 6 mV during that time. When it stopped charging, It said 4mV.
Now the car will stand for a few hours, to see the actual imbalance after this.

 

[Candleflame]

Too late to edit my post?( 1hr limit?)


Charging just finished at 90% SOC.
6mV balance all the way up to last check point 73% SOC.
I saw 4mV at 87% SOC.
I sat in the car for the last 10 minutes of the charge and SMT showed alternating 4 and 6 mV during that time. When it stopped charging, It said 4mV.
Now the car will stand for a few hours, to see the actual imbalance after this.

should have charged to 80 and 85% to see if it balances then too! how much % imbalance would 5mv cause? cant be that much, maybe 1-2%?

 

[AAKEE]

should have charged to 80 and 85% to see if it balances then too! how much % imbalance would 5mv cause? cant be that much, maybe 1-2%?

I didnt have the time for that step this time. Seems that no balancing occured below 73% at least.

5mv is not much.
10mV might be about 1% on a single cell

 

[Devils son]

Anyone, but you, obviously. I just did a test run on my way today, will present the results shortly.

I don’t really care for your testing and I was talking about the (S)COP of heat pumps in relation to the European energy label. Which fully contradicts your statement regarding maximum achievable COP of heat pumps.

 

[TimothyHW3]

I don’t really care for your testing

Yes, people like yourself usually don't care for actual tests, facts or anything related to the way things work.

They have "oppinions" and spam us with those...

For anyone else who relies on actual facts here is the test:

1. The heat pump is pulling around 5.5kW when there is nothing to scavenge from and in the beginning. The car was driven the day before, hence there was still some leftover heat in the cells. Start of the run was at -1C or 30F end was 4C. There were no cars on the streets so very little stop and go (actually none, so even better for the test. On a real working day with cars and stop and go the car would've pulled more)

2. The heat pump is scavenging heat from the battery even at low cell temperatures (which is not very ideal, since there is higher internal resistance, we have seen this with the drive tests), but still puling around 3kW into the drive.

3. For any short drives below 20 minutes, mostly with stop and go city driving, the heat pump hence has zero effect. Or very minimal.

4. I have shown in previous videos of the older 2019/20 models, how you can drop the consumption of the PTC even below these heat pump numbers I got today. You preheat the cabin for 5 minutes, turn on the seat heater and turn off the AC, put heater on manual 3. It will then be pulling below 2kW.

5. When the battery cells are very cold (below 5C) then there is nothing to scavenge so the heat pump is not very efficient.

6. The most efficiency of the heat pump system comes from the battery scavenging. Which is being proven by the canadian test and by the VW ID tests done recently. I guess the VW system is not as good as the octovalve.

When there is nothing to scavenge from (battery is cold as in the canadian test), and the temperatures are below 0C, there is almost zero gain (if not zero)

 

[AlanSubie4Life]

I don’t really care for your testing and I was talking about the (S)COP of heat pumps in relation to the European energy label.

The only way anyone will be able to contradict @TimothyHW3 ’s data-backed claims with owner data is to actually take careful data. There are reasons why he gets the results he gets - he’s not making it up. You need to record temperatures, vehicle speeds for entire trip, etc., etc. However, I am not suggesting that doing so would convince Timothy.

It’s probably true there are situations where the heat pump provides no advantage. I am not sure whether it can be worse - that depends on how effectively the resistive heat from the heat pump (when in that mode) is transferred to the cabin as compared to the PTC.

But there are certainly situations where the heat pump will do far better.

For example: Tesla’s run of the EPA 20F FTP test shows that the 2021 gets much better efficiency than the 2020 under those specific test conditions. This data taken under standardized conditions is readily available for all to see, in the public domain; no secrets. People can draw their own conclusions about when those specific test conditions will apply to them (which will not be all the time!). What’s important about that datapoint is it shows definitively that under some conditions, with a cold-soaked 20F battery, the heat pump is more efficient than the PTC.

 

[AAKEE]

For anyone else who relies on actual facts here is the test:

1. The heat pump is pulling around 5.5kW when there is nothing to scavenge from and in the beginning. The car was driven the day before, hence there was still some leftover heat in the cells. Start of the run was at -1C or 30F end was 4C. There were no cars on the streets so very little stop and go (actually none, so even better for the test. On a real working day with cars and stop and go the car would've pulled more)

Air to air Heatpumps in general can give some COP of 2 at -7C and 1.8-1.5 at -15C.
Theres a good chance this heatpump is not as effective as the best air-to-air heat pump. The heat pump efficiency depends on the source temperature and the outoput temperature(to lower output temp the better COP).

I have not turned mine M3P in for fioxing the heat pump issues. I lost heat totally once when it was one week old and I had a bit variyng heat results dirung january. After this things settled, with stable temps every day. (Our coldest month is february with daily -15 to -25C and as cold as -35 to -40C in the dips, but this year we only had about -30 as the lowest point.)

There was some kind of software fix, which made the heat pump go into some kind of emergency mode. Easy detected by listening to the heat pump(more angry noise), vibrations that could be felt in the steering wheel and also very increased consumption. SMT showed batterty power about 6kw or slighlty more.
I had the oppurtunity to do a lot of drives to work and back with the heat pump both working normal and not normal. For -7C I had a couple of trips both ways with the heat pump in both modes.
Its a 47km drive on diffgerent speeds, not much och real highway but mostly 80/90km/h and some 100/110km/h but less amount.
The car sits in the garage every night and outside at work, not connected to any charger.

When the heat pump was in normal mode I had about 165-170wh/km both ways(not a big difference between the garage and outside, but I use planned departure so cabin and battery preheated).
When the heatpump not worked as it supposed to, 230-240Wh/km.
Sincve it got hotter, I think the heatpump hasnt been in the emergency mode anymore.

I would think that the heatpump use the best available source at every moment so if battery is cold but its some +/- 0C outside, it would use this heat and still get good efficiency.

Yesterday we left my mother in laws house, the car was outside during the night(parked for about 18 hours, it was freezing temps during the night and 2C when we left, this drive was 200km to a stop at IKEA and then another 130km.
I did start ”preheat cabin” before we left but just for some 10-15 minutes, as I wasnt aware of my wifes IKEA plan, and need to time the opening hours. I didnt check values with scan my tesla but I cannot think the battery was that hot. I did a partial charge during the night with the travel charger with 2kw charging. I think the charging was done a couple of hours before we left.
Still, I had really low consumption from start and it really didn change up or down during the 200 km drive( about 156wh/km), temps was 2-3C and a slight headwind/sidewind.

 

[AAKEE]

So, so far its been 3 1/2 hrs since charge to 90% was complete. I just couldnt wait 5 hrs as the other times. Already down to 2mV imbalance.
Ive never charged over 90% and had it parked afterwards. Ive also never seen a imbalance below 4mV.
For all my driving for 3 months, Ive charged to 80% the colder period, 70% when the coldest days was gone and since a couple of weeks to 60%.
The imbalance has always been 4mV, I was keen to believe that the BMS stopped balancing at 4mV delta. Just checked some 30 print screens from SMT, always 4mV if not during drive or home with low SOC, and a few high SOC charges with more than 4mV.

So, from this and my daily charging routine with 60/70/80% SOC I can say that it sure do not balance below 4mV these times( probably no balancing at all).
And just 3 1/2h with 90% the imbalance is down to 2mV.

 

[TimothyHW3]

Air to air Heatpumps in general can give some COP of 2 at -7C and 1.8-1.5 at -15C.

Have you seen the size of the Tesla heat pump and where it is located inside the car? There is not even a distinct comparison between this and a high end home heat pump. I also doubt that there is 1.8C at -15C, but since I can't proove it I will be left with the doubt.

Actually my heat pump, car built November 2021 never had issues and it gets the cabin warm pretty fast. But I guess there might be software variations. The point is that when there is nothing to scavenge from and the outside temps are negative C there is simply not enough benefit on short commutes.

And with the old system, if you go smart about it, you can actually achieve better results. Also a thing my car does - when I put it on manual it pulls more kW than in Auto, same setting, where it was the opposite before that with the PTC on the 2019

 

[AAKEE]

Have you seen the size of the Tesla heat pump and where it is located inside the car? There is not even a distinct comparison between this and a high end home heat pump. I also doubt that there is 1.8C at -15C, but since I can't proove it I will be left with the doubt.

Absolute!
We are not thinking different about the heat pump most probable COP values.

I think that the octovalve can use a combination of battery heat, engine heat ambient heat, and always try to use the best available energy?

I have a refresh Performance, and my collegue has a 19 LR. We commute the same way to/from work, both cars in garage at night and outside at work.
Im on 19” studded tyres on aftermarket wheels and he use 18” studded on stock wheels with the aero caps.

What we experienced this winter is that my car is much more efficient in cold wx.
We did a comparison the other week in about +5C, and I think my values for that drive was some 10wh/km lover.

I have no values for short drives with cold battery as my car stands inside the heated garage at home.

 

[Devils son]

The only way anyone will be able to contradict @TimothyHW3 ’s data-backed claims with owner data is to actually take careful data. There are reasons why he gets the results he gets - he’s not making it up. You need to record temperatures, vehicle speeds for entire trip, etc., etc. However, I am not suggesting that doing so would convince Timothy.

It’s probably true there are situations where the heat pump provides no advantage. I am not sure whether it can be worse - that depends on how effectively the resistive heat from the heat pump (when in that mode) is transferred to the cabin as compared to the PTC.

But there are certainly situations where the heat pump will do far better.

For example: Tesla’s run of the EPA 20F FTP test shows that the 2021 gets much better efficiency than the 2020 under those specific test conditions. This data taken under standardized conditions is readily available for all to see, in the public domain; no secrets. People can draw their own conclusions about when those specific test conditions will apply to them (which will not be all the time!). What’s important about that datapoint is it shows definitively that under some conditions, with a cold-soaked 20F battery, the heat pump is more efficient than the PTC.

My data is carefully taken and analysed. If you take two weeks to do the commute with both cars, switching them back and forth, in comparable circumstances, then you have 10 trips in both cars. If you then have one car that will have trips ranging from 185-195 wh/km and the other car from 150-160, it is safe to say that that car is more efficient during those circumstances. Both being on 18 inch and wintertires.

But that wasn’t even the argument I made. I said that the overall efficiency of the MY2021 is better compared to the MY2019. A conclusion I do not only base on my data, but also the EPA and WLTP, plus common sense that adding a heat pump is going to make the car more efficient. That is exactly why Tesla made that improvement.

The EPA and WLTP are tested professionally, and frankly I have more confidence in those numbers than someone that is testing on youtube.

Call it professionalism.

Timothy is clearly not interested in other data, he needs to be right. Well, he isn’t. I am giving up on putting some sense in his believes regarding the efficiency of heat pumps, especially at freezing temperatures (which are not common where I live, though we did have very cold days this year, and the MY2021 still was more efficient compared to the data I have regarding the MY2019 that I have).

What Timothy might have proven is that under certain circumstances the MY2021 is in fact not more efficient. But that is irrelevant. Because I am taking about overall efficiency under real world conditions. Why would I care how my car behaves at -10 degrees Celcius without heating on? Who does that?

Interesting data if that is what you’re researching, but does not counter my simple statement:

MY2021 is definitely more efficient. Even the Performance 2021 is more efficient then the LR 2019. And has the bigger battery.

If Timothy chooses to think otherwise: great. Be my guest.

He could have already checked my statement regarding SCOP and the energy label. If he can’t even admit that he was in the wrong there, there is no data from my testing that will ever satisfy him.

Which is ok. I didn’t pick the discussion.

 

[TimothyHW3]

Timothy is clearly not interested in other data, he needs to be right. Well, he isn’t.

Said the joker who is yet to present single piece of data... And me, the one "who isn't right", actually presenting said data, on MULTIPLE occasions. Backed up with actual live videos...

He could have already checked my statement regarding SCOP and the energy label

Nobody is talking about heat pumps in houses. If again, you read what is being written, the heat pump in the car is so efficient, because of the heat scavenging from the battery and from the motors, not so much because of the heat pump itself.

And I actually look at kW usage, thank you very much. I am sure that for someone like you, who doesn't understand simple data points, an "Energy label" is more telling, but actual kW usage is what counts here for a comparison, not what a random EU commision thought was "good".

Why would I care how my car behaves at -10 degrees Celcius without heating on? Who does that?

Nobody, this is why the test was actually done WITH THE HEATING ON. AND THERE WAS NO EFFICIENCY GAIN!11! Because there was nothing to scavange from and the air is too cold. Physics, my friend, physics.

Oh I forgot, you don't care about "Youtubers testing on Youtube".
You call it "professionalism".

But then you quote your God Bjorn Nyland as a point of reference (on two occasions).

Contradictions, contradictions...

My data is carefully taken and analysed. If you take two weeks to do the commute with both cars, switching them back and forth, in comparable circumstances, then you have 10 trips in both cars.

I think you just contradicted yourself within two sentences. Switching cars back and forth is in no way reproducable. Even the slightest variation in temperature or cell levels will scew the results by a long shot...

Plus, if you actually listen or read what I wrote - the old PTC system is very ineffecient if you put it on Auto, because A) Tesla pushes everything to heat up the cabin in short time B) turns on the AC for no reason.

So if you have a 2019/20 car and you put in on Auto and you had let's say 2-3C variation, sure, you might even get 20% difference in some cases.

If you manually adjust the controls like I explained (which again, backed up by data and valdiated with tests) you can lower the consumption of the 2019/20 car down to negligable numbers and on par with the heat pump (even better in some cases, because it seems the heat pump can't be fine tuned like the old PTC)

Actually, I believe that I can even make the 2019/20 car more efficient in a short commute below 10 minutes if I put it on manual 2, turn on the seat heater. Because then the total combined pull will be around 1.5kW (again, tested in previous videos). And for some reason, I can't get the heat pump to this level as in manual it actually pulls more than in Auto.

And in Auto the heat pump does all these weird shenanigans by pulling 5.5kW from the already cold battery.

But once again - you are not interested in data or facts. Man, some people...

 

[AAKEE]

Teslabjörn camped with a M3P refresh for tre days, temps between -3 an -1 or about zero. In the beginning the battery had heat to share, but not that long. Initially the battery draw( including the heatpump) was some 300watts but after the battery cooled the total draw was about 1000Watt(1 kW). If you try to heat a car with a passenger compartment heater as we do in Sweden, 1000W will not keep a car that warm at -3. 2000W normally feels fine if not too cold outside, I guess the heat needed to keep the cabin 21C is in the ball park of 2kW.
Teslabjörn heat pump test

 

[TimothyHW3]

Teslabjörn camped with a M3P refresh for tre days, temps between -3 an -1 or about zero. In the beginning the battery had heat to share, but not that long. Initially the battery draw( including the heatpump) was some 300watts but after the battery cooled the total draw was about 1000Watt(1 kW). If you try to heat a car with a passenger compartment heater as we do in Sweden, 1000W will not keep a car that warm at -3. 2000W normally feels fine if not too cold outside, I guess the heat needed to keep the cabin 21C is in the ball park of 2kW.
Teslabjörn heat pump test

I believe that there was also a test where he drove the car, not sure if the P or the 21LR and he showed how inefficient the heat pump was in these conditions. Not sure if it was a live video or another video, can't bother to search for it.

But I don't see what else he can test, basically the same car. And my car doesn't have an issue with the heat pump so probably as good as it can get. The canadian guy also confirms my tests, so it seems pretty confirmed to me. 3 people on 3 continents.

Most of his other tests he runs after the car is warm and he is just finished charging and mostly HPC fast DC charging, which gets the battery to 40C+.
This is something most viewers don't understand. This to me is a very, very unrealistic test. Nobody supercharges the car every day before they go to work.

To me the most interesting thing from my test is that the heat pump will scavange the heat from the cells even at 6C. I don't really like that. While still pulling 2-3kW. The other thing is that you can't really control it.

With the old system you can put it on 2 manual, turn of the AC and turn the seat heater on and get below 2kW.
Sure, you wouldn't get it as fast as warm as with the heat pump, but it was good enough.

By the way, for anyone who is just not into the technicalities - the heat pump of course has its advantages, I have tests to show that one too. But in very specific conditions - like if you have heated Garage, the cells are warmish, the outside temp is in the +5C/10C region etc. Unfortunately not for short commutes where the battery is very cold and the outside temp is negative C.

 

[TimothyHW3]

What’s important about that datapoint is it shows definitively that under some conditions, with a cold-soaked 20F battery, the heat pump is more efficient than the PTC.

The EPA chargers the cars to absolute full right before they start their tests. This usually brings the car to positive 10C/15C 60F on its own. I don't see how this could be a "cold-soaked 20F battery"...
To get the battery to -6C you have to leave it out in the open for many many hours to days with negative degrees. I had the car outside in Norway at negative -15C and the cells barely went to 0C in the morning...

I don't know how they manage to get the cells down to 20F, did they put it in a freezer?

 

[AAKEE]

And in Auto the heat pump does all these weird shenanigans by pulling 5.5kW from the already cold battery.

Id say, when you see 5.5-6kW power draw your heat pump is in the same emergency mode( or maybe direkt mode, 1:1 energy in, heat out) and there isnt any heat drawn from the battery. Energy in form of Electric power, yes but not picking heat up via the heat pump function.
This 1:1 direct mode is described as one of the modes, supposed to be used when the heat from other sources is not enough to keep the cabin warm. But, at least for me, I have had the heatpump working fine without the angry 5.5-6kW mode at below zero a lot of times. Down to -10C (at least) it seem to be able to work very effective.

I think maybe you have seen the elevated consumption because the heatpump goes into the emergency mode because it has the defective pressure sensors?
my car is also built in november and I know for sure I need to get it fixed. If you wherent too unlucky you didnt have the heating to stop completely buth still you could have seen the 1:1 mode covering this issue.
I guess you did not have pressure sensors changed( two or three, heat pump system parts) ?

 

[TimothyHW3]

heat drawn from the battery. Energy in form of Electric power, yes but not picking heat up via the heat pump function.

Well, actually if you look at the video, the cell temp was dropping from 6.5C to around 5.5C. So I guess it might be still pulling something from the battery.
But yes, most likely just some PTC 1:1 equivalent.



"on because the heatpump goes into the emergency mode because it has the defective pressure sensors?"

No, it dropped to about 3kW right after 1 minute or less and it continued to drop to about 2kW. The 5.5kW was just from the cold cabin, cold car, cold cells and heat pump firing up. And I guess after the 20 minutes mark it was to about 1.5kW, which is still pretty okish.

What exactly is the typical use case when the sensors fail - what do you experience? I haven't had any heat pump outtages, even in Norway at -15C for days and the heat comes pretty fast normally.

What else is a typical indication of these sensors not working correctly?

 

[AAKEE]

What exactly is the typical use case when the sensors fail - what do you experience? I haven't had any heat pump outtages, even in Norway at -15C for days and the heat comes pretty fast normally.

What else is a typical indication of these sensors not working correctly?

I have been in contact with tesla SC about it, still 1000km one way to the nearest SC so I have postponed the fix. Its supposed to open one, ”only 300km” (one way) away soon.

The most common problem seem to have been the heating just stops working. There have been a lot of these cases in Sweden. Also, Tesla seem to still ship cars with the faulty sensors. I did read about a new car delivery in a facebook group this week, heting broke down during delivery.

I had the car about on or two weeks before the heating stopped. So I had the time to learn how it is supposed to work, and could see that they probalby made som changes via software fixes. After a period of unstable heat(not stopping totally but big variations in heat when it was -15 to -30, there was a software update and after this the cabin temp is stable. But from time to time, when it is cold outside the heatpump goes into this angry noise mode, 6kw on battery power when stationary in SMT and also vibration in the steering wheel. I dont feel it at 100km/h but in lower speeds I can feel it after learning how it feels stationary. When it does the angry mode there is no heat drawn from the battery. I had 12C as a battery temp that was reached after leaving the garage or after starting coild at work(if warmer it went down, and if colder it went up), and when the angry mode was on, the battery got hotter than the usual 12C when commuting between work/home.

I absolutely do not distrust your experience. I believe you. But also, my experience do not match to 100%, thats the reason for asking about the heat pump fix. As it seems, most cars is affected. Tesla SC talked about one or two high pressure sensors, and alos one low pressure sensors that needed to be changed. This was without looking at my car, just from some internal information letter about the issue. There is a lot of info about this on the net.

 

[TimothyHW3]

I don't have any of these things so far, 6000km down an temperatures anywhere between -15 to +20C.
Heat pump is working the same way.

I know there is a silent recall for the heat pump, but so far it works good on mine.

Never touch a running system.

 

[Devils son]

An air to air heat pump has a maximum 2.5 at optimum levels of above 5°. I know because I own one. At -7 it is close to 1, yes. And this for big ass household heat pumps.

Nobody is talking about heat pumps in houses.

Except, it was you...

Nobody, this is why the test was actually done WITH THE HEATING ON.

However, this is you...

Oh yeah, a link to that test where they used the same exact cars with heater off

Switching cars back and forth is in no way reproducable. Even the slightest variation in temperature or cell levels will scew the results by a long shot...

Actually, this is normal in scientific testing. You get multiple data points and depending on number of data points and their consistency, the fault margin decreases. With the data I have, it is impossible that the 2019 LR is in fact more efficient then the 2021 P in those conditions.

And I never use the auto setting for the HVAC, so I don't know why you are going on about that.

It is very simple: you have your data and only you can verify that, I have my data and only I can verify that.

If you say that there is no difference in efficiency during your testing: fine. However, it is unclear what you are actually saying, because you then talk about the efficiency of the motors, then efficiency difference because of heat pump. So I am truly lost, because you really lack consistency. And mainly because you seem to be talking about different things.

Whereas my comment was simple: overall the MY2021 is more efficient. And that is exactly why Tesla decided to use a heat pump. To make the car more efficient.