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

[Devils son]

By the way @TimothyHW3, I haven't seen your data, because I really cannot watch all these YouTube video's. They are almost all a drag, take way too long to get to a conclusion and more often then not make conclusions that are not following from the data provided (more than 9 out of 10 times when I see some conclusion that really makes me unable to resist to check the video).

I did not get the data regarding the COP from the Björn video, but it was mentioned at Electrek.co website and then I still had to watch a small portion of the video to know at which temperate, so I could get a COP of 2.95 at 3 degrees Celcius. And that 1 minute was more then enough for me.

 

[TimothyHW3]

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

No, it is very simple indeed - I have actual data, which I present. From the BMS coming directly from the car. With kW and everything. The video is 2 minutes long go watch it.

You don't have any data and you haven't presented anything so far.

And on top of that you use TeslaFi, which is very inaccurate, because it bases their calculation on the rated range without actually factoring the way Tesla hides the buffer. So it is off by as much as a factor of 5%.

So yeah...

 

[Devils son]

No, it is very simple indeed - I have actual data, which I present. From the BMS coming directly from the car. With kW and everything. The video is 2 minutes long go watch it.

You don't have any data and you haven't presented anything so far.

And on top of that you use TeslaFi, which is very inaccurate, because it bases their calculation on the rated range without actually factoring the way Tesla hides the buffer. So it is off by as much as a factor of 5%.

So yeah...

So I have no data, and it is inaccurate. Really, make up your mind.

I never said my data is 100% accurate, I said that the samples are such that factoring the margin of error, it still shows that for that particular trip the 2021 P is more efficient then the 2019 LR.

Let me ask you a couple of direct questions, because you are still jumping from one foot to the other:

1. Would you agree that the heat pump is more efficient providing normal circumstances in a country like the Netherlands (temp between 5 and 15 degrees Celcius)?

2. Would you agree that if the circumstances are extreme (-20 degrees Celcius, or +30 degrees Celcius) the heating system from the 2021 Model 3 will not be less efficient significantly?

 

[TimothyHW3]

Would you agree that the heat pump is more efficient providing normal circumstances in a country like the Netherlands (temp between 5 and 15 degrees Celcius)?

2. Would you agree that if the circumstances are extreme (-20 degrees Celcius, or +30 degrees Celcius) the heating system from the 2021 Model 3 will not be less efficient significantly

I have no data, and it is inaccurate. Really, make up your mind.

Yes, in your case both are true.
You haven't presented any data so far(not a single comparison video or anything, only vague estimates by memory ...) your data points are flawed because not on the same commute and on top of that you use miscalculated data from TeslaFi as a reference point.

1. Depends on what you mean by "more efficient" and what the commute is(avg speed, heater setup etc.). If it is a very short commute, car is cold, cells are cold, under 15 minutes I would say no, not much MAYBE 5%. I have such a similar test of cold battery and I have to evaluate it.
But it was done on the highway and it was way longer drive at high speed. Then yes, it is probably 10% or so more efficient, when there is something to scavenge from while drivinf and the heat pump has enough time to crank up.

2. Again, not sure what you mean here, very vague terms.

Not only in extreme temperatures, but even in moderate negative ones the heat pump has zero effect, unless you have heat to scavenge. And if setup properly, on short commutes, the PTC can be made more efficient, by using smart manual setup of 2 with AC off, heater blowing towards you - so far my experience.

From the Canadian test it seems the PTC is actually better then, at below -10° on cold cells.

+30° you don't use heater just the AC so I don't see any advantage here. The old system was very efficient in the summer, below 800W so I don't see how much more efficient the new system can be.

I have a few non SMT tests here with 20°

 

[Devils son]

Yes, in your case both are true.
You haven't presented any data so far(not a single comparison video or anything, only vague estimates by memory ...) your data points are flawed because not on the same commute and on top of that you use miscalculated data from TeslaFi as a reference point.

1. Depends on what you mean by "more efficient" and what the commute is(avg speed, heater setup etc.). If it is a very short commute, car is cold, cells are cold, under 15 minutes I would say no, not much MAYBE 5%. I have such a similar test of cold battery and I have to evaluate it.
But it was done on the highway and it was way longer drive at high speed. Then yes, it is probably 10% or so more efficient, when there is something to scavenge from while drivinf and the heat pump has enough time to crank up.

2. Again, not sure what you mean here, very vague terms.

Not only in extreme temperatures, but even in moderate negative ones the heat pump has zero effect, unless you have heat to scavenge. And if setup properly, on short commutes, the PTC can be made more efficient, by using smart manual setup of 2 with AC off, heater blowing towards you - so far my experience.

From the Canadian test it seems the PTC is actually better then, at below -10° on cold cells.

+30° you don't use heater just the AC so I don't see any advantage here. The old system was very efficient in the summer, below 800W so I don't see how much more efficient the new system can be.

I have a few non SMT tests here with 20°

Well, it cannot both be true, because non existing data has no characteristics.

And apparently you’ve missed my post yesterday 9.23 pm.

Furthermore you are once again focussing at specific situations, whereas I just made a statement in general.

Why?

 

[AAKEE]

Not only in extreme temperatures, but even in moderate negative ones the heat pump has zero effect, unless you have heat to scavenge

There is always heat/energy in the ambient air outside. Some information say the heatpump only use energy from the outside air down to -10, but I do not really think thats true. I think its use the energy byt with a gradual decrease in efficiency.

The lowest COP you will get is 1, and then it is similar to the PTC heater, which never ever go above 1.

Did you look upp the descriptions on how the octovalve (and teslas patent that preceeds the octovalve) is supposed to work.
For example on a dat when you need cooling and need to preheat the battery, the heat from the cabin and engines( without doing afterburning) can be used together to at the same time cool the cabin and heat the battery without any unnessesary loss of energy

 

[Devils son]

For those that want to know more, I think this is an interesting read:

Tesla Model Y Heat Pump Details Infrequently Discussed By The Media

Tesla’s new system “boosts” the heat pump to give it better efficiency than regular heat pump systems and uses unique methods for heating below -10C.

insideevs.com

According to this, even at -10 ambient, the COP is not 1, but getting closer to 1.

Looks like it is very interesting engineering and highly effective under multiple circumstances.

 

[Devils son]

I think the point is that while there might actually be circumstances that the heat pump does not make a significant difference, in most drives it definitely will have a significant impact.

We can discuss percentages (somewhere I have read a 10% difference overall for the Model Y), but bottom line will be that it will make the car more efficient overall.

For that I don’t need to test, this would be obvious. If from testing there is no gain recorded, I would first look at the test itself and then proper functionality of the heat pump.

 

[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 did do a 45minute drive yesterday after the 90% balancing was done. Back home at 82%, 4mV imbalance when I parked the car. Still 82% this morning after 15 h of parking in the garage, WC not connected.
As it seems, it doesnt balance at 80 or 82% SOC either.
I feel its probably worth a new thread about balancing the battery on the model 3 with these findings.

 

[AAKEE]

I so far havent found any really good articles. Searched and did read all before i put a tick in the M3P buy box. There are some good explanations about the different heating modes on youtube. These arent very precise about COP and ambient.

The Tesla Patent, though, is.

We can se in this picture, that unless the battery AND ambient temp is below -10C The heat pump can do better than COP 1.
If the battery is between -10 and +10 the heat pump/heating is in blend mode and COP between 1 and 2.

If the battery is 10C or above we can keep the heat pump at COP 1.5 to as low as -30C ambient temps.
The heat pump takes heat from the battery and produced heat from inverters/motors so the total inlet temp to the heatpump is not that low, keeping the heat pump efficiency( COP) much higher than if only outside air was used.

There is described as two main modes for the heatpump, efficient and lossy. Lossy should be the mode it used for me when the hestpump didnt work as planned, as an emergency alternative. This mode is also used in some other modes. When the car is just started, if the battery do not contain enough heat and also to quickly gain high temps and get the cabin heated quickly. This lossy mode can be in combination with taking heat from the battery etc. The lossy mode comes from changing the phase timing to the electric motor, making it use a lot more electricity that becomes heat right away.

Thats the mode when using the heatpump as a heater with COP1. Only done when needed and as soon as other options is available these are prioritized.
Optimal source electric vehicle heat pump with extreme temperature heating capability and efficient thermal preconditioning

 

[TimothyHW3]

The heat pump takes heat from the battery and produced heat from inverters/motors so the total inlet temp to the heatpump is not that low, keeping the heat pump efficiency( COP) much higher than if only outside air was used.

That is pretty much confirming the findings among various videos. The main source of benefit comes from heat scavenging and not from ambient air.

To be honest, I don't really like the idea of battery scavenging at cell level around or below 5C. Also Tesla should integrate the heating into the navigation and check if the commute is very short, then prioritize the heating accordingly - maybe ask the driver what they want - better range or more heat and save it into the profile (then heat up the cabin, turn on the seat heater etc). Obviously will not work for people who don't use the nav, but still.

I think there is still a lot to be optimized with this system to make it better.

 

[AAKEE]

Also Tesla should integrate the heating into the navigation and check if the commute is very short, then prioritize the heating accordingly -

Well, according to the patent it does, or can/could do. Short trips, then its best to use battery heat.
If a trip is known to be long in forehand, heating cabin with less battery heat might be better. ( this info is present in the patent).

 

[Devils son]

I so far havent found any really good articles. Searched and did read all before i put a tick in the M3P buy box. There are some good explanations about the different heating modes on youtube. These arent very precise about COP and ambient.

The Tesla Patent, though, is.

View attachment 651080
We can se in this picture, that unless the battery AND ambient temp is below -10C The heat pump can do better than COP 1.
If the battery is between -10 and +10 the heat pump/heating is in blend mode and COP between 1 and 2.

If the battery is 10C or above we can keep the heat pump at COP 1.5 to as low as -30C ambient temps.
The heat pump takes heat from the battery and produced heat from inverters/motors so the total inlet temp to the heatpump is not that low, keeping the heat pump efficiency( COP) much higher than if only outside air was used.

There is described as two main modes for the heatpump, efficient and lossy. Lossy should be the mode it used for me when the hestpump didnt work as planned, as an emergency alternative. This mode is also used in some other modes. When the car is just started, if the battery do not contain enough heat and also to quickly gain high temps and get the cabin heated quickly. This lossy mode can be in combination with taking heat from the battery etc. The lossy mode comes from changing the phase timing to the electric motor, making it use a lot more electricity that becomes heat right away.

View attachment 651091
View attachment 651093
Thats the mode when using the heatpump as a heater with COP1. Only done when needed and as soon as other options is available these are prioritized.
Optimal source electric vehicle heat pump with extreme temperature heating capability and efficient thermal preconditioning

This is good info and in line with my findings.

Thx.

 

[Devils son]

That is pretty much confirming the findings among various videos. The main source of benefit comes from heat scavenging and not from ambient air.

To be honest, I don't really like the idea of battery scavenging at cell level around or below 5C. Also Tesla should integrate the heating into the navigation and check if the commute is very short, then prioritize the heating accordingly - maybe ask the driver what they want - better range or more heat and save it into the profile (then heat up the cabin, turn on the seat heater etc). Obviously will not work for people who don't use the nav, but still.

I think there is still a lot to be optimized with this system to make it better.

Agree.

But with this info, would you agree that the MY2021 is more efficient?

 

[AAKEE]

I don't really like the idea of battery scavenging at cell level around or below 5C.

I think there is still a lot to be optimized with this system to make it better.

I think very much is done already. Anything can be made better, off course.
Tesla probably comes up with new solutions that can be implemented via software, and maybe some additional hardware to make it better.
Still, I think the engineering of this is outstanding.
I dont think you normally can use a pre-refresh M3 and make it draw less. Worst case refresh is COP1, and best case pre-refresh is COP 1.


We had a couple of inches snow today, but the weather is clear and the temps are falling.
If I let the car be outside one night, what ambient temp should i aim for and what battery temp range would tou like me to be inside?

 

[AlanSubie4Life]

The main source of benefit comes from heat scavenging and not from ambient air.

It would be interesting to know what the average cold-soaked outdoor temperature is for heat use. It seems to me that for a large number of people in temperate climates during the winter, 10C temperatures, cold-soaked, would be very common. There are huge benefits there. (Think Northern California and Oregon & Washington, but I would think this would also be common in Europe through most of the year.)

There it looks like COP of 2.5 which is a huge advantage.

But in any case this data shows also that the heat pump will have only a small advantage at -6C (cold-soaked), as is done in the 20F FTP test procedure. I went back and looked at the data and it looks like only a 6% improvement in efficiency for that cold test (comparing 20" 2020 Performance to 20" 2021 Performance - all the other comparisons are flawed or missing data for one reason or another), which I think aligns pretty well with the above graph.

I think that very small improvement is pretty consistent with not seeing much improvement at all in very cold (20F/-6C) weather in real-world tests, but it definitely will depend on the speed of the test and myriad other factors, for such a small difference.

This improvement along with the hot weather improvements also aligns fairly well with the scalar increase of 7% (the hot cycle SC03 also showed improvement of 12%, which is actually more interesting than the cold weather results!) that has previously been documented and discussed here. (For full disclosure, the scalar increase for the Performance was only 3% though, going from 0.7250 to 0.755. This has to do with the relative performance on the 5-cycle tests vs. the 2-cycle tests for the 2020.)

Regarding the very surprising hot weather results: Seems like the next test to do will be to compare the 2019 to the 2021 in hot weather with the AC blasting to see where that improvement might be coming from (it's quite surprising to me since a simple observation might be: "They both just use AC which will be similarly efficient." But apparently the octovalve or other changes have helped with hot weather performance).

Certainly in very cold weather that's where things are always the worst for EVs, and I don't really see any significant range improvement for 2021 over 2020 in those specific conditions - you'll still need to be very careful to allocate margin between charging stops mid-winter. But we shouldn't let that distract from the much more common scenario of chilly temperatures above freezing, where users will be using their cars most of the time.

Anyway, here is the solid data, straight from Tesla, showing measured efficiencies of these two models for the various test cycles they've run (it's easy to look up what these cycles are, but SC03 and 20F FTP are the key ones for hot/cold testing). Ignore the AC recharge event energy for the Performance 2021 (in red), as it's irrelevant here (it only affects MPGe numbers, not the quoted range or the battery capacity), and it's also an error in their testing which at some point will be resolved.

 

[Devils son]

Certainly in very cold weather that's where things are always the worst for EVs, and I don't really see any significant range improvement for 2021 over 2020 in those specific conditions - you'll still need to be very careful to allocate margin between charging stops mid-winter. But we shouldn't let that distract from the much more common scenario of chilly temperatures above freezing, where users will be using their cars most of the time.

But wouldn’t that be different on the longer trips? Going on the freeway at 80 mph will get heat in the battery and that will get the COP to a decent level, I would assume.

 

[AlanSubie4Life]

But wouldn’t that be different on the longer trips? Going on the freeway at 80 mph will get heat in the battery and that will get the COP to a decent level, I would assume.

Results might differ at -6C on a long trip. Especially after a supercharge which is not captured by EPA results of course. There are tons of variables, like speed, which is important to note. The EPA results represent a blend of these short trip and long trip results (usage is higher at the start of the tests, both with and without heat pump, but the final published result is a blend of the early part of the test and the later part of the test when the car and battery has warmed up).

But even for very long journeys, the difference between the heat pump and the PTC is clearly going to diminish at very cold temperatures. I’d expect minimal difference by -15C, even with some battery heat scavenging going on (see plots above). Whether you exceed COP of 1 is a question of how warm the battery is going to get in those temperatures, steady state, if the heat pump is always pulling heat from it. The whole idea of an EV is to not have waste heat, so there isn’t THAT much heat in the pack. That’s really cold though! For most owners and use scenarios, this is a rare event (it’s obviously more and more common as you head north). To me it looks like a large advantage for the heat pump at much more typical 10C temps. And apparently, as mentioned above, at very high temperatures the 2021 has an advantage - not sure if it is the new windows (aren’t they laminate now or something?), or octovalve, or what is responsible for that. Worth investigating!

 

[Devils son]

Results might differ at -6C on a long trip. Especially after a supercharge which is not captured by EPA results of course. There are tons of variables, like speed, which is important to note. The EPA results represent a blend of these short trip and long trip results (usage is higher at the start of the tests, both with and without heat pump, but the final published result is a blend of the early part of the test and the later part of the test when the car and battery has warmed up).

But even for very long journeys, the difference between the heat pump and the PTC is clearly going to diminish at very cold temperatures. I’d expect minimal difference by -15C, even with some battery heat scavenging going on (see plots above). Whether you exceed COP of 1 is a question of how warm the battery is going to get in those temperatures, steady state, if the heat pump is always pulling heat from it. The whole idea of an EV is to not have waste heat, so there isn’t THAT much heat in the pack. That’s really cold though! For most owners and use scenarios, this is a rare event (it’s obviously more and more common as you head north). To me it looks like a large advantage for the heat pump at much more typical 10C temps. And apparently, as mentioned above, at very high temperatures the 2021 has an advantage - not sure if it is the new windows (aren’t they laminate now or something?), or octovalve, or what is responsible for that. Worth investigating!

Of course, the less heat the heat pump has to work with, the more COP will reach 1.

For me it is not such a big deal. The trips that are most important to me regarding range are those to the wintersport. These are typically at temperatures slightly above freezing for most of the trip.

Unfortunately I haven’t been able to test the car on those trips this year due to a pandemic, so I won’t know more until December (hopefully...).

 

[AAKEE]

Results might differ at -6C on a long trip. Especially after a supercharge which is not captured by EPA results of course. There are tons of variables, like speed, which is important to note. The EPA results represent a blend of these short trip and long trip results (usage is higher at the start of the tests, both with and without heat pump, but the final published result is a blend of the early part of the test and the later part of the test when the car and battery has warmed up).

But even for very long journeys, the difference between the heat pump and the PTC is clearly going to diminish at very cold temperatures. I’d expect minimal difference by -15C, even with some battery heat scavenging going on (see plots above). Whether you exceed COP of 1 is a question of how warm the battery is going to get in those temperatures, steady state, if the heat pump is always pulling heat from it. The whole idea of an EV is to not have waste heat, so there isn’t THAT much heat in the pack. That’s really cold though! For most owners and use scenarios, this is a rare event (it’s obviously more and more common as you head north). To me it looks like a large advantage for the heat pump at much more typical 10C temps. And apparently, as mentioned above, at very high temperatures the 2021 has an advantage - not sure if it is the new windows (aren’t they laminate now or something?), or octovalve, or what is responsible for that. Worth investigating!

For a 100km/h run with 150Wh/km you use for example 15kW engine power, at best these engine are said to have about 95% effeiciency - leaving about one kW of heat loss that goes to the heating circiut. The battery would produce more heat the colder it is, due to internal resistance, but when the battery is +10C the internal resistance is low and the might be a couple of hundred Watts or so. Battery heat is not much according to my calculation, but togheter theres more than one kilowatt of waste power we can use top heat the cabin. Its for free.
So if we input 1kw heat pump motor, use the 1kw we got from the motors and use 1kw from the ambient air, we have three kilowatt of heat for the cost of one.
The combined temperature from the at least 1kW motor/inverter heat plus battery heat togheter with 1kW of ambient air would not be that cold. The battery and engine heat is at least +10C. In the diagram from the patent we see that COP might be between 1.5 to 5.

The consumption for the heat pump version would be about (15+1kW) = 16kw or 160Wh/km.
The old school M3 would need (15 + 3kw) = 18kW or 180Wh/km.