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

[Spoiler Alert + Mild Speculation] Tesla has created a monster!

This site may earn commission on affiliate links.
I don't think they changed the C rate at all. I think the faster charge rate in mph can be attributed to a combination of less parasitic drain as @arg posited combined with the higher efficiency of the car. The Model S is rated around 285 Wh/Mi and the Model 3 around 237.

Let's assume for the simplicity lossless charging at 2.8 KW. In one hour:

Model S - (1 Hr * 2.8KW)/(.285 KWh/Mi) = 9.8 Mi of range added

Model 3 - (1 Hr * 2.8KW)/(.237 KWh/Mi) = 11.8 Mi of range added

Calculating the effective power from the losses

Model S - P = (7 Mi * .285 KWh/Mi)/ 1 Hr = 2 KW

Model 3 - P = (11 Mi * .237 KWH/Mi)/ 1 Hr = 2.6 KW

We also don't know how Tesla rounded the numbers. They may have rounded down the Model S and X numbers and rounded up the Model 3 numbers to make them look better. Tesla also doesn't make clear which version of the cars they are using for these calculations. The P100D is not as efficient as the 100D or the 75D (S or X). I'm going with the 237 WH/Mi leaked by Tesla for the Model 3, but I don't know if that is the small pack or big pack car. For AC charging the size of the pack isn't a big factor, but the driving efficiency is for the mph charging.

In any case, it looks like the Model S has about 800 W overhead when charging, but the Model 3 only 200 W. Due to rounding the two numbers are probably a bit closer, but they probably figured out how to charge with the Model 3 shutting down a number of unneeded systems that may be left running on the older cars. The Model 3 also has less electronics on board in general. The new battery management electronics might be contributing too.

I didn't run the numbers for all charging scenarios or the Model X, but that could be done fairly easily in a spreadsheet. If I didn't have to run off and take care of some chores that have been sitting in a few minutes, I'd slap one together. I might do so this evening if nobody beats me to it.
 
It's also possible that Tesla has redesigned the M3 cells for faster charging since early reports don't seem to show an increase in energy density. Trading density for C rate.

This is AC charging, so it really comes down to the on board charger, not the charge rate of the cell.

So if you are limited to 10 kW from the wall, the Model 3 will definitely charge more mph, since it's more efficient. If you look at kWh added over time, so basically how much usable power the car gets, you can estimate charger efficiency, which might very well have been improved.

But as long as you don't go well above 20 kW, the C-rate doesn't matter too much.

Of course if you have a bigger OBC, then the car can also charge quicker, if there is sufficient power supply from the wall.
 
2B23fIf.gif
Found this helpful overview of the model 3 range (EPA and tested). The Consumer Reports range tests are surprising. I didn't know these range tests were actually performed. Where can I find the source?
 
Hi, @ElecFan. The Consumer Reports test is not a standard test they perform for all cars. The only Tesla they have tested so far is the Model S 75D and it scored 235 miles at 65 mph. See the video here. The table you see shows the range numbers at 65, 70, 75 mph I calculated for all Teslas. I have used EPA highway dyno test scores to extrapolate.

In the EPA highway dyno test, the Model S 75D scored 358.49 miles. See page 19 here. The reason this number is high is that the test is performed at a low speed (48 mph on average). That means the range at 65 mph is 235/358.49= 65.55% of EPA highway dyno score. It would be excellent if the EPA took 66% of EPA highway dyno score and used that for EPA rated range. That would make EPA rated range very accurate. Instead, they use 70% for almost all EV's including the Model 3. However, historically, they have used higher multipliers for the Model S. For example, in 2012, the multiplier was 79.6%. Then it dropped to 75.4% and then to 73.8%. Therefore EPA rated range numbers are messed up and you have to go back to dyno test scores to have comparable data.

The good thing about this test is that the EPA has published all the dyno test scores for all Tesla's including the Model 3 80 (aka Model 3 LR). The Model 3 80 scored 454.64 miles in the same test. See page 7 here. Therefore I have calculated that the Model 3 80's range at 65 mph would be 454.64*65.55%= 298 miles. I expect Consumer Reports to test the Model 3 as well. That should show how accurate my calculations are.

There was some criticism towards the Consumer Reports test. The Bolt beat the Model S 75D and Tesla wasn't happy about it and they had some objections. However, I think the 235 miles score is a realistic score because a new Model S 75D displays 241 miles range at 100% charge outside of North America. Tesla does not use EPA rated range outside of North America. Instead, they use a range unit they created called "Typical range".
 
It's also possible that Tesla has redesigned the M3 cells for faster charging since early reports don't seem to show an increase in energy density. Trading density for C rate.

JRP3 and others who discussed the EPA doc: What do you say about the new piece of info from Sandy Munro at the at the 32:00 minute mark?


He quite clearly say that they measured the cells to be 50% the volume, but 100% more mA. "From 3000 ma to 5700. Actually, we think it is 6000". There are a few possible explanations, all interesting:
1) Munro's people cant measure battery cells properly, feels strange, 3000ma for the 18650 is correct or perhaps conservative depending on how wrong you think the 85 kWh figure of the early pack was. Anyways, that number is not exaggerated, so why would the 2170 be? And he refers to his guy as a battery expert who helped build factories in china, and that they have torn down the majority of hybrids and EVs sold in the US. They should know what they are doing by now.
2) Tesla is sandbagging the capacity because they have prioritized a pack design that withstand higher C rates for longer, capping the cells at lower max SOC, also increasing life time towards the goal of "1 million miles".
3) Tesla is sandbagging the capacity to not beat Model S/X on range. If you do the calculations, I end up at more than 300Wh/kg at cell level, and more than 180Wh/kg at pack level, which is much more in line with public statements given by Straubel and Elon before on the trends from Roadster to Model S, to Model S P100 with more silicon anode, and then to Model 3. The are good reasons to not beat the current model S X architecture as has been discussed before.
4) Any other explanations? Combination of the above?

(Edited for clarity)
 
The LR pack is 80.5 kWh and 4416 cells.
A cell is nominally 3.6 v

Works out to 5060 mAhr per cell

By the way, if my arithmetic is correct the 2170 is 1.46x the volume of the 18650, not 1.5x.

Yes, you are right that volume wise it somewhat lower than 1.5x. And using actual 80.5 kWh instead of usable 78.3 kWh give us 5Ah. But still, this is off by 20% compared to 6Ah which is a lot. And 20% take us from 250 to 300 Wh/kg, a huge deal at cell level, if their measurement is correct. I.e., it is not enormously important for Model 3 as we have the needed range and specs anyhow, but it has quite some implications on the outlook of Tesla in terms of cell chemistry progress. If it continues as promised by Elon and Straubel on multiple accounts, or if it is stagnating.
 
And the current 18650 cells are over 3400 I believe, so even if 6000 was correct it's not double.

Yes, the more recent cells are 3400 mAh. So definitely not doubling from that basis. My guess is that Munro tore down, or refer to cells of an original S85 when it came out, which according to @wk057 had useful 77.5 kWh = 3000 mAh cells, and actual 81.5 kWh so 3150 mAh cells, so around 3Ah and this is what they are comparing with when they round to double the energy capacity, which I think is fair, if the measurement of 6Ah is true.
 
1) Munro's people cant measure battery cells properly, feels strange, 3000ma for the 18650 is correct or perhaps conservative depending on how wrong you think the 85 kWh figure of the early pack was. Anyways, that number is not exaggerated, so why would the 2170 be? And he refers to his guy as a battery expert who helped build factories in china, and that they have torn down the majority of hybrids and EVs sold in the US. They should know what they are doing by now.
2) Tesla is sandbagging the capacity because they have prioritized a pack design that withstand higher C rates for longer, capping the cells at lower max SOC, also increasing life time towards the goal of "1 million miles".
3) Tesla is sandbagging the capacity to not beat Model S/X on range. If you do the calculations, I end up at more than 300Wh/kg at cell level, and more than 180Wh/kg at pack level, which is much more in line with public statements given by Straubel and Elon before on the trends from Roadster to Model S, to Model S P100 with more silicon anode, and then to Model 3. The are good reasons to not beat the current model S X architecture as has been discussed before.
4) Any other explanations? Combination of the above?

(Edited for clarity)

I think it's all but #1. Measuring an individual cell capacity is trivial and doesn't take any expertise. What we don't know is what parameters were used to determine "full" and "empty".

2. We don't know what maximum / minimum cell voltages that the Model 3 BMS uses. If Tesla were to cap the voltage at 4.0v/cell then the overall lifetime of the battery would be significantly higher, and it would have ~80% of the capacity of a fully charged cell.
5000/6000 is 83%.

3. It would be a win/win for Tesla to only go to 4.0 or similar volts per cell. They get to not worry about warranty work due to degradation (ever) since they could at some point reprogram the voltage cut off to a higher voltage to "gain" energy back. It also allows Model S 100D to maintain it's higher mileage rating.

-Jim
 
I think it's all but #1. Measuring an individual cell capacity is trivial and doesn't take any expertise. What we don't know is what parameters were used to determine "full" and "empty".

2. We don't know what maximum / minimum cell voltages that the Model 3 BMS uses. If Tesla were to cap the voltage at 4.0v/cell then the overall lifetime of the battery would be significantly higher, and it would have ~80% of the capacity of a fully charged cell.
5000/6000 is 83%.

3. It would be a win/win for Tesla to only go to 4.0 or similar volts per cell. They get to not worry about warranty work due to degradation (ever) since they could at some point reprogram the voltage cut off to a higher voltage to "gain" energy back. It also allows Model S 100D to maintain it's higher mileage rating.

-Jim

Yes, this is what I'm leaning towards too. Pure speculation until someone proves that the cells only charges to lower than the 4.15V we know from Model S/X in max range. But very interesting if they indeed apply this strategy now suddenly. Before, they have pushed the limit. Note that the charge screen (to my knowledge) on Model 3 only give kW as Model S and X do nowadays when supercharging, not V and A as they used to do. So we cant infer anything from that. But happy to be corrected!

Another option is that the chemistry is more "fundamentally" different, and that they now need to limit to say 4.0V, but on the other hand, they now can tolerate higher C rates, which others have speculated is coming up, given the much faster charge times / rates that (again) has been flagged by both Elon and Straubel on multiple occasions, but not yet been implemented...

(edited for clarity)
 
3400 * 1.46 = 4964, compared to the 5063 calculated.

Close enough* to say that the chemistry is the same in the two cells.


*I don't know how thick the metal wrapping the cell is but the actual volume increase of active material will be a little more than 1.46x

Yes, sure, totally agree, the chemistry is the same by this math. The point is that the 6Ah quoted in the video is 20% more than this roughly 5Ah, a bit less if we go by the 5.7Ah. So, the Munro data point is significantly different from what we can calculate from either the EPA Model 3 data, or from the volume change from the 18650, right?
 
  • Like
Reactions: omgwtfbyobbq
JRP3 and others who discussed the EPA doc: What do you say about the new piece of info from Sandy Munro at the at the 32:00 minute mark?


He quite clearly say that they measured the cells to be 50% the volume, but 100% more mA. "From 3000 ma to 5700. Actually, we think it is 6000". There are a few possible explanations, all interesting:
1) Munro's people cant measure battery cells properly, feels strange, 3000ma for the 18650 is correct or perhaps conservative depending on how wrong you think the 85 kWh figure of the early pack was. Anyways, that number is not exaggerated, so why would the 2170 be? And he refers to his guy as a battery expert who helped build factories in china, and that they have torn down the majority of hybrids and EVs sold in the US. They should know what they are doing by now.
2) Tesla is sandbagging the capacity because they have prioritized a pack design that withstand higher C rates for longer, capping the cells at lower max SOC, also increasing life time towards the goal of "1 million miles".
3) Tesla is sandbagging the capacity to not beat Model S/X on range. If you do the calculations, I end up at more than 300Wh/kg at cell level, and more than 180Wh/kg at pack level, which is much more in line with public statements given by Straubel and Elon before on the trends from Roadster to Model S, to Model S P100 with more silicon anode, and then to Model 3. The are good reasons to not beat the current model S X architecture as has been discussed before.
4) Any other explanations? Combination of the above?

(Edited for clarity)

I watched the entire video. The battery analysis is interesting. Tesla has been working with a cell structure that was established 20 years ago when the 18650 was introduced. There may have been a more efficient structure possible with the 2170 layout that allowed them to squeeze more battery in there? We probably won't know until somebody takes apart a 2170, which won't be soon. That's not something a hobbyist wants to do in their garage. You basically need a hazmat lab setup because if you do it wrong, the cell will catch on fire.

The battery stuff aside, his analysis was fascinating to me and I think his conclusion was spot on. Tesla's weakness is in dinosaur tech: making metal beasts. He said if the construction quality and mechanical design were to the same standard as the electronics, every car maker in the world should be soiling their underwear.

The pros:
1) The quality of design of the electronics is on par with the best in the business. He compared the autopilot hardware design and build quality to what you see in state of the art military aircraft or very expensive super computers. Years ahead of the rest of the car industry.
2) Tesla's battery pack construction and battery management was beyond what he thought was possible. He had been very impressed with the BMW i3, but the Model 3 blew that away.
3) The car drives like a much more expensive sports car.

The cons:
1) He compared the build quality to a 1990s Kia. He admitted some are built well, but the quality varies all over the place.
2) Some of the mechanical design elements had him scratching his head. Despite using a lot of aluminum, the body of the car is heavier than any steel body in its class.

He believed if Tesla got some guys like him in there to restructure production and clean up the mechanical design, they would be untouchable. Right now he said they are vastly ahead of everyone in some areas, but way behind in others. And he said the places where Tesla is behind are all problems that were solved many years ago.

I really think Tesla should buy Munro's report and hire him as a consultant to clean up manufacturing and tighten up the mechanical design. Munro said he wants to see Tesla succeed.

I may watch that video again. There was a lot of good stuff in it.

I think it's all but #1. Measuring an individual cell capacity is trivial and doesn't take any expertise. What we don't know is what parameters were used to determine "full" and "empty".

2. We don't know what maximum / minimum cell voltages that the Model 3 BMS uses. If Tesla were to cap the voltage at 4.0v/cell then the overall lifetime of the battery would be significantly higher, and it would have ~80% of the capacity of a fully charged cell.
5000/6000 is 83%.

3. It would be a win/win for Tesla to only go to 4.0 or similar volts per cell. They get to not worry about warranty work due to degradation (ever) since they could at some point reprogram the voltage cut off to a higher voltage to "gain" energy back. It also allows Model S 100D to maintain it's higher mileage rating.

-Jim

The voltage of a cell is mandated by the laws of Physics with the chemistry used. Different chemistries have different voltage levels, even different Li-ion chemistry mixes have different voltage levels, but two cells of the same chemistry will always have the same voltage at the same level of charge.

Tesla is unlikely to go with a li-ion chemistry with a higher voltage unless it also provides benefits over the existing chemistry with few or no drawbacks.
 
We probably won't know until somebody takes apart a 2170, which won't be soon. That's not something a hobbyist wants to do in their garage. You basically need a hazmat lab setup because if you do it wrong, the cell will catch on fire.

There are plenty of people comfortable with taking a Tesla battery pack apart around here. EVTV has already gotten a salvage Model 3 that they are taking apart, but I don't know if they plan on taking the battery pack apart to the cell level. @wk057 would be a good candidate but he thinks that the salvage Model 3s have been going for too much, but I suspect the price will start to drop as more of them get wrecked.