Model S Plaid Battery Details, sourced from EPA Docs, Vehicle Observations, & Supercharging/Charging data

[Zoomit]

MotorTrend has published a decent article on their Plaid V3 Supercharging experience.

Tesla Model S Plaid Fast-Charging and Range Test: How Far Can It Go?

The resulting charge profile is slightly lower than the limited data points I’ve seen posted here, but I do expect that it’s “typical”. Meaning, we can go to great extent to empirically derive an ideal charging profile (been there, done that on Model 3), but it’s exceedingly rare for all the conditions to align such that the ideal charging profile is seen in practice. This MotorTrend Plaid charging profile data is actually fairly similar to the Model 3 profile under ideal conditions. The article summarizes the time spent to charge to different levels and discusses other relevant topics. It is worth the read for those interested.

 

[MorrisonHiker]

Plenty of vids showing the MS modules. Some are clownish, like Rich rebuilds, some are long winded, like Jack Ricard.

This is a short one to help those orient themselves with the existing MS pack:

Just watching on my phone so I can't see closely. He mentions 444 cells but there are a lot more than 444 18650 cells in a battery. Isn't it something like 8256 18650 cells in an older 100 kWh battery?

Too bad he didn't get video of the sign at the display.

 

[MarcG]

This fellow has crunched some numbers, and is convinced that the pack has 4680's, despite everyone saying it doesn't.
Who is going to be the first in their new 2022 Model S to rip out their new pack and tear it apart to let us know the facts?

This guy’s math (or maths, if he’s british) is/are so off it’s laughable. He’s getting incorrect battery pack weights from the total weight differences of the car, then he’s making direct correlations to the battery capacity differences between LR and Plaid from those weight differences, not taking into account non-cell component weights. Major fail.

 

[MarcG]

MotorTrend has published a decent article on their Plaid V3 Supercharging experience.

Tesla Model S Plaid Fast-Charging and Range Test: How Far Can It Go?

The resulting charge profile is slightly lower than the limited data points I’ve seen posted here, but I do expect that it’s “typical”. Meaning, we can go to great extent to empirically derive an ideal charging profile (been there, done that on Model 3), but it’s exceedingly rare for all the conditions to align such that the ideal charging profile is seen in practice. This MotorTrend Plaid charging profile data is actually fairly similar to the Model 3 profile under ideal conditions. The article summarizes the time spent to charge to different levels and discusses other relevant topics. It is worth the read for those interested.

Thanks for sharing! I was hoping the Plaid would charge at peak rates for longer, or at least have a less aggressive taper past 30%, but at least this puts it on par with the Model 3 (but with a decent amount of extra range per % SoC).

 

[MarcG]

Just watching on my phone so I can't see closely. He mentions 444 cells but there are a lot more than 444 18650 cells in a battery. Isn't it something like 8256 18650 cells in an older 100 kWh battery?

Too bad he didn't get video of the sign at the display.

This is a single Model S module of 18650 cells from pre-prefresh packs. There used to be 16 modules in a battery pack.

 

[AlanSubie4Life]

250kW up to about 33% SoC

Looks like it noses over at 31% or so (for this charge session). So the direct scaling method seems pretty close so far (since this curve will probably push out some more in idea conditions).

but at least this puts it on par with the Model 3 (but with a decent amount of extra range per % SoC).

I think it is on par with an ideal Model 3 session. I think we’ll likely see it push out a bit more in other sessions, eventually (or with updates), so will be comfortably faster than Model 3 on a kW basis.

Model S is a less efficient vehicle, of course (but not by that much for the LR), so maybe close to on par overall (I suspect it will end up slightly better eventually but that would take higher than 250kW rates).

For the LR, 25-28% larger battery than Model 3, but on the raw efficiency I think it is only going to be about 12% worse on the highway (a bit more at high speeds since aero drags start to climb, but not 25% worse). Raw Unscaled Hwy FET numbers are 223Wh/mi (S LR) vs. 198Wh/mi (Model 3) for reference (ignore the absolute numbers - obviously not representative of true efficiency).

 

[AlanSubie4Life]

This guy’s math (or maths, if he’s british) is/are so off it’s laughable.

Yeah. It’s so bad there is nowhere to start. He seems confused about what 4680s will do, too. The guy’s maths just mostly don’t exist, is the main problem.

He has convinced himself the Model S LR has a much smaller battery somehow. Though that is clearly contradicted by the EPA data.

 

[aerodyne]

Just watching on my phone so I can't see closely. He mentions 444 cells but there are a lot more than 444 18650 cells in a battery. Isn't it something like 8256 18650 cells in an older 100 kWh battery?

Too bad he didn't get video of the sign at the display.

There was no signage, but I believe there was nomenclature on the circuit board.

Those were 18650 cells, 444 in this single module, usually 16 modules in a MS HV pack.

 

[MorrisonHiker]

This is a single Model S module of 18650 cells from pre-prefresh packs. There used to be 16 modules in a battery pack.

There was no signage, but I believe there was nomenclature on the circuit board.

Those were 18650 cells, 444 in this single module, usually 16 modules in a MS HV pack.

Oops. Thanks for pointing that out. I was trying to view it on my phone without my contacts in and like I said, I couldn't really see it closely. The video was referring to cells in one module and I was referring to cells in an entire pack.

 

[AlanSubie4Life]

Some of the discussion in the more general Model S battery sticky thread seems to be pointing towards a ~455V Plaid pack (which would be 108s, approximately). That would make the Owner’s Manual a little inconsistent but that alone would not be a big deal.

This is based on somewhat incomplete Supercharging voltage data.

 

[AlanSubie4Life]

The documents are released. There's tons of info in these; I don't have time to go through them all now - so others can comment on items of interest that they notice.

The Model S Plaid has a 99.3kWh capacity battery with 85.6% charging efficiency. (Lower capacity than the 103kWh I expected, but also much lower charging efficiency. This is similar to the Model 3 Performance 2021, and I don't understand these terrible numbers (typically expect 88-89%).)

Nominal voltage is 410V. (vs. 360V previously)

537kg, 186Wh/kg. (Vs. 625kg, 165Wh/kg previously). A big improvement!

https://iaspub.epa.gov/otaqpub/display_file.jsp?docid=52951&flag=1

The Model S Long Range also has the same 410V battery:

98.3kWh, 86% charging efficiency.

https://iaspub.epa.gov/otaqpub/display_file.jsp?docid=52949&flag=1

(You can compare this document to the older 2021 document located at the same site, and see all the differences in vehicle weights, etc.)

Standard verbiage in these about the new heat pump, but worth a read for anyone interested.

These are lower capacities than I expected, and lower charging efficiency, but this means the on-road efficiency is much better than I speculated previously, by a few %.

However, the lower capacity does make sense, since previously we had calculated the degradation threshold to be about 96kWh (below this is when capacity loss will start to show). That 2kWh gap between EPA energy and the threshold is fairly typical.

The lower charging efficiency is odd. I don't really get it. We have seen such outliers before, but it's really strange. See the older 2021 document - the Performance got 88% charging efficiency and the Model S Long Range Plus got 86.5% efficiency. The inconsistency is weird for a well-controlled test like this.

 

[AlanSubie4Life]

A little more editorializing:

Raw Efficiency (these are all battery to wheels, directly from the documents - not wall to wheels):

Model S Long Range: 192Wh/mi (hwy).
Model S Long Range Plus (older version without heat pump): 198Wh/mi (hwy)

A 3 % efficiency improvement, but note this may not be equivalent tires (no idea the rubber on each of these vehicles).

Model S Plaid: 224Wh/mi (hwy)

Model S Performance 21" (older version without heat pump): 237Wh/mi (hwy)

(That's a 5.5% efficiency improvement! Presumably tire choices are similar here even if the widths differ and the Plaid is at a disadvantage.)

Obviously these results are impacted by tire choice and these numbers do not properly capture any aero improvements because the highway cycle is done at such a low speed.

Going to the US06 drive cycle (the most aggressive available) for comparison, and back calculating using the charging efficiency to get the battery to wheels numbers:

City portion: (The beginning and end of the cycle)
Model S Plaid: (116.3kWh AC, 99.3kWh DC)
295.8Wh/mi (AC) => 252.6Wh/mi (DC)

Model S Performance 21": (117.6kWh AC, 103.5kWh DC)
326.8Wh/mi (AC) => 287.6Wh/mi (DC)

That's a 12% efficiency improvement! That's significant and quite impressive.

Highway portion:
Model S Plaid: 274.5Wh/mi (AC) => 234.4Wh/mi DC
Model S Performance 21": 294.66Wh/mi (AC) => 259.3Wh/mi (DC)

A 9.6% efficiency improvement, battery to wheels. Also impressive.

I don't really understand exactly why the efficiency improvement is greater on this more aggressive test, including the "city" portion (I'd understand the highway portion, as being driven by aero improvements). But it is what it is. Large improvements, regardless.


One thought on the difference in charging efficiency is that in some cases I suppose it may take a long time to complete the charging on some vehicles (as we all know completing a charge to 100% can be kind of variable depending (perhaps) on balancing, etc.). The longer that takes on the test article the worse the efficiency would be due to the 200-250W of overhead (or whatever it is).

In these particular documents there are no times given, but the earlier documents don't bear out this theory, one way or the other (they're within 8 minutes of charging time for two vehicles with similar (low) efficiencies, and they don't give a charging time for the vehicle with better efficiency). So no idea what the deal is with this.

 

[omarsultan]

Interesting they still reference a 72A charger.

 

[Zoomit]

Interesting they still reference a 72A charger.

There's typically a lot of crap in these documents that Tesla doesn't bother to update, or purposely leaves incorrect or vague. For example:

"The vehicle is also capable of accepting DC current up to 225A from an off‐board charger (Supercharger)."

That's equivalent to only ~100 kW.

 

[AlanSubie4Life]

There's typically a lot of crap in these documents that Tesla doesn't bother to update, or purposely leaves incorrect or vague. For example:

"The vehicle is also capable of accepting DC current up to 225A from an off‐board charger (Supercharger)."

That's equivalent to only ~100 kW.

Yep. The old verbiage I tend to ignore. Hopefully the new information on battery weight, and energy density, is correct (no guarantees though).

And of course all the measured numbers presumably need to be correct.

Nominally 3x 250kW motors, which I assume is close to correct.

 

[BigNick]

So the LR has one fewer kWh of battery capacity than the Plaid? (98.3 vs 99.3)?
It doesn't make sense that they'd use a different battery pack if they're that close.
Could this just be manufacturing variances in the cells adding up to the 1 kWh difference?

 

[ItsNotAboutTheMoney]

So the LR has one fewer kWh of battery capacity than the Plaid? (98.3 vs 99.3)?
It doesn't make sense that they'd use a different battery pack if they're that close.
Could this just be manufacturing variances in the cells adding up to the 1 kWh difference?

They upped the nominal voltage of the pack to 450V.

 

[BigNick]

They upped the nominal voltage of the pack to 450V.

I meant the refreshed LR is 98.3 and the Plaid is 99.3 but both batteries are 450V max/410V nominal.

 

[jbcarioca]

...

Model S is a less efficient vehicle, of course (but not by that much for the LR), so maybe close to on par overall (I suspect it will end up slightly better eventually but that would take higher than 250kW rates)....

It's too early to conclude that. There are only a couple of indications of Wh/mi equivalent comparisons. It seems possible that the Model S in both variants might actually be more efficient than is a P3D in terms of Wh/mi. I would like to see clear apples-apples comparisons before concluding anything at all. All we know for certain is that the new Model S in both variants is amazingly capable in unanticipated ways. The PLAID 19" is rated at 390 miles and the LR at 405. They both benefit from better aerodynamics, new cell chemistry, new motors, new BMS (including everything from heating/cooling, inverters, and numerous other things. We also know the actual battery capacity is slightly smaller than were the preceding models.

Until a Sandy Munro teardown and a handful of owner comparisons crop up we should be very cautious about expecting 'less efficient'.

 

[aerodyne]

It's too early to conclude that. There are only a couple of indications of Wh/mi equivalent comparisons. It seems possible that the Model S in both variants might actually be more efficient than is a P3D in terms of Wh/mi. I would like to see clear apples-apples comparisons before concluding anything at all. All we know for certain is that the new Model S in both variants is amazingly capable in unanticipated ways. The PLAID 19" is rated at 390 miles and the LR at 405. They both benefit from better aerodynamics, new cell chemistry, new motors, new BMS (including everything from heating/cooling, inverters, and numerous other things. We also know the actual battery capacity is slightly smaller than were the preceding models.

Until a Sandy Munro teardown and a handful of owner comparisons crop up we should be very cautious about expecting 'less efficient'.

One take away for me and other owners of older MS is that the LR is a hella more efficient.

192 vs my 290 Wh/mi. And charges three times faster.

With high used car prices, I could trade in my '15 85D for only 30k. Really tempting...