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

[henderrj]

I was really hoping the new battery pack could be used in the old model S, but it doesn't look like that's going to be possible. Too bad, could really make those cars last a long time! I have nearly a quarter million miles on mine, with the original battery, and almost 2 years left on my battery warranty. Kind of hoping for a new one!

 

[BPeter]

EPC has been updated with MS Refresh
Charger is in front penthouse, pretty sure.

Circled the interesting component, but item 2 is marked as that metal cover looking thing? Maybe the electronics are below the cover.
PCS is Power Conversion System, which is the charger I believe.
ASY,PCS,48A,1PH&3PH,P2 MSX

Interesting, is this charger capable of both 1 and 3 phase?

Model 3 part is ASY,PCS,48A,1PH,MDL3 (1135558-00-C) so different for 1PH and 3PH

 

[BPeter]

This was also interesting

Charge port is "interim"

HV INLET,ASY,NA INTERIM,SX,P2

 

[MorrisonHiker]

This was also interesting

Charge port is "interim"

HV INLET,ASY,NA INTERIM,SX,P2

View attachment 673629

Not only is it "interim", but it's free!

Also...what is part #2?

 

[BPeter]

Not only is it "interim", but it's free!

Also...what is part #2?

Yes part #2 is also very interesting. Looks like it has the same cable as part #1. Could be the new charge port after the rear redesign.

 

[AlanSubie4Life]

plaid changes this. 450V implies 108S, the number of parallel cells will decrease proportionally. So the outgoing Model S pack was 96s86p and the new one will be something like 108s76p. This would imply a decrease in cell count from 8256 cells to 8208 cells. Of course this is just guesswork. But guesswork that can’t be more than a few percent off.

It’s interesting that none of the serial string counts that give near 450V are divisible by by 8256 so cell count almost certainly changed

Trying to keep this discussion out of the Model S Delivery thread; the above quote is from that thread.

Remember 450V is the nominal pack voltage not the max pack voltage. There is more detail in the first post here, with links.

As mentioned in the beginning of this thread, based on how the voltages are specified, I'd guess 120s. Furthermore I would guess:

120s69p (Since Elon likes the number 69, haha.) That gives 8280 cells (If they are 18650; about 40-50% fewer cells and more like 45-50p if they are 2170 (probably aren't)).

It seems like it would make sense to step up from 400V (~403V, ~4.2V per cell) max to 500V max, but who knows. It could be 475V, or 525V. Or anything. I guess it depends on the Supercharger capabilities, and also the capabilities of the MOSFETs in the inverters (how high a voltage they can sustain...though I'm not sure how all of those limits work out at the device itself). Porsche is doing 800V so I guess very high voltage devices exist, but they're probably more expensive.

 

[aerodyne]

So, my old 85 pack is labelled on the data "plate" 400VDC. I know that it is near the max rated voltage, at 4.2 VPC.

Why do we think the new pack is 450V, nominal, not max?

For the record, I think that chemistry and cooling, with the higher voltage, are enabling the higher pack performance.

 

[AlanSubie4Life]

Why do we think the new pack is 450V, nominal, not max?

See the explanation and Owner’s Manual links in the OP for the rationale.

 

[aerodyne]

See the explanation and Owner’s Manual links in the OP for the rationale.

I did, but I guess I should have mentioned I don't trust the OM. With all the changes going on, it is likely to have errors. The data tag is almost certainly correct, understand no one has seen it yet on the plaid.

For instance, OM says the MS are 320V nominal. That does not make sense for my 85 pack.

 

[AlanSubie4Life]

For instance, OM says the MS are 320V nominal. That does not make sense for my 85 pack

Yeah precision is impossible. However, Model 3 and the latest Model S OMs both say 350V/360V. And we know these are ~403V max packs.

So, extrapolate from there. At least I think we know with reasonable confidence that the OM value is always lower than the max value. It seems like it is at around 50-60% SoC or something (not sure exactly what the curves look like). So if it is declared at 450V for Plaid...500V or more max seems likely.

 

[aerodyne]

Yeah precision is impossible. However, Model 3 and the latest Model S OMs both say 350V/360V. And we know these are ~403V max packs.

So, extrapolate from there. At least I think we know with reasonable confidence that the OM value is always lower than the max value. It seems like it is at around 50-60% SoC or something (not sure exactly what the curves look like). So if it is declared at 450V for Plaid...500V or more max seems likely.

The 100 KwH and Raven packs appear to be close to thier advertised capacity, per SMT, unlike the 85's.

In advance of CAN bus data on the Plaid, I'll be looking at the EPA data from the newer cars and comparing them to the refresh LR.

Really curious how they managed to reduce weight, increase range and efficiency and charging speed by just chemistry in the 18650 form. I'll bet they applied some tech from the 4680 development.

 

[AlanSubie4Life]

advance of CAN bus data on the Plaid, I'll be looking at the EPA data from the newer cars and comparing them to the refresh LR.

Definitely EPA data will tell everything, once it is available.

managed to reduce weight,

Do we know this? I imagine there is a lot of learning from the past, various tweaks. I guess we will see about the actual cell itself (lots of those so if they somehow found a way to reduce any weight that would help.

increase range and efficiency

I am sure the carbon-sleeve rotor motors have managed to squeeze out tiny additional savings, tweaks to electronics of drive inverter as well. Higher voltage helps since you can either run thinner wires with the same loss or the same thickness with less loss. Aero is slightly better. Heat pump may fundamentally allow slightly better efficiency during the standard city/hwy tests (though I would expect minimal help)

Beyond that there are the details of the EPA test - specifically the scalar factor used, which is likely going to be a couple % higher due to better cold and hot cycle performance, due to the heat pump’s ability to save energy. This can add an extra 10 or more miles to the EPA range even if the city and highway test results don’t change at all.

and charging speed

I’ve heard that in the past wiring was a limiting factor for Supercharging speeds. So that would be one place relatively easy to improve. The voltage helps with this too (higher voltage lower current for the same speed).

by just chemistry in the 18650 form

Yeah, wonder whether the recent 2170 density improvement was also moved to the 18650. Along with tweaks allowing higher charging and discharging rates. That would also allow them to use 5% fewer cells. So maybe if it really is just 103kWh, it might be just 120s66p.

I guess I still expect peak Supercharging rates could be as high as 330kW, but not sure any Superchargers are capable of that at the moment. Maybe the hardware anticipated this moment, though.

Baseline highest rate per capacity is Model 3 250kW for 78kWh pack, I guess:

At any SOC, new pack is 25% higher voltage.

With 20% higher voltage, the parallel capacity of the pack is going to be about 5.6% higher than Model 3.

Product is just 1.25*1.056 = 1.32 (which is of course just the ratio of the 103kWh to 78kWh pack capacities).

So that gives 330kW. With nearly exactly the same wiring losses as at 250kW due to the higher voltage.

Would end up around 1300mi/hr (peak) at current constants. A little higher even, maybe, for the LR. The best!

 

[TNGuy99]

This seems like a fairly pedestrian observation in contrast to the great data and analysis in this thread, but here goes...

What caught me by surprise about the Plaid Model S is how quickly it can precondition it's battery when selecting "Dragstrip Mode." With roughly the same battery conditions that @omarsultan described in a recent post, his Plaid took about 5 minutes to precondition the battery for Dragstrip Mode, while my 2020 Model S Performance takes approx. 45 minutes to precondition the battery for "Ludicrous+ Mode" (which is essentially the same as Dragstrip Mode) under similar battery conditions.

I realize that Tesla has made big improvements to battery temp management, but this is a drastic improvement and has me wondering how they did it! The time it takes to precondition my 2020 MSP makes Ludicrous+ kind of a non-feature due to the amount of time needed, but the Plaid changes all of this.

Unfortunately I don't have much battery preconditioning data to provide other than Omar's post, some glances that I've caught on recently posted Plaid videos where they briefly show the preconditioning message on the binnacle and my personal experience with my car, but it's very promising and had to require some serious wizardry.

 

[BPeter]

This seems like a fairly pedestrian observation in contrast to the great data and analysis in this thread, but here goes...

What caught me by surprise about the Plaid Model S is how quickly it can precondition it's battery when selecting "Dragstrip Mode." With roughly the same battery conditions that @omarsultan described in a recent post, his Plaid took about 5 minutes to precondition the battery for Dragstrip Mode, while my 2020 Model S Performance takes approx. 45 minutes to precondition the battery for "Ludicrous+ Mode" (which is essentially the same as Dragstrip Mode) under similar battery conditions.

I realize that Tesla has made big improvements to battery temp management, but this is a drastic improvement and has me wondering how they did it! The time it takes to precondition my 2020 MSP makes Ludicrous+ kind of a non-feature due to the amount of time needed, but the Plaid changes all of this.

Unfortunately I don't have much battery preconditioning data to provide other than Omar's post, some glances that I've caught on recently posted Plaid videos where they briefly show the preconditioning message on the binnacle and my personal experience with my car, but it's very promising and had to require some serious wizardry.

Precondition for the new pack may be targeting a lower temperature than the old pack? Does it show the preconditioning temperature on the new one?

 

[AlanSubie4Life]

realize that Tesla has made big improvements to battery temp management, but this is a drastic improvement and has me wondering how they did it!

In addition to batteries that might not require quite as much heat to be able to draw high current, I would guess the heat pump capacity in addition to the stator heating are big factors.

Stator heating probably can put out more heat.

Heat pump probably has very large output capacity as well - over 10kW when operating in its heating mode?

Would add up to 20kW of heating capacity.

Guess someone needs to watch their battery gauge when doing this prep and see how quickly the rated miles drop, to get an idea of the amount of energy that gets dumped into the pack (divide by time for the power of course).

Maybe at some point @omarsultan could take a video of the prep process, starting with a battery that has sat overnight, with his distance display mode set to “distance,” and capture the instrument panel sequence over 5 minutes or so? Would also capture the sounds, etc.

The EPA docs will have a description of the basics of the thermal management system, when they become available, so when they are posted here that is another place to look for details. (Though it might just be a repeat of the Model 3/Y verbiage.)

 

[LostWages]

Single data point isn't super useful, but this looks to be a Model S Plaid supercharging at ~246kW @ 34% SoC:

https://twitter.com/x/status/1405010773076627459

I'm super curious to see a full V3 Supercharging curve ...

 

[BPeter]

Charge port diagram on EPC is no longer loading... Everything else does.

 

[AlanSubie4Life]

Single data point isn't super useful, but this looks to be a Model S Plaid supercharging at ~246kW @ 34% SoC:

https://twitter.com/x/status/1405010773076627459

I'm super curious to see a full V3 Supercharging curve ...

I would expect in the short term to see no better than 250kW but with a broad plateau at 250kW (much broader than Model 3). And the plateau is what really matters, mostly, of course (brief spike to 330kW would only help very slightly). From that we should be able to roughly extrapolate the potential peak (fitting to Model 3 curve vs. SoC and scaling up, then imagining no hard limit). But unless they “unlock” SuC V3 I would expect 250kW max for a while.

We will see!

 

[AlanSubie4Life]

Back to the apparent penthouse location…seems a little odd that it is not under the rear seat (wonder what is there, if anything, as someone else was wondering). Looks like this means they will have to drop the pack to service the DCDC, AC-DC, etc. (these fail much more frequently than the pack, AFAIK).

Hopefully they have made pack removal easier than it is on Model 3. But on Model 3 I thought they just removed the rear seat to perform penthouse service.

 

[rhuber]

Another annoying single datapoint. This one is doing 54kw (edit: or 56 or 58. low res twitter phottos...)kw at 90%. Again promising, but we need about 100 more datapoints to know anything useful about the actual curve.

https://twitter.com/x/status/1405243407476297728