What does Powerwall 2 tell us about the Model 3 battery?

[J1mbo]

Does comparing the specs of Powerwall 1 against Powerwall 2 tell us anything useful about the Model 3 battery and 2170 cells?

Dimensions (L x W x D)

PW1: 52.1" x 33.9" x 7.1" (130cm x 86cm x 18cm) => 201,240 cm3
PW2: 44" x 29" x 5.5" (115cm x 75.5cm x 15.5cm) => 134,578 cm3

Based on the above, PW2 is 60% the volume of PW1, but with 1.4x the total storage (14kWh vs 10kWh) and just a 10% increase in weight.

The PW2 "buffer" is 66% smaller as well (0.5kWh vs 3kWh for PW1) giving more available power.

PW2 also has 5kW continuous (7kW peak) vs 2kW continuous (3kW peak), suggesting the new cells can support a higher discharge rate, or the BMS strategy for the original cells was more cautious.

If the same factors were applied to an 85 battery, that would give a 119 kWh battery with ~112kWh usable.

Assuming the M3 battery is physically 30% smaller than the current MS/MX (30% is taken from comments on an investor call last year, maybe someone has real data?), that gives a maximum 84kWh battery with about 78kWh available.

Happy to be corrected on any of the above!

 

[electracity]

Yes! I agree that the specs are highly significant

 

[Garlan Garner]

Time will tell.

 

[ecarfan]

Based on the above, PW2 is 60% the volume of PW1, but with 1.4x the total storage (14kWh vs 10kWh) and just a 10% increase in weight.

The Powerwall 2 enclosure apparently contains an inverter which must take up a significant amount of volume. It

And the Powerwall 1 is 6.4kWh, not 10kWh.

 

[Garlan Garner]

The Powerwall 2 enclosure apparently contains an inverter which must take up a significant amount of volume. It

And the Powerwall 1 is 6.4kWh, not 10kWh.

Hello ecarfan.

 

[J1mbo]

The Powerwall 2 enclosure apparently contains an inverter which must take up a significant amount of volume. It

And the Powerwall 1 is 6.4kWh, not 10kWh.

PW1 is 10kWh but only 6.4 kWh are usable for daily cycling.

PW1 has a sculpted front which increases the apparent volume when using the simple LxWxH calculation - my assumption is that that extra volume nets off against the space taken by the PW2 inverter.

 

[bredi]

Model 3 will also be lighter.

 

[Yggdrasill]

PW1 is 10kWh but only 6.4 kWh are usable for daily cycling.

PW1 has a sculpted front which increases the apparent volume when using the simple LxWxH calculation - my assumption is that that extra volume nets off against the space taken by the PW2 inverter.

No, the PW1 is 6.4 kWh using NMC cells. There were plans for a 10 kWh PW1 version with NCA cells, but it was discontinued. Why Tesla’s grid batteries will use two different chemistries

The PW2 is likely still using NMC cells, but my match suggests that Tesla must have been able to improve the chemistry by 10-20%. Plus a ~35% improvement in volumetric energy density.

The volumetric energy density improvement should carry over to the Model 3, while the chemistry improvement may carry over, it may partially carry over, or it might not carry over at all. I think I'll try to recalculate the max size of the Model 3 battery again, given the new info surrounding Tesla Energy. My previous calculations got me to around 85 kWh.

 

[Yggdrasill]

Okay, some thoughts and back-of-envelope calculations:

- The Model 3 has 8 modules that are roughly 70 cm x 35 cm, which is roughly half the size of the PW2 battery pack. (My previous estimate was 8 modules with the dimensions 80 x 37 cm.)
- Each of these modules would be 7 kWh using the NMC cells, but assuming the NMC cells are 60-72% as energy dense (depending on how much of the chemistry improvement carries over), they will contain 9.7-11.5 kWh.
- Total pack size is then 77.6-92 kWh. (85 kWh is still a good guess.)
- Pack mass will be around 350 kg.

 

[J1mbo]

No, the PW1 is 6.4 kWh using NMC cells. There were plans for a 10 kWh PW1 version with NCA cells, but it was discontinued. Why Tesla’s grid batteries will use two different chemistries

The PW2 is likely still using NMC cells, but my match suggests that Tesla must have been able to improve the chemistry by 10-20%. Plus a ~35% improvement in volumetric energy density.

The volumetric energy density improvement should carry over to the Model 3, while the chemistry improvement may carry over, it may partially carry over, or it might not carry over at all. I think I'll try to recalculate the max size of the Model 3 battery again, given the new info surrounding Tesla Energy. My previous calculations got me to around 85 kWh.

Thanks for the link - so PW1 is 7kWh with 6.4kWh usable. Interesting that we both came to ~85kWh. Question is, does PW2 still use NMC (DoD suggests they do) or have they moved to NCA?

 

[Yggdrasill]

Thanks for the link - so PW1 is 7kWh with 6.4kWh usable. Interesting that we both came to ~85kWh. Question is, does PW2 still use NMC (DoD suggests they do) or have they moved to NCA?

I don't at all see how the PW2 could have the needed durability with NCA. It would suggest a battery revolution if Tesla has managed to improve durability for NCA by a factor of 5-10.

 

[J1mbo]

Right, is it possible that Tesla will move to NMC for M3? There were good reasons for using NCA in Model S/X, but maybe tweaks to the NMC chemisty have been able to offset that. Can see that the lower cost of manufacture for NMC would have been pretty attractive.

 

[Garlan Garner]

To answer the Topic Title......

I don't think the powerwall 2 reveals anything at all conclusive about the m3 at all.

 

[Yggdrasill]

Right, is it possible that Tesla will move to NMC for M3? There were good reasons for using NCA in Model S/X, but maybe tweaks to the NMC chemisty have been able to offset that. Can see that the lower cost of manufacture for NMC would have been pretty attractive.

I think NCA is substantially cheaper per kWh, so I don't think there's any chance Tesla will use NMC for it's vehicles.

Also, using NMC, my estimate for the Model 3 pack capacity would be 7 kWh x 8 = 56 kWh. That's not great.

 

[Nuclear Fusion]

Assuming the M3 battery is physically 30% smaller than the current MS/MX

Based on dimensional specs, the M3 battery area is 15% smaller than the MS

 

[22522]

I have not seen that the power wall even uses Panasonic. Where is that information?

 

[Yggdrasill]

I have not seen that the power wall even uses Panasonic. Where is that information?

The Gigafactory has been built to supply Tesla Energy and the Model 3 with cells. Google it.

 

[J1mbo]

I think NCA is substantially cheaper per kWh, so I don't think there's any chance Tesla will use NMC for it's vehicles.

Agreed - my cheaper comment was after misreading the profile of NCA vs NMC (here). There is quite a big difference in capacity over PW1, which was also NMC, I wonder if they are using "2nd generation NMC"?

How much of the improvement is down to new chemisty, and how much is down to the improved packaging and cooling?

 

[Electroman]

One thing though is a Powerwall will never get supercharged or super-discharged like how a S or X would. So the cooling requirements are perhaps far less. So a direct translation of specs from Powerwall to a Tesla car would be incorrect.