(P)100D Battery capacity question

[quenbo]

Hello,
recent owner of a P100D, I have some questions about battery that I'd like to submit and validate by you.

I have read that the 100D has a "real" battery capacity of 102,4 kWh.

But, from what I have read and understood, 4 kWh are reserved for the car electronic to avoid bricking.
=> 102,4 - 4,0 = 98,4 kWh are available for driving.

I have also read somewhere that some tesla sellers tells customers that there is a reserve of 30 km @70kmh below the calculated zero range, i.e. 4 kWh
=> 98,4 - 4,0 = 94,4 kWh

Can someone please confirm that the "real life" capacity (above zero range) is thus 94,4 kWh, and that these 94,4 kWh are use to calculate range ?

I got my inventory P100D last week (with 3300 kms) and I have a 100% typical calculated range of 482 kms. I'd like to calculate the range loss in % due to battery degradation.

if i look in the car, i got
- Range based on last 10 km (241 Wh/km) is 102 km => 24,4 kWh remains in the pack
- Range based on last 25 km (198 Wh/km) is 124 km => 24,6 kWh
- Range based on last 50 km (241 Wh/km) is 97 km => 24,6 kWh
My calculated typical range is 129 km, so I conclude that the typical range is calculated on 190 Wh/km basis (24500 Wh / 129 km)

I have today (without eco mode) a typical range of 482 km.
Considering typical basis of 190 Wh/km, battery capacity is 482 x 0,19 kWh = 91,58 kWh, which is 97% of theoretical 94,4 kWh and a lost of 3%.
But if I'm wrong in my reflexion and that the battery capacity is 98,4 kWh, then I have a lost of 7% which seems huge.

Can someone tell me if I'm right or wrong?

 

[Tiger]

If you have a P and if you drive like me, the thing that makes most of the difference is your foot on the gas pedal. Just keep chargers close on your planned trip and enjoy the ride!

 

[quenbo]

Very strange...

I had another look in the car 4 h after and the typical range basis is different from my previous post :

- Range based on last 10 km (266 Wh/km) is 219 km => 58,3 kWh remains in the pack
- Range based on last 25 km (220 Wh/km) is 265 km => 58,3 kWh
- Range based on last 50 km (225 Wh/km) is 260 km => 58,6 kWh

The calculated typical range is 292 km, so now the typical range is based on 200 Wh/km (58400 Wh / 292 km)

 

[bob_p]

Since receiving our S 100D in late March, it appears Tesla has made several updates to the software estimating the current charge. Originally we were getting about 306-307 miles at 90% (which put the 100% charger over 340 miles), And after some of the software updates, we noticed that the 90% charge has dropped a little, at least once, possibly twice - so we're now getting 296-298 miles at 90%.

Since the 90% charge level dropped quickly, it seems more likely the change was due to software - and not to actual battery capacity loss.

With our S P85, we've seen about 5% range loss over time, and I expect to see something similar with the 100D - but not for a couple of years...

 

[Troy]

Hi, @quenbo. The 102.4 and 98.4 kWh look correct but I don't think your assumption about hidden capacity is correct. Also, Typical range is not based on a universal Wh/km number for all Model S variants and it is not based on any tests. If you want to know how your degradation compares to other Teslas, you could enter your data here and then go to the Charts tab and select your username.

 

[SucreTease]

I have also read somewhere that some tesla sellers tells customers that there is a reserve of 30 km @70kmh below the calculated zero range, i.e. 4 kWh....
Can someone tell me if I'm right or wrong?

This is not correct. Others have written about that here (see below). When the car tells you that it is at zero % left (by its best estimate), it will shut down any moment. Tesla only reserves sufficient charge to protect the battery from damage; if a battery cell is completely drained, it can reverse polarity. Battery charge is estimated by measuring voltage (the only instantaneous metric that can obtained from the pack).

"Zero" means "Zero" - Caution about Vegas to LA trip!
How to read the Energy chart?
Latest thinking on buffer after "Charge Now" screen appears

 

[David99]

when the car tells you that it is at zero % left (by its best estimate), it will shut down any moment.

There is a chance it will, but more times than not it will continue to go for a little more. It depends on a lot of things, but having zero miles left on the estimate does not always mean the car will shut down. Many many examples where it did not.

It does not mean there is a buffer than you can use.

Battery charge is estimated by measuring voltage (the only instantaneous metric that can obtained from the pack).

That is not correct. Battery capacity is estimated by keeping track of the energy that went in and the energy that was taken from it. Since you can't directly measure the remaining capacity of a battery you have to measure the energy that was charged and discharged. You can get a pretty good idea of the capacity doing so if you also consider charge/discharge rate and temperature together with the known battery efficiency. The algorithm can be a little off especially when partially charge and discharging the battery. The car will always estimate on the lower side to be safe. That's why often you can go a little beyond zero. It's not a buffer, it's just a margin of error that is most of the time on your side.

 

[SucreTease]

That is not correct. Battery capacity is estimated by keeping track of the energy that went in and the energy that was taken from it.

Please read what I said more carefully. What I said is indeed correct: voltage is the only instantaneous metric you can obtain from a battery of this type.

Keeping track of extracted current is a referring to a history of past measurements—not an instantaneous measurement of the battery; this method is inherently inexact, as it doesn't account for internal losses, which are not constant across charge level.

 

[David99]

Please read what I said more carefully. What I said is indeed correct: voltage is the only instantaneous metric you can obtain from a battery of this type.

No, you wrote this: Battery charge is estimated by measuring voltage (the only instantaneous metric that can obtained from the pack).

This is wrong on two levels. First, that's not how the BMS calculates the charge of the battery. Voltage depends on many factors. It is one rough indicator of many others. It is also not the only insantenous metric you can get. Impedance can also be measured which makes the mentioned measurement of energy in and out pretty accurate. And that's how the BMS estimated range.

 

[SucreTease]

Voltage depends on many factors. It is one rough indicator of many others. It is also not the only insantenous metric you can get. Impedance can also be measured which makes the mentioned measurement of energy in and out pretty accurate.

And how do you measure the impedance (internal resistance) of a battery? You measure its output voltage with and without a reference load on it. And how do you measure its current output? You measure the voltage drop across a small reference resister in series with the load. All measurements of a battery come down to measuring output voltage, sometimes in combination with another measurement. Voltage is the only direct measurement. All other measurements (impedance, current, etc.) are measured indirectly through other voltage measurements.

I should warn you: you are arguing basic E&M with a physicist. ;-)

 

[quenbo]

So, thanks to all for your responses (and debate ).
I learned a lot and do understand now that the zero mile reserve isn't really "programmed to be reserved by the car", thank you !

So after some measure and experiments, I noticed that the typical RR is sometimes calculated by the car on the basis off 190 Wh/km, or sometimes 200 Wh/km.
(il looks like it depends on the remaining charge level, but I'm not 100% sure about that)

So, as I have a 482 km typical range at 100% charge, I assume my current battery capacity is 482 * 0,2 = 96,4 kWh.

If I consider the 4 kWh reserve for the electronic to avoid bricking
=> 102,4 - 4,0 = 98,4 kWh are theoreticaly available for driving.

So 96,4/98,4 = 98%
=> I finally know that I have 2% of battery capacity loss after 5000 km !

 

[Viking1]

Does anyone know if the 100 battery has the silicon anode like the 90 battery, or did they go back to the old chemistry (i.e. 85 battery)?

 

[quenbo]

Does anyone know if the 100 battery has the silicon anode like the 90 battery, or did they go back to the old chemistry (i.e. 85 battery)?

Yes I think so.
Battery for P100D is, from what I read :

- 600 kg (540 kg for P85D)
- 8256 cells (7104 for P85D)
- 12.4 Wh/cells (11.5 for P85D and 12.1 for 90D)
- improve chemistry with silicon
- 171 Wh/kg of battery (152 for P85D)

 

[widodh]

Yes I think so.
Battery for P100D is, from what I read :

- 600 kg (540 kg for P85D)
- 8256 cells (7104 for P85D)
- 12.4 Wh/cells (11.5 for P85D and 12.1 for 90D)
- improve chemistry with silicon
- 171 Wh/kg of battery (152 for P85D)

12.4Wh per cell is that with 3.6V or 3.7V?

Those cells would be 3.3 or 3.4Ah then. These INR/NCR cells can be purchased from Samsung or LG online as well.

See: NKON | Samsung INR18650-35E 3450mAh - 10A - 18650 size - Rechargeable batteries

8256 of these cells would give you 102kWh

3.6V * 3450mAh * 8256 = 102.5kWh

 

[quenbo]

12.4Wh per cell is that with 3.6V or 3.7V?

I don't know the voltage.
12,4 Wh is indeed the battery capacity (102420 Wh) divided by 8256 cells

 

[Viking1]

If the 100 battery has the same new chemistry as the 90 (silicon in anode) then it will be interesting to see its degradation.

However, whereas the 85 pack and 90 pack have the same number of batteries and the same pack design (resulting in higher temps for the 90), from what I've read the 100 pack has more batteries and is a new design. Better for removing the battery heat build up. For that reason, I'm thinking the 100 pack may have a better degradation record then the 90. Time will tell.

 

[David99]

If the 100 battery has the same new chemistry as the 90 (silicon in anode) then it will be interesting to see its degradation.

However, whereas the 85 pack and 90 pack have the same number of batteries and the same pack design (resulting in higher temps for the 90), from what I've read the 100 pack has more batteries and is a new design. Better for removing the battery heat build up. For that reason, I'm thinking the 100 pack may have a better degradation record then the 90. Time will tell.

The cooling system in the 'old' battery pack design is just fine. It is very capable and keeping the battery at a reasonable temperature. I have monitored the battery temperature on the CAN bus for over a year now in all kinds of conditions. The reason the 100 packs have a different cooling design is to allow more space for the extra cells.

 

[Viking1]

The cooling system in the 'old' battery pack design is just fine....

That may be true. But from reading this forum and other sources, it seems the 90s experience a higher rate of battery degradation then the 85s. If the pack was designed for the older battery chemistry, then it seems to me it would not be optimal for the new chemistry batteries. I have no evidence of that. Just observation of what I see reported for degradation.

 

[Lanber]

The only way to know how much is hidden below 0 range is to drive till it stops ;-)

That is somewhat impractical. You can however read out what the car estimates remains in the battery with a CanBus reader.
My 2013 S85 is very stable over time with both range and hidden reserve, it is reported to be 3.4kwh there. Have driven it down to -2kwh without issues to see if range calculated would go up. It did with 1km, ie nothing.

 

[David99]

That may be true. But from reading this forum and other sources, it seems the 90s experience a higher rate of battery degradation then the 85s. If the pack was designed for the older battery chemistry, then it seems to me it would not be optimal for the new chemistry batteries. I have no evidence of that. Just observation of what I see reported for degradation.

Definitely, from what we can see the 90 definitely seem to age quicker. It has nothing to do with running hotter, though. The cooling system has software set temperature targets, if the new cells needed different temperatures it would be set so in the software. The system can use up to 6 kW of either cooling or warming power to keep the battery well. It's not like the cooling system has one setting that matched the old cells and Tesla just didn't bother to adjust it for the new cells.