Battery Capacity Test: Do these numbers make sense ?

[eggy]

Hi guys

I was wondering about the health of my battery pack and wanted to test how much kwh I could draw out of my battery pack.
I charged to 100% (257 rated range) and drove within 3.5h, 235 miles with a remaining 5 miles of rated range, when I pulled into the supercharger in Edison. My dashboard showed me a consumption of exactly 70 kwh. If I add 1.5 kwh for the 5 miles remaining rated rage it sums up to 71.5 kwh usable energy.

According to other TMC postings the usable capacity should be ~75.9kwh

As you can see from my signature below, my BROWNY is about 2 1/2 years old and the current odometer shows 26,300 miles.
What is your opinion about the difference of 4.5 kwh between the theoretical available energy (75.9 kwh) and my test result (71.5 kwh).
Is it a "normal" difference due to the battery degradation caused by age and miles driven ?

Looking forward to your opinions.:biggrin:

Best
Eggy

 

[fwgmills]

That seems correct. I just watched the Bjørn video about world record distance in a P85D and they went 77.5kwh from 100% to a few kilometers past "charge now" using a battery with about 12,000 km on it.

 

[jerry33]

The graphic is someone's guess about an "A" battery pack. My opinion is that it doesn't match reality with any later pack. In particular, there is no zero mile protection anymore--even if there once was.

 

[AmpedRealtor]

The graphic is someone's guess about an "A" battery pack. My opinion is that it doesn't match reality with any later pack. In particular, there is no zero mile protection anymore--even if there once was.

I always thought zero mile protection (buffer) was nothing more than a byproduct of a range algorithm that has drifted significantly out of calibration.

 

[David99]

This graphic is floating around this forum a lot, but it's just a nice visual of someone's guesswork.

The most important thing to understand is that a battery does not have a certain capacity like a bottle filled with liquid. Depending on temperature and how fast or slow you draw the energy from the battery, the capacity is different. It's not a single number, it depends on several conditions. Since driving conditions change a lot, you cannot say you have exactly that much energy in the battery. Slower driving with lower power draw will allow you to get more out of the battery then when driving fast up the hill at high power draw. I'm not talking about range, I'm talking about available energy from the battery.

When my car was new I once drove from 100% to one mile passed zero and got 76.5 kWh out of the battery. How far could I have gone passed zero? I don't know. On a different day I had to go beyond zero for 9 miles to reach a Supercharger. That would be about 2.7 kWh. But the conditions were quite different then.

There will never be an exact answer when it comes to you driving the car as the battery conditions and power draw makes such a difference. Tesla estimates range based on some more or less realistic assumptions. As far as I can tell, this estimate is pretty good.

 

[Troy]

TLDR: 75.9 kWh shown in that graph matches survey data.

The 75.9 kWh number is not just a guess. If I try to calculate the battery capacity of a new 85 kWh pack using trip data from 12 different survey entries, I get the same number. The following data is from this survey. The survey didn't exist at the time the graph was created. The survey belongs to a group of TMC users from the Netherlands. They have a forum topic here.

Let me explain the calculation using the data in the first row:
David started his trip with 250 miles rated range and finished with 11 miles rated range left. He consumed 66.2 kWh energy during this trip. The rated range of this model when new is 265 miles. His consumption was 250-11= 239 rated miles.

If 239 rated miles equals to 66.2 kWh
Then 265 rated miles equals to X
X= 265*66.2/239= 73.4 kWh

In this example the result is a little low. In other examples it is a little higher. The average is 75.89 kWh which is almost exactly the same number. I think the 75.9 kWh number is a credible number. It is not just a guess. They must have done some tests.

Username (can be anything)	Model	Typical range when new	Typical range at the beginning of the trip	Typical range at the end of the trip	kWh consumption for trip	Battery capacity when new
DavidH	Model S 85	265	250	11	66.2	73.40
Muffinman	Model S P85	400	356	0	68.5	76.97
Go4IT	Model S P85	400	366	277	17.1	76.85
Go4IT	Model S P85	400	370	270	18.7	74.80
gelden	Model S 85	400	371	0	68.2	73.53
ErikR	Model S 85	400	371	79	56.5	77.40
Go4IT	Model S P85	400	372	281	17.1	75.16
ErikR	Model S 85	400	376	243	26	78.20
ErikR	Model S 85	400	381	94	54.9	76.52
QueenieEye	Model S 85	400	335	283	9.7	74.62
JimmyJam	Model S 85	400	384	88	57.1	77.16
Dimivdb	Model S 85	400	389	52	64.1	76.08
Average	75.89

 

[stopcrazypp]

I have been reading for a long time that the energy consumption shown does not represent all the kWh that the car actually consumed during the trip. There are some accessories that are not counted. Is this not correct?

Also the zero mile has never been consistent. There are some that reported still having it, but some where zero means zero.

 

[David99]

TLDR: 75.9 kWh shown in that graph matches survey data.

Let me explain the calculation using the data in the first row:
David started his trip with 250 miles rated range and finished with 11 miles rated range left. He consumed 66.2 kWh energy during this trip. The rated range of this model when new is 265 miles. His consumption was 250-11= 239 rated miles.

If 239 rated miles equals to 66.2 kWh
Then 265 rated miles equals to X
X= 265*66.2/239= 73.4 kWh

In this example the result is a little low. In other examples it is a little higher. The average is 75.89 kWh which is almost exactly the same number. I think the 75.9 kWh number is a credible number. It is not just a guess. They must have done some tests.

OK that helps! Actually the example you used is from me It's on the lower side as my car has more than 60k miles on it and some battery degradation has happened. I did this test last week. The first time I did the same test (when the car was virtually new) it came out to 76.5 kWh. But either way, the total amount depends a lot on how you drive. That's why having a single number can be misleading.

The so called 'zero mile protection' and 'bricking protection' must be guessing, though, to add up to 85 kWh. There is no consistent data about how much you can go passed 0. Some have been able to do 18 miles, others none. In rare cases it even shut down before 0. And according to the person who took a battery apart the capacity he measured from a single cell doesn't add up to 85 kWh.

- - - Updated - - -

I have been reading for a long time that the energy consumption shown does not represent all the kWh that the car actually consumed during the trip. There are some accessories that are not counted. Is this not correct?

The energy total consumption on the trip includes everything. It's measured at the battery, so everything is automatically included. For some reason the trip measurement stops counting energy as soon as the car stops. So at every red light or stopped in traffic or when you're parked, the energy the car uses isn't accounted for. I'm not sure why Tesla does it that way.

 

[Troy]

Hi David. I didn't know that was you. That's some coincidence! I see that you are interested in battery related subjects. Let me add some interesting data about what you said here:

And according to the person who took a battery apart the capacity he measured from a single cell doesn't add up to 85 kWh.

There is a PDF file about this subject here. They took measurements from the actual Model S cells too and the rated capacity adds up almost exactly to 85 kWh. The exact voltage numbers in the PDF are behind a pay wall therefore I don't want to use them. Without using the exact numbers, the calculation is similar to this:

Rated Capacity: 7104 x ~3.7V x 3.25Ah = 85426 Wh = 85.43 kWh
Actual Capacity: 7104 x ~3.6V x 3.1Ah = 79280 Wh = 79.28 kWh

Update: Let me more info:
The numbers are on page 12. 7104, 3.25 Ah and 3.1 Ah are visible. To read the rated and actual voltage numbers you can copy and paste that section.

 

[David99]

That's interesting! Thanks for the info. I'm aware that a battery that is rated at 85 kWh won't have that available for daily use unless you want to kill it quickly. Getting 76.5 out of it for daily use is actually pretty good considering how long these cells last.

BTW I really wish more people would add their data to the google sheets. There is just 50 from the US. But the US had the first Teslas delivered so there should me many with lots of miles on them. I'm really interested to see how the battery is on those having 200k miles and more.

 

[stopcrazypp]

Hi David. I didn't know that was you. That's some coincidence! I see that you are interested in battery related subjects. Let me add some interesting data about what you said here:



There is a PDF file about this subject here. They took measurements from the actual Model S cells too and the rated capacity adds up almost exactly to 85 kWh. The exact voltage numbers in the PDF are behind a pay wall therefore I don't want to use them. But you can read the numbers if you copy and paste. Without using the exact numbers, the calculation is similar to this:

Rated Capacity: 7104 x ~3.7V x 3.25Ah = 85426 Wh = 85.43 kWh
Actual Capacity: 7104 x ~3.6V x 3.1Ah = 79280 Wh = 79.28 kWh

I didn't see that in the first go through, I only noted the higher nominal voltage does not match NCA cells at 3.6V. However, that report uses a 4.35V charging voltage to get to 85kWh. Tesla uses a 4.2V charging voltage and Panasonic does not make an NCA cell that uses a 4.35V charging voltage.

 

[David99]

I didn't see that in the first go through, I only noted the higher nominal voltage does not match NCA cells at 3.6V. However, that report uses a 4.35V charging voltage to get to 85kWh. Tesla uses a 4.2V charging voltage and Panasonic does not make an NCA cell that uses a 4.35V charging voltage.

Yes Tesla charges the Model S (85 battery) to 403 Volt which comes out to 4.2 Volt per cell. According to the measurement, this would yield a total capacity of 80 kWh. Being able to pull 76.5 out is pretty impressive then. But it also shows that the image in the original post is definitely wrong. Knowing that the Model S only charges to 4.2 Volt, the capacity of the battery is 80 kWh at max. Knowing we can get up to 76/77 kWh out of it leaves no room for zero mile protection and bricking protection as the image suggests.

 

[stopcrazypp]

Yes Tesla charges the Model S (85 battery) to 403 Volt which comes out to 4.2 Volt per cell. According to the measurement, this would yield a total capacity of 80 kWh. Being able to pull 76.5 out is pretty impressive then. But it also shows that the image in the original post is definitely wrong. Knowing that the Model S only charges to 4.2 Volt, the capacity of the battery is 80 kWh at max. Knowing we can get up to 76/77 kWh out of it leaves no room for zero mile protection and bricking protection as the image suggests.

I'm just not convinced if it was empirically determined rather than back calculated from known assumptions. The 3.6V and 3.1 volt spec was what Tesla announced in 2010:
http://www.engadget.com/2010/04/23/panasonics-3-1ah-batteries-to-be-used-in-the-tesla-model-s-hav/

Assuming it was empirically determined, the Tesla pack is actually 80kWh using standard capacity rating conventions. While a 4.2V cell can be charged at 4.35V, it dramatically shortens the lifetime unless it was designed for it. Panasonic has no NCA cells that are specified as suitable for 4.35V charging (nor do other manufacturers; it is the manganese based cells that are designed for that kind of charging). Thus Panasonic's spec sheet never specifies capacity assuming a 4.35V charging voltage.

When I have the time I can overlay some discharge curves from the one wk057 did on a Model S cell and see what matches better.