Baseline for HV Battery SoH Performance

[ciaranob]

It's common practice (in my non-EV but highly battery related industry) to use "equivalent full cycles" as a figure of merit. Essentially, energy throughput divided by nominal energy capacity. To calculate this, you'd need lifetime efficiency averages from users as well as mileage.

Hey - so I've determined that I've 'charged in' ca 3500 kWh of energy to my battery to date over ca 2 years - 83 AC charges and 16 DC charges - with mileage at 7620.

The Porsche 'advertised' spec (accessible) capacity of the PB+ is 83.7 kWh of the 93.4 kWh battery pack - we know from other's work (see Post 1) that the 'real world' accessible/usable capacity is more likely closer to 86.7 kWh.

Cycles = [total kWh charged into the battery] / [net battery capacity]

@83.7 I'd have the equivalent of ca 42 full charge cycles done to date
@86.7 I'm closer to 40 equivalent charge cycles.

Do you then simply compare extracted Soh to the #cycles for degradation rates etc. - example for my car it equates to losing 0.3 Soh through every 'equivalent' charge cycle (and that number does make me squirm a bit!)?

I'm betting there is a more sophisticated measure to evaluate Soh against this i.e. given endless chemistries in different battery designs there prob is no std value to apply how increasing charge cycles relate to SEI build-ups or other chemical transformations (oxides) that would impact/reduce capacity over time - or is there? Clearly I am WAY out of my technical expertise but always up to learn !

MyPorsche app seems to suggest my average efficiency close to 2.7 kWh per mile.
However 3500/7620 yields a value closer to 2.2 kWh per mile!

Can't recall offhand what the expected no. of full cycles the Porsche battery is supposed to handle before any significant degradation - guess we can estimate a minimum from the 3 and 8 year warranty min. capacity values they publish (70% and 80%) but perhaps someone here has some data on that?

Cheers, C.

PS: as I dig into this a bit including other threads here in the Forum, the prediction of Soh decline is extremely complex and dependent on a host of variables - so poss. a bit unrealistic to ask above question(s)!

Sponsored

 

[800v]

Hi all, here is an update on my "bad" previous reading from Dec 4, 2022, where the SOH was 88.14% with 5810 miles (Turbo 2020 model).

Now, Sep 9 2023, the SOH is 88.39% (!) with 8060 very happy miles.

Great, the SOH increased a bit ! I almost exclusively use fast charging to about 85%, but I saw one cell was a bit less than the others (this post).
Recently I charged a few times to 100% using a slow charger to entice the battery to rebalance the cells (I read somewhere this happens mostly when charging to 100%). Maybe that helped increase the SOH ?

ps. Also, the difference between the best (3.878V) and worst cell (3.868v) is about 0.010v -- that seems fine (You can read this on page 7 in the dashboard of carscanner). This may be good to monitor given that @Scandinavian measured a rather big difference at some point (0.264v !), and right after had a battery failure .

(@ciaranob : maybe we should ask people in this thread to also report the voltages of the best and worst cell (page 9 of dashboard, or top of page 7) so we can track the differences?)

 

[ciaranob]

@2P168S
@800v

Your latest data added to the main charts in Post 1.

 

[Yaibal]

Ok, I just got my Vlink device. My Taycan was recently purchased used. It’s a 2022 Taycan 4S CT. I had the vehicle delivered to me in TN, but I believe it was purchased originally at Porsche exchange and sold to me from Porsche Danbury in Connecticut. So I think it’s seen mostly northern US weather.

1. unknown
2. 90.55
3. 13,300 Miles
4. Unknown (the app won’t show me the previous owners history)
5. Build 12/21 delivery 2/22

Now, my wife drives a 2022 Audi Etron S. We purchased it new in 8/22. I also hooked the vlink up and looked over the info provided. This is just anecdotal, but I figured it’d be interesting to provide.

1. unknown
2. 6524.24% (what does this mean
3. 10,700 miles
4. 272 kWh
5. Delivery 8/22

Another interesting detail is the SoC hr is lower (97.25) than the doc display which shows 100%. The profile for the car scanner app is the same for the etron or the taycan. It’s interesting the info is reported a bit differently. Thoughts?

cheers,
Jordan

so it’s been a couple of months so here’s an update.


1. unknown
2. 88.39
3. 15,316 Miles
4. Aprox. 80kwh since delivery in 06/23.
5. Build 12/21 initial delivery 2/22

I don’t love to see the SoH dropping more than 2 percentage points. I may try a 100% charge to see what effect that has. I’ve kept the car at 85% charge with the only exception being a dealer visit where they charged it to 100 via dc fast charging.

the other factor is temperature. I live in a warm part of the us. The summer has been miserably hot.

 

[2P168S]

Hey - so I've determined that I've 'charged in' ca 3500 kWh of energy to my battery to date over ca 2 years - 83 AC charges and 16 DC charges - with mileage at 7620.

The Porsche 'advertised' spec (accessible) capacity of the PB+ is 83.7 kWh of the 93.4 kWh battery pack - we know from other's work (see Post 1) that the 'real world' accessible/usable capacity is more likely closer to 86.7 kWh.

Cycles = [total kWh charged into the battery] / [net battery capacity]

@83.7 I'd have the equivalent of ca 42 full charge cycles done to date
@86.7 I'm closer to 40 equivalent charge cycles.

Do you then simply compare extracted Soh to the #cycles for degradation rates etc. - example for my car it equates to losing 0.3 Soh through every 'equivalent' charge cycle (and that number does make me squirm a bit!)?

I'm betting there is a more sophisticated measure to evaluate Soh against this i.e. given endless chemistries in different battery designs there prob is no std value to apply how increasing charge cycles relate to SEI build-ups or other chemical transformations (oxides) that would impact/reduce capacity over time - or is there? Clearly I am WAY out of my technical expertise but always up to learn !

MyPorsche app seems to suggest my average efficiency close to 2.7 kWh per mile.
However 3500/7620 yields a value closer to 2.2 kWh per mile!

Can't recall offhand what the expected no. of full cycles the Porsche battery is supposed to handle before any significant degradation - guess we can estimate a minimum from the 3 and 8 year warranty min. capacity values they publish (70% and 80%) but perhaps someone here has some data on that?

Cheers, C.

PS: as I dig into this a bit including other threads here in the Forum, the prediction of Soh decline is extremely complex and dependent on a host of variables - so poss. a bit unrealistic to ask above question(s)!

Close-

Cycles = [total energy expended by vehicle] / [gross battery capacity]

If you use the total amount charged, you will include the charging losses. Don't want to do that. So have the numerator be the total energy used by the vehicle, calculated by mileage and average lifetime efficiency. For the denominator, you technically want to use the gross battery capacity, though if you are only comparing Taycans to Taycans it won't really matter. If you were to ever compare degradation of a Taycan to another vehicle, calculating cycles based on gross capacity becomes important since every vehicle has different buffer sizes.

And yes, the method then is simply to replace the miles in your analysis with the number of cycles. This should reduce the noise in the result since degradation is more closely correlated to cycles than miles. As for doing more sophisticated analyses from usage data, this is an active area of both public and private research.

 

[Jupeman]

Where/can one see SoH in PIWIS? I was plugged my car in today and looked at PIWIS (v40) for the first time today but couldn’t figure out if there was a place to view SoH.

 

[ciaranob]

Close-

Cycles = [total energy expended by vehicle] / [gross battery capacity]

If you use the total amount charged, you will include the charging losses. Don't want to do that. So have the numerator be the total energy used by the vehicle, calculated by mileage and average lifetime efficiency. For the denominator, you technically want to use the gross battery capacity, though if you are only comparing Taycans to Taycans it won't really matter. If you were to ever compare degradation of a Taycan to another vehicle, calculating cycles based on gross capacity becomes important since every vehicle has different buffer sizes.

And yes, the method then is simply to replace the miles in your analysis with the number of cycles. This should reduce the noise in the result since degradation is more closely correlated to cycles than miles. As for doing more sophisticated analyses from usage data, this is an active area of both public and private research.

OK - quick and dirty calculation results of what you advised (which I should have figured out - lazy - thx!) and my potential of dipping below 80% Soh by the end of my 3rd year of ownership. Results ENTIRELY dependent on the validity/accuracy of the OBDII tool's 'flash' readouts. I will have a data point for my car from a PIWIS Soh readout hopefully this week (car now at dealership for different reason) to compare against my last OBDII readout.

For those reading, BEC here is 'Battery Equivalent Cycle' i.e. the energy in to complete a 100% charge cycle (despite in our case comprising multiple charge events) and that energy used for miles driven.

The correct calculation for the number of BEC’s (using energy used by the car excluding losses - thx again to @2P168S ) for an Soh drop of 12% over 7668 miles for my specific car gives a Soh decline rate of 0.38% per BEC. A near identical result was obtained by using an average BEC mileage range of 250 miles which yielded 1.3 less cycles. This decline rate is 3 times higher than that calculated (in a simple linear sense) using the basic warranty mileage-time numbers i.e. Porsche’s Soh decline rate is ca. 0.1% per BEC.

The total energy I have charged into my car is 3500.6 kWh (over 7668 miles) - however this total kWh number includes losses to/within the car. Using the average lifetime efficiency for my car from MyPorsche app (I can prob later indeed calculate this to compare), I can calculate the total energy used by the car to gain these miles (less those losses) which is 2814.2 kWh. This implies losses are accounting for ca 20% of all energy put into the vehicle (seems high!?).

Using again the App's lifetime efficiency data i.e. energy used by the car to result in my 12% Soh drop, if I maintain my current use-charge rates I would add another ca 16 BECs by end of year 3, a total of 48 BECs over 3 years and have less than 12000 total mileage. This (assuming a linear decline rate - almost certainly not!) would shave another 6% off my Soh bringing it to 82% before 3 years are up!

Clearly if, as we actually plan to do, travel a much greater distance in 2024, I will be at a much higher risk of going under the 3 year/37,500 mile 80% Soh battery warranty date.

We plan at least two ca. 3000 mile road trips next year so mileage for 2024 could be ca 10,000 (6000+4000). This is equates to ca. 40 additional BECs in 2024. If Soh decline again was linear (not likely) this would subtract another 15% off my Soh i.e. total Soh drop after 3 years and ca 18,000 miles of 27% to a potential SoH reading of 73%!! Most industry data shows the rate of Soh decline tends to flatten out rapidly close to an Soh of 90% - nonetheless my apparent higher than normal Soh decline rate still predicts a high risk of going below 80% Soh before my 3 years are up.

Possibly all recalibrated once I receive the results of the PIWIS SoH test!

Here is another view of the Forum SoH data to date in respect SoH and warranty coverage - I'd say Porsche has left ca 5% buffer for even the worst case scenario (like mine or other Hot climate zone cars) in the bank ...... but maybe I'll break that trend!

 

[Daniel]

@ciaranob Thank you for this very good work

Sad result here ... the same SoH with 50% or 100% SoC


1) unknown
2) 84.96
3)17030 miles 27400 km
4) 715 kWh DC charger,mainly 11 kw at home from 30 % to 85 % , never under 10 %
-About 15 charges to 100 % before long trip or after minutery bug... , but always driving immediatly after
5) july 2021 delivery september CT4

44° 30' altitude 800 m , beginning of the warm zone

ODBlink CX Model CX101 and Car Scanner

 

[ciaranob]

@ciaranob Thank you for this very good work

Sad result here ... the same SoH with 50% or 100% SoC


1) unknown
2) 84.96
3)17030 miles 27400 km
4) 715 kWh DC charger,mainly 11 kw at home from 30 % to 85 % , never under 10 %
-About 15 charges to 100 % before long trip or after minutery bug... , but always driving immediatly after
5) july 2021 delivery september CT4

44° 30' altitude 800 m , beginning of the warm zone

ODBlink CX Model CX101 and Car Scanner

Thx Daniel - uploaded and your entry rounded to 85% on the chart - but indeed quite low!

 

[andb]

@Daniel Its a bit strange to have 15% battery loss after two years at very low mileage with your car.
I wonder if you notice reduced range during your driving. I mean your should see around 60 km less range on average.

 

[ciaranob]

@Daniel Its a bit strange to have 15% battery loss after two years at very low mileage with your car.
I wonder if you notice reduced range during your driving. I mean your should see around 60 km less range on average.

Just for ref - he is not alone - I have 12% SoH loss in less than half his mileage over 2 years at 7600 miles!

As per my last post above you can see that that rate of degradation is ca 3 times that calculated albeit roughly for the warranty degradation rates!

I’ll be getting a PIWIS SoH reading when I get the car back from the dealership this week - will be interesting to compare to the OBDII readouts.

And so far in 2023 I’m seeing close to 10% less range in average than I attained in 2022.

PS: all the data to date are confirming that we’re seeing the worst degradation rates in hot climates - but we’re talking percentage pt differences here so all in context.

 

[SergeyIndy]

I only charge at home on 48Amp out and car receiving solid just below 11kW during charging.
I use Timer function to charge overnight from 9pm to 6am.

1) Delivery SoH = 99.95% (Measured July 1, 2023)
2) Current SoH = 95.00% (Measured September 15, 2023)
3) Current mileage = 1008
4) kWh DC charging: 1 time during PDI to 100% and 1 time at EA as test to go from 10% to 35%
5) Build date April 2023

This is a pretty dramatic drop for the first 1,000 miles.
I assume to be expected, before it slows down.

Also, interesting to see the cell voltage from my last reading at 500 miles (August 18, 2023):

500 miles read: Min 3.869 Max 3.875 Avg 3.8716
1000 mile read: Min 4.012 Max 4.023 Avg 4.018

 

[Scandinavian]

Also, interesting to see the cell voltage from my last reading at 500 miles (August 18, 2023):

500 miles read: Min 3.869 Max 3.875 Avg 3.8716
1000 mile read: Min 4.012 Max 4.023 Avg 4.018

Great data to add to the collection.

I would not pay any attention to the cell voltage. If you have that page displayed when you drive you will see how it dips when you press the accelerator, and how it increases when regen kicks in, especially if you put your foot on the brakes.

You need to have exactly the same SOC if you like to observe one value against the other. It varies with the SOC etc.

 

[ciaranob]

Great data to add to the collection.

I would not pay any attention to the cell voltage. If you have that page displayed when you drive you will see how it dips when you press the accelerator, and how it increases when regen kicks in, especially if you put your foot on the brakes.

You need to have exactly the same SOC if you like to observe one value against the other. It varies with the SOC etc.

Exactly - as stated before. in this thread the cell voltages entirely a function of SoC - more useful to look at comparatively between cells/modules to check for anomalies.

 

[ciaranob]

I only charge at home on 48Amp out and car receiving solid just below 11kW during charging.
I use Timer function to charge overnight from 9pm to 6am.

1) Delivery SoH = 99.95% (Measured July 1, 2023)
2) Current SoH = 95.00% (Measured September 15, 2023)
3) Current mileage = 1008
4) kWh DC charging: 1 time during PDI to 100% and 1 time at EA as test to go from 10% to 35%
5) Build date April 2023

This is a pretty dramatic drop for the first 1,000 miles.
I assume to be expected, before it slows down.

Also, interesting to see the cell voltage from my last reading at 500 miles (August 18, 2023):

500 miles read: Min 3.869 Max 3.875 Avg 3.8716
1000 mile read: Min 4.012 Max 4.023 Avg 4.018

My first OBDII reading was at mileage of 4000 and it had plummeted to 92.7! Hopefully yours indeed flattens out around 90 - mine has not.

Sponsored