Interested in an unofficial Apple Watch app?

[electricsidecar]

The battery modules appear to be read in modules, with one module returning a block of data in the following CAN format:

// 7ED102E62187051916E
// 7ED21394145454A916E
// 7ED2239918C3982916E
// 7ED23399182398C916E
// 7ED2439916E398C916E
// 7ED253991823964916E
// 7ED263991643878AAAA

Which translates to the following data bytes:

62187051916e394145454a916e39918c3982916e399182398c916e39916e398c916e3991823964916e3991643878

62 18 70 appears to be the header; though I'm not sure what the 62 represents. Still working on decoding the remainder of the data.

Sponsored

 

[electricsidecar]

Rough guestimate of data meanings so far:

// 83  91 82 39 41 45 45 4a  91 64 38 91 78 39 6e  91 5a 38 91 64 38 78  91 64 38 91 64 38 6e  91 46 38 91 6e 38 5a  91 64 38 91 46 38 64
// 83  91 78 39 41 45 45 4a  91 64 38 91 78 39 6e  91 64 38 91 64 38 78  91 6e 38 91 64 38 6e  91 46 38 91 6e 38 5a  91 64 38 91 46 38 64
// 83  91 82 39 41 45 45 4a  91 6e 38 91 78 39 6e  91 64 38 91 64 38 78  91 6e 38 91 64 38 6e  91 46 38 91 6e 38 5a  91 64 38 91 46 38 64
// 83  91 6e 38 41 45 45 4a  91 50 38 91 6e 38 6e  91 5a 38 91 5a 38 78  91 5a 38 91 5a 38 6e  91 46 38 91 64 38 5a  91 5a 38 91 3c 38 64
// 83  91 6e 38 41 45 45 4a  91 64 38 91 6e 38 6e  91 5a 38 91 5a 38 78  91 5a 38 91 5a 38 6e  91 3c 38 91 64 38 5a  91 5a 38 91 3c 38 64
// 83  91 6e 38 41 45 45 4a  91 64 38 91 6e 38 6e  91 50 38 91 5a 38 78  91 64 38 91 5a 38 6e  91 32 38 91 64 38 5a  91 50 38 91 3c 38 64
// 83  91       41 45 45 4a  91    38 91       6e  91    38 91    38 78  91    38 91    38 6e  91    38 91    38 5a  91    38 91    38 64
//     3.723 56 1.6709       3.72  56 3.723 1.4446 3.72  56 3.721 1.4456 3.722 56 3.721        3.717 56 3.722        3.72  56 3.718

 

[electricsidecar]

I think that Car Scanner might be showing only half of the battery data. Based on what I can intuit from the raw data, these packets seem to show 12 cells, not 6, which aligns with Porsche's newsroom article: https://media.porsche.com/mediakit/taycan/en/porsche-taycan/die-batterie

> The two-deck Performance Battery Plus (standard in the Taycan Turbo and Taycan Turbo S) contains 33 cell modules consisting of 12 individual cells each (396 in total). There are five fewer cell modules in the Performance Battery

I have the performance battery and Car Scanner's correctly showing 33 modules but only 6 cells per module. If I'm reading the raw data right, there's actually 12 cells being reported per module.

// module  1: 83  91 82 39  41 45 45  4a  91 64 38  91 78 39  6e  91 5a 38  91 64 38  78  91 64 38  91 64 38  6e  91 46 38  91 6e 38  5a  91 64 38  91 46 38  64
// module  1: 83  91 78 39  41 45 45  4a  91 64 38  91 78 39  6e  91 64 38  91 64 38  78  91 6e 38  91 64 38  6e  91 46 38  91 6e 38  5a  91 64 38  91 46 38  64
// module  1: 83  91 82 39  41 45 45  4a  91 6e 38  91 78 39  6e  91 64 38  91 64 38  78  91 6e 38  91 64 38  6e  91 46 38  91 6e 38  5a  91 64 38  91 46 38  64
// module  1: 83  91 6e 38  41 45 45  4a  91 50 38  91 6e 38  6e  91 5a 38  91 5a 38  78  91 5a 38  91 5a 38  6e  91 46 38  91 64 38  5a  91 5a 38  91 3c 38  64
// module  1: 83  91 6e 38  41 45 45  4a  91 64 38  91 6e 38  6e  91 5a 38  91 5a 38  78  91 5a 38  91 5a 38  6e  91 3c 38  91 64 38  5a  91 5a 38  91 3c 38  64
// module  1: 83  91 6e 38  41 45 45  4a  91 64 38  91 6e 38  6e  91 50 38  91 5a 38  78  91 64 38  91 5a 38  6e  91 32 38  91 64 38  5a  91 50 38  91 3c 38  64
// module  1: 83  91        41 45 45  4a  91    38  91        6e  91    38  91    38  78  91    38  91    38  6e  91    38  91    38  5a  91    38  91    38  64
// module  1:     3.723 56  1.671 60      3.72  56  3.723 56      3.72  56  3.721 56      3.722 56  3.721 56      3.717 56  3.722 56      3.72  56  3.718 56

I'm still trying to figure out what the delimiter bytes mean but I'm about 70% confident in the voltage + percentage interpretations. Will need to change the car's voltage a bit and re-measure to see if the values still make sense.

 

[electricsidecar]

Module 33's codes for comparison:

// module 33: 51  91 6e 39  41 45 45  4a  91 6e 39  91 8c 39  82  91 6e 39  91 82 39  8c  91 6e 39  91 6e 39  8c  91 6e 39  91 82 39  64  91 6e 39  91 64 38  78

 

[electricsidecar]

I'm quite curious to see what the voltage map looks like over time and how power spreads around the cells as you drive and charge the car. Once I get everything confidently decoded I've got some pretty fun ideas for visualizing the data nicely

 

[romed]

as i need for another documentation, following the ECU list for EV / MHV control modules. highlighted are mounted in Taycan. ECU19 = Gateway lots of data in also as alredy posted. Maybe it helps as overview

21 = Battery Energy Control Module 2
C4 = DC/DC Converter Control Module
C5 = Thermal Management
C6 = Battery Charger Control Module
8C = Battery Energy Control Module
BD = High Voltage Battery Charge Management
8105 = DC/DC Converter Control Module (MEB Plattform)

 

[romed]

I think that Car Scanner might be showing only half of the battery data. Based on what I can intuit from the raw data, these packets seem to show 12 cells, not 6, which aligns with Porsche's newsroom article: https://media.porsche.com/mediakit/taycan/en/porsche-taycan/die-batterie

> The two-deck Performance Battery Plus (standard in the Taycan Turbo and Taycan Turbo S) contains 33 cell modules consisting of 12 individual cells each (396 in total). There are five fewer cell modules in the Performance Battery

I have the performance battery and Car Scanner's correctly showing 33 modules but only 6 cells per module. If I'm reading the raw data right, there's actually 12 cells being reported per module.

// module  1: 83  91 82 39  41 45 45  4a  91 64 38  91 78 39  6e  91 5a 38  91 64 38  78  91 64 38  91 64 38  6e  91 46 38  91 6e 38  5a  91 64 38  91 46 38  64
// module  1: 83  91 78 39  41 45 45  4a  91 64 38  91 78 39  6e  91 64 38  91 64 38  78  91 6e 38  91 64 38  6e  91 46 38  91 6e 38  5a  91 64 38  91 46 38  64
// module  1: 83  91 82 39  41 45 45  4a  91 6e 38  91 78 39  6e  91 64 38  91 64 38  78  91 6e 38  91 64 38  6e  91 46 38  91 6e 38  5a  91 64 38  91 46 38  64
// module  1: 83  91 6e 38  41 45 45  4a  91 50 38  91 6e 38  6e  91 5a 38  91 5a 38  78  91 5a 38  91 5a 38  6e  91 46 38  91 64 38  5a  91 5a 38  91 3c 38  64
// module  1: 83  91 6e 38  41 45 45  4a  91 64 38  91 6e 38  6e  91 5a 38  91 5a 38  78  91 5a 38  91 5a 38  6e  91 3c 38  91 64 38  5a  91 5a 38  91 3c 38  64
// module  1: 83  91 6e 38  41 45 45  4a  91 64 38  91 6e 38  6e  91 50 38  91 5a 38  78  91 64 38  91 5a 38  6e  91 32 38  91 64 38  5a  91 50 38  91 3c 38  64
// module  1: 83  91        41 45 45  4a  91    38  91        6e  91    38  91    38  78  91    38  91    38  6e  91    38  91    38  5a  91    38  91    38  64
// module  1:     3.723 56  1.671 60      3.72  56  3.723 56      3.72  56  3.721 56      3.722 56  3.721 56      3.717 56  3.722 56      3.72  56  3.718 56

I'm still trying to figure out what the delimiter bytes mean but I'm about 70% confident in the voltage + percentage interpretations. Will need to change the car's voltage a bit and re-measure to see if the values still make sense.

are you able to send commands? Or you just logging?

With $22 you can read data identifier in UDS ECUs which we talk here on taycan model.

Do on Ecu 19 = Gateway: $224613
($22 Read DataIdentifier and 4613 = DEC for $1205 can identifier for EM_prcSoh)

If you are able to send raw data on bus then you can restart all UDS Ecus simply with:
$1101
will ack you 5101 back for sucessful reset.

 

[ciaranob]

Cell voltages of course entirely SoC dependent and will vary with SoC. Full range something like 3.2V to 4.2 V - also if cell map displayed when driving you'll see it change on acceleration or with region kicking in etc.

BTW, not sure if you're aware of this angle re how you the popular ABRP app can be used with an OBDII reader:
https://www.iternio.com/abrp-obd

The principle fields used by most of us to date are HV Soh, Battery Energy, Display Soc and the Cell maps.

Another measure of 'Battery Equivalent Cycles' would be useful but might not be possible to extract the correct charging history data via OBDII although you'd think it would - I managed to calculate to some degree from my car using my own logged data.

 

[electricsidecar]

here you go: ECU 8C

you want more? specify it and i will search for you. what you need additional is so called calcution formula. i can export all in one shot. lot of data is completly different in just formula output so thats main work - search needed data field and have also the right calc formula for.

ECU C6 Battery Charger

in current models the most data is stored in ECU 19 = gateway. here you go, the so called 'holy grail' = SoH

Is this gateway ECU 19 something you can send commands to to generate a human readable list of PIDs like this? Or are you using software that already has the PIDs coded in? I am able to send commands to my OBDII dongle so if there’s a way to scan this ECU 19 for data that could be a huge accelerant for my understanding!

 

[romed]

all UDS Ecu refering to manufacturers ODX which describes the data identifier as well as the formulas for the data identifier results in order to measurement values. as wrote: send me what you need and i will create an list for you with specifications.

 

[romed]

Is this gateway ECU 19 something you can send commands to to generate a human readable list of PIDs like this? Or are you using software that already has the PIDs coded in? I am able to send commands to my OBDII dongle so if there’s a way to scan this ECU 19 for data that could be a huge accelerant for my understanding!

i gave you an example how to query ECU data to get an result back. Then we go ahead with structured interperting

 

[ciaranob]

I'm quite curious to see what the voltage map looks like over time and how power spreads around the cells as you drive and charge the car. Once I get everything confidently decoded I've got some pretty fun ideas for visualizing the data nicely

Hey just wanted to mention, and I'm sure you're entirely aware, that the OBDII dongles can be used to access/change ride heights - there is a person/company that sells 'OBDAPP' sw that does exactly this and a couple of threads on it here e.g. https://www.taycanforum.com/forum/threads/lowering-links.12588/ - likely WAY beyond scope of what you want ElectricSideCar to offer, but interesting

Also in addition to David's great shares here, forum members @rs38 and @Dee amongst others have deep knowledge on all things OBDII related and might be good to ping for queries as to certain attribute functions etc.

Cheers, C.

 

[romed]

Is this gateway ECU 19 something you can send commands to to generate a human readable list of PIDs like this? Or are you using software that already has the PIDs coded in? I am able to send commands to my OBDII dongle so if there’s a way to scan this ECU 19 for data that could be a huge accelerant for my understanding!

do you already have PIWIS or ODIS? I think you should catch it, so you are way faster to sniff your needs out. my screens are from ODIS. there you can trace the hole CAN data raw as well as take a look in dedicated ECUs for measurement values and all other stuff.

with piwis iam struggling, dont got it run yet. have to force this. but i can readout with odis also actually.

 

[electricsidecar]

do you already have PIWIS or ODIS? I think you should catch it, so you are way faster to sniff your needs out. my screens are from ODIS. there you can trace the hole CAN data raw as well as take a look in dedicated ECUs for measurement values and all other stuff.

with piwis iam struggling, dont got it run yet. have to force this. but i can readout with odis also actually.

Not yet, but will look into it!

Latest update on the battery readouts, feeling decently confident in the decoding of the voltage and SoC % for each cell.

// module  1: 83  91 6e 38  41 45 45  4a  91 64 38  91 6e 38  6e  91 50 38  91 5a 38  78  91 64 38  91 5a 38  6e  91 32 38  91 64 38  5a  91 50 38  91 3c 38  64
//    values:     3.723 56  1.671 60      3.72  56  3.723 56      3.72  56  3.721 56      3.722 56  3.721 56      3.717 56  3.722 56      3.72  56  3.718 56

// module  1: 83  90 60 35  3e 44 44  49  90 56 35  44 48 35  00  90 4c 35  90 56 35  00  90 4c 35  90 4c 35  00  90 38 35  90 56 35  00  90 42 35  90 2e 35  00
//    values:     3.696 53  1.594 68      3.72  53  3.723 53      3.72  53  3.721 53      3.722 53  3.721 53      3.717 53  3.722 53      3.72  53  3.718 53

The mystery is still the bytes acting as padding around each cell. What I know so far:

- There are 7 of these values, one at the start of the cell block, one at the end, and one in between each cell pair.
- The values seem fairly stable within a narrow window of time, but do change over time.
- The values seem fairly different between different modules.
- The values I've seen so far include:

// 83 = 1000_0011 = 131
// 4a = 0100_1010 = 75
// 49 = 0100_1001 = 73
// 6e = 0110_1110 = 110
// 78 = 0111_1000 = 120
// 6e = 0110_1110 = 110
// 5a = 0101_1010 = 90
// 64 = 0110_0100 = 100
// 8c = 1000_1100 = 140

There's no discernible bit pattern here, so I'm inclined to think they're not bitmasks. I'm going to try streaming more values over a longer drive to see if there's a discernible change over time.

 

[romed]

where exactly you fetching this data from? ecu number?

Sponsored