Welcome to Tesla Motors Club
Discuss Tesla's Model S, Model 3, Model X, Model Y, Cybertruck, Roadster and More.
Register

Let the hacking begin... (Model S parts on the bench)

This site may earn commission on affiliate links.
How often does the VIN appear on CAN3 ... perhaps once a minute?

In my TDC to OBD adapter for the ELM-type OBD dongle that TM-Spy uses,
I take ground and power from TDC pins 9 and 10, to OBD pins 4-5 and 16.
The CAN3 H, L come from TDC 1 and 6, and connect to OBD pins 6 and 14.

The wired female 16-wire molded OBD connectors that I ordered should
be here next week, and then I should be able to try TM-Spy Beta on my car,
and log multiple CAN buses simultaneously.

For logging systems like mine (that worked with the LEAF), I will probably connect:
CAN6 to the LEAF's EV-CAN pins, and
CAN2 to the LEAF's AV-CAN pins.

Then, connect CAN4 to OBD pins 1 and 9 ... unless others have better suggestions.

Do we know for sure which CAN bus is on pins 1 and 9 of the Tesla OBD connector?
Thanks, Gary
 
i didn't have a chance to calculate the voltages from the previous data yet, but from your parse it looks like your pack is in great shape.

View attachment 111520
I am fully charged here. SOC says 69%
Battery temperatures were around 7°C

I am still trying to figure out what my degradation % is. Any idea of how to do that using these numbers?
The problem is that I cannot charge to 100%, so I cannot see the max charge.
 
Only my calculations ... :wink:
58 = 56.6 + 1.4
96.6% = 58/60

81.6 = 77.6 + 4
96.6% = 81.6/85

and next story...
so, please forget to data from CANbus and calculated with me:
assuming for same cells...

7104 / 5880 / 5040 = 85 / 70 / 60
so ratio ca. 84
TS60 5040 = 84 * 60
TS70 5880 = 84 * 70
TS85 7140 = 84 * 85 -> 7104/7140 = -0.5% (only -402W from 85000W)
TS90 different chemistry ... ???
I calculate it wrong? :wink:
thanks
 
I am fully charged here. SOC says 69%
Battery temperatures were around 7°C

I am still trying to figure out what my degradation % is. Any idea of how to do that using these numbers?
The problem is that I cannot charge to 100%, so I cannot see the max charge.

Only thing I can think of is to take the nominal full energy figure (56.6) and see how it compares to a fresh pack.

Now my 60 is 18 months old with 12.5k miles on the clock and shows 58.1. (So newer but not fresh)

BUT Mine is a "B" pack, and I assume yours is an "A". Also the energy buffer figure is different :confused:

If you can find another 40 owner maybe you could make a fairer comparison?
 
For a while now, we've been making DIAG connectors for OVMS on the Tesla Roadster. I notice that some guys here are using those in Model S now (even though only 1 CAN bus is wired).

I'm considering getting some fully wired connectors made for my own use. The same as the existing OVMS cable (https://www.fasttech.com/products/0/10000001/1000400-ovms-data-cable-for-tesla-roadster-1-x-2-x), with the same DIAG plug, but with all 12 pins cabled. Perhaps 1m (3feet) of open ended cable on the end. Something easy to solder on to a DB9 (or whatever is required).

1000400-1.jpg


Would that be something useful to the community here? If so, I can easily get them to make up a batch. Is 1m cable enough? Cut and pre-tinned for soldering useful? Any other suggestions?
 
From earlier post (different logging session though):

Here are some of my numbers when fully charged:
Nominal Energy Pack Full: 56.6 kWh
Nominal Energy Remaining: 39.2 kWh
Expected Energy Remaining: 40.8 kWh
Ideal Energy Remaining: 39.2 kWh
Energy (Bricking?) Buffer: 1.4 kWh
Pack Voltage: 336.50
Battery Odometer: 34172
SoC UI: 69%

69% of nominal 56.6 + 1.4 buffer = 40.02 kWH vs expected remaining of 40.8

Also look at nominal remaining vs expected remaining: 39.2 vs 40.8

edit

Looks like very little degradation by either of these comparisons.

336.5V was under charging, so use the other reading.
326 V / 84 = 3.881 volts per cell at 69%, now that agrees with the brick voltages you measured.

add this to Simon's data

69% ~ 3.881 V
79% ~ 4.09 V
89% ~ 4.179 V
 
Last edited:
How often does the VIN appear on CAN3 ... perhaps once a minute?

In my TDC to OBD adapter for the ELM-type OBD dongle that TM-Spy uses,
I take ground and power from TDC pins 9 and 10, to OBD pins 4-5 and 16.
The CAN3 H, L come from TDC 1 and 6, and connect to OBD pins 6 and 14.

The wired female 16-wire molded OBD connectors that I ordered should
be here next week, and then I should be able to try TM-Spy Beta on my car,
and log multiple CAN buses simultaneously.

For logging systems like mine (that worked with the LEAF), I will probably connect:
CAN6 to the LEAF's EV-CAN pins, and
CAN2 to the LEAF's AV-CAN pins.

Then, connect CAN4 to OBD pins 1 and 9 ... unless others have better suggestions.

Do we know for sure which CAN bus is on pins 1 and 9 of the Tesla OBD connector?
Thanks, Gary

Gary,

I realize that you are probably trying to reuse existing hardware but the size of an OBD-II connector compared to the TDC is enormous, think it might be more worthwhile to interface directly with that, even if requires some soldering, it would be pretty easy to desolder the male OBD on the Leaf Spy and in the same through holes solder wires leading to a male TDC, no funky adaptors or long cabling required, you might even be able to tuck it into the back of the cubby. Tesla's CAN bus structure is significantly different than the leaf with much more interconnection of the various systems through gateways (unlike the leafs EV bus and AV bus). The real high value targets are CAN3 Powertrain (battery, drive unit and chargers) CAN6 Chassis (TC/ESC, body control module, ABS, EPB, TPMS) and finally CAN2 Body (lights, locks, windows, sunroof, radio etc). CAN4 is a fault tolerant module for the seat modules, PTC heater, and RCCM not much interesting going on there for most peoples purposes.

As for what pins are brought out to the OBDII port. To make a long story short the only thing available that we would care about is 12V and GND plus the CAN6 chassis bus, not of interest for our particular application.

More details:
1 = CAN6+
P4 = GND
P5 = GND
P6 = CAN1+
P7 = K/Ser line
P9 = CAN6-
P14 = CAN1-
P16 = Battery Voltage (nom 13V)

CAN1 is a special bus that connects directly to the center console stack and nothing else, what it's used for is anyones best guess at the moment (diagnostics or firmware loading?). The K/Ser line connects directly to the RCCM (remote climate control module) and is (I'm guessing) some kind of LIN interface, the TDC under the center console taps off this line and connects to P3 of the diag connector.

Also the VIN is on ID 0x508, not sure what the rate is though.


- - - Updated - - -

For a while now, we've been making DIAG connectors for OVMS on the Tesla Roadster. I notice that some guys here are using those in Model S now (even though only 1 CAN bus is wired).

I'm considering getting some fully wired connectors made for my own use. The same as the existing OVMS cable (https://www.fasttech.com/products/0/10000001/1000400-ovms-data-cable-for-tesla-roadster-1-x-2-x), with the same DIAG plug, but with all 12 pins cabled. Perhaps 1m (3feet) of open ended cable on the end. Something easy to solder on to a DB9 (or whatever is required).

Would that be something useful to the community here? If so, I can easily get them to make up a batch. Is 1m cable enough? Cut and pre-tinned for soldering useful? Any other suggestions?

Mark,

I don't know if it would be a better solution (it has worked for my needs reverse engineering) is a TDC to ethernet cable, with that I can go into a standard Ethernet to DE9 adaptor and then connect the DE9 cable to things like the CANtact or Kvaser Leaf Light (the reason I built it), it is a little bulkier but it enables the swapping of different buses just by changing out the Eth-TDC cable and keeping the same interface on the DE9 end, meaning I can have any bus without re-wiring on my logger.

Probably too bulky for stashing under the console but food for thought, I'll try and get some pictures uploaded later.
 
Last edited:
69% of nominal 56.6 + 1.4 buffer = 40.02 kWH vs expected remaining of 40.8

Also look at nominal remaining vs expected remaining: 39.2 vs 40.8

336.50 V / 84 = 4.006 volts per cell at 69%, how does that compare to the cell soc chart?

336.5V was not an accurate number as it was still trickle charging because of HVAC. The total voltage was 326 when not charging.
 
This was from a log while driving. I'll have a look more into it.

Yes, it seems to be a 5 minute heartbeat message that is sent any time the car's computer systems are operating. I've seen it on driving captures (but only captures that are minutes long, otherwise you're likely to never see it). I can't really be sure as to *why* it sends its own VIN number every 5 minutes. Who is listening for that message? The VIN never changes. Maybe some components check the expected VIN number and fault if it does not match - sort of like how radios and transmissions have been tied to the original vehicle in many cars. But, I haven't seen a device that seems to care. So far I've heard of people using drive trains and chargers outside the car and they seem to work without any VIN number communication. I wonder if they'd fault if they got the VIN message and it didn't match their expected VIN? Anyone brave enough to try this?
 
markwj,
We could use a very short TDC to female OBD adapter cable, with CAN3 wired
to OBD pins 6 and 14, and power and ground, of course.

We have AVR-CAN based loggers, with DB9 input that might use your TDC to DB9
cable, depending upon the pins used. What is your DB9 pinout, please?

I have a few manufacturers across the border (I am in Hong Kong) that I have worked with in the past for custom OBDII cables. Simple to get this done, but the TDC pins and plugs (Tyco) are only available in China in 10,000+ quantities, so need to provide those to them directly. I think the way to do it would be to get the manufacturer to provide their standard OBDII socket and cable (something they are very familiar with), and then customise the TDC end with the four pins required. MOQ is usually around 50 pieces.

We used the pinout that was popular at the time on a couple of different USB-CAN loggers:

DB9-F Signal
3 Chassis/Power GND
7 CAN-H
2 CAN-L
9 +12V Vehicle Power

20120811-teslacable.png


Mark,

I don't know if it would be a better solution (it has worked for my needs reverse engineering) is a TDC to ethernet cable, with that I can go into a standard Ethernet to DE9 adaptor and then connect the DE9 cable to things like the CANtact or Kvaser Leaf Light (the reason I built it), it is a little bulkier but it enables the swapping of different buses just by changing out the Eth-TDC cable and keeping the same interface on the DE9 end, meaning I can have any bus without re-wiring on my logger.

Probably too bulky for stashing under the console but food for thought, I'll try and get some pictures uploaded later.

If the non-TDC end is open, anything can be put there. The wires should be compatible with ethernet connectors, and it is a couple of minutes job to crimp those.

I know what Tesla used to do (on the roadster) is bring out the connector onto three separate DB9 connectors (one for each bus). That is great for a single-CAN-bus logger as you can just plug into the bus you want, but not so good for those loggers capable of logging multiple buses at the same time.
 
Last edited: