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.

Hmm! I haven't gotten that to work... Nothing I have has the Model X assets files............ *wonders*

Edit: Nevermind... they're there... in the Model S folder... lol

So every car has a unique token every day (rather than all cars have the same new token every day)? How do the techs obtain the token to use on a specific car?

So now the car sends the token twice? Once to send it, then second time to confirm? Why not just get an ACK back the first time around (with either the token back at you or a hash of the token received)?

Whoever decided this needed to be a 2 transaction operation (send token, then ask for confirmation that the token got there) probably didn't want include the token in the confirmation request, meaning they needed some way to identify that the ACK was in fact for the current token, not some previous one. Just a guess...

Hey, I didn't design it. lol. Seems to be built around ensuring Tesla always has at least one of the tokens. And yes, every car has a unique token... every day. How the techs get them I assume is via their diagnostic software that connects to some authenticated Tesla service server.
 
Last edited:
OK... fruits of much CAN decoding labor..... GRAPHS! :D

p85d-power1.png


Power plot of a 0-60-ish launch. Notice how total power stays pretty steady after the ramp, and the rear motor starts to ramp down while the front ramps up. Going to need to see something like a 0-120 MPH graph or something to see how that pans out... somewhere around 333 kW total to the shafts, or about 446 HP.

p85d-torque1.png


Torque! Right around ~680 ftlbs of torque for this launch in total.

This is the same CAN log I used for one of my first graphs. I'll need to get some new data to graph at 100% charge or something on a warmer day soon. I didn't use full resolution for the graphs this time since it's so much data. I clipped it at about 25 Hz.

More to come. :D
 
Power plot of a 0-60-ish launch. Notice how total power stays pretty steady after the ramp, and the rear motor starts to ramp down while the front ramps up. Going to need to see something like a 0-120 MPH graph or something to see how that pans out... somewhere around 333 kW total to the shafts, or about 446 HP.

So...446 is less than 6.. ...nah, never mind. :)
 
So...446 is less than 6.. ...nah, never mind. :)

Yeah, well, I was reasonably certain that was already pretty well established. lol. I guess it's also worth noting that neither motor maxes out on it's nameplate power during a launch... not even close.

- - - Updated - - -

Just to note, these are just about all of the variables I've been able to decipher CAN messages for. I used Tesla's naming conventions from their CAN diagnostic tool. I added a few derived ones like _ftlb and the human readable gear selections...

Code:
MSG: BMS_energyBuffer
MSG: BMS_energyToChargeComplete
MSG: BMS_expectedEnergyRemaining
MSG: BMS_idealEnergyRemaining
MSG: BMS_kwhChargeTotal
MSG: BMS_kwhDischargeTotal
MSG: BMS_maxDischargePower
MSG: BMS_maxRegenPower
MSG: BMS_nominalEnergyRemaining
MSG: BMS_nominalFullPackEnergy
MSG: BMS_odometer
MSG: BMS_packCurrent
MSG: BMS_packPower
MSG: BMS_packVoltage
MSG: BMS_socMin
MSG: BMS_socUI
MSG: DCDC_inletTemperature
MSG: DCDC_inputPower
MSG: DCDC_outputCurrent
MSG: DCDC_outputPower
MSG: DCDC_outputVoltage
MSG: DIS_aps12V
MSG: DIS_dissipation
MSG: DIS_drivePowerMax
MSG: DIS_mechPower
MSG: DIS_statorCurrent
MSG: DIS_torqueEstimate
MSG: DIS_torqueEstimate_ftlb
MSG: DI_analogSpeed
MSG: DI_aps12V
MSG: DI_brakePedal
MSG: DI_brakePedalState
MSG: DI_dissipation
MSG: DI_drivePowerMax
MSG: DI_epbInterfaceReady
MSG: DI_epbParkRequest
MSG: DI_gear
MSG: DI_gearHuman
MSG: DI_gearRequest
MSG: DI_gearRequestHuman
MSG: DI_mechPower
MSG: DI_motorRPM
MSG: DI_pedalPos
MSG: DI_regenPowerMax
MSG: DI_statorCurrent
MSG: DI_torqueEstimate
MSG: DI_torqueEstimate_ftlb
MSG: DI_torqueInterfaceFailure
MSG: DI_vehicleSpeed
MSG: PMS_torqueMeasured
MSG: PMS_torqueMeasured_ftlb
MSG: PM_pedalPosA
MSG: PM_pedalPosB
MSG: PM_torqueMeasured
MSG: PM_torqueMeasured_ftlb

DI/PM are related to the rear drive unit. DIS/PMS are for the front drive unit.

Edit: No, BMS_energyBuffer is not what you think it is. 0 miles still means 0.
 
4.15/4.16V at 100%. Never looked at zero.
This is very good thing! Means they are not charging the cells to 100%!!! At 4.15, thats only about 93-95% of "full" since 4.2 is considered "Full". So Charging to 100% on the car really doesn't stress the cells anywhere near what I was thinking. And even then, my own experience with my packs mirrors the above. Almost no degradation even with the tons of miles I put on.
 
Yeah, well, I was reasonably certain that was already pretty well established. lol. I guess it's also worth noting that neither motor maxes out on it's nameplate power during a launch... not even close.

Reasonably established as well as already published by Tesla. However, if you could read that out form the CAN messages, that means Tesla knew all along and not realized some time down the line that they couldn't reach this number (there is no way a sane person would think they could squeeze out another 50% power from somewhere). So, when JB Straubel posted his Tesla All Wheel Drive (Dual Motor) Power and Torque Specifications | Tesla Motors he also could have simply come out and stated the max power but chose instead to continue to snow people with a "chemical reactions happen, electrons flow - it's complicated" argument rather than an honest "691 will never be seen unless you strap on a couple of spaceX boosters". If on the other hand JB did not see max power numbers by Sept 21 2015 (when he posted his update), that would make him one astronaut CTO - completely disconnected from earth, floating in orbit somewhere, a Donald Trump of EV's. :)
 
Last edited:
  • Like
Reactions: KyleDay
This is very good thing! Means they are not charging the cells to 100%!!! At 4.15, thats only about 93-95% of "full" since 4.2 is considered "Full". So Charging to 100% on the car really doesn't stress the cells anywhere near what I was thinking. And even then, my own experience with my packs mirrors the above. Almost no degradation even with the tons of miles I put on.

4.15/4.2=0.988?
 
Reasonably established as well as already published by Tesla. However, if you could read that out form the CAN messages, that means Tesla knew all along and not realized some time down the line that they couldn't reach this number (there is no way a sane person would think they could squeeze out another 50% power from somewhere). So, when JB Straubel posted his Tesla All Wheel Drive (Dual Motor) Power and Torque Specifications | Tesla Motors he also could have simply come out and stated the max power but chose instead to continue to snow people with a "chemical reactions happen, electrons flow - it's complicated" argument rather than an honest "691 will never be seen unless you strap on a couple of spaceX boosters". If on the other hand JB did not see max power numbers by Sept 21 2015 (when he posted his update), that would make him one astronaut CTO - completely disconnected from earth, floating in orbit somewhere, a Donald Trump of EV's. :)

You're definitely preaching to the choir with me.

In any case, let's keep the 691 HP discussion over in one of the 691 HP discussion threads to keep things clean over here.
 
OK... fruits of much CAN decoding labor..... GRAPHS!

Flipping cool.

Dumb question: During the ramp-up, I would (very naively) expect the power from batteries and mechanical motor power to be monotonically non-decreasing, especially since there's no transmission like there would be on most ICE cars. But the graphs have steps and jaggies all the way to the top at 1.5s. Why? Are these measurement artifacts or actual fluctuations in measured power? Thanks for a clue...
 
Flipping cool.

Dumb question: During the ramp-up, I would (very naively) expect the power from batteries and mechanical motor power to be monotonically non-decreasing, especially since there's no transmission like there would be on most ICE cars. But the graphs have steps and jaggies all the way to the top at 1.5s. Why? Are these measurement artifacts or actual fluctuations in measured power? Thanks for a clue...
My guess is T/C

- - - Updated - - -

Q. for WK57/Ingineer, any V2G stuff buried in the firmware?
 
No, you need the ratio between full and empty. Empty is somewhere between 3 and 3.2V IIRC - so 0.05V is indeed about 5% of the potential capacity.

FYI: Lithium Ion (and in fact most all) battery chemistry has a non-linear discharge curve. It's hard to accurately discern capacity from voltage, which is why Tesla (and most everyone else) uses coulomb counting to more accurately determine SoC.
 
So, I went out to log a launch-mode run... first time trying it, and I failed to get it to enable. Just kept complaining about both pedals pressed. Any trick to this that I'm not aware of? I followed the directions exactly and even checked the dev screen after a few tries to make sure my steering wheel was perfectly straight. lol.
 
Turn on "Max Power Mode". If it's not blue, then it wont work. (Battery does not need to be heated though, just max power on)

Hold brake with left foot HARD. Stab right foot all the way to the floor and let off fast. You should see the "Launch Mode" message. Then press and hold accel and then when ready, drop the brake.
 
Turn on "Max Power Mode". If it's not blue, then it wont work. (Battery does not need to be heated though, just max power on)

Hold brake with left foot HARD. Stab right foot all the way to the floor and let off fast. You should see the "Launch Mode" message. Then press and hold accel and then when ready, drop the brake.

Hmm... yeah, did all of that. I even tried parking, enabling max power, drive, brake, accel, etc... no dice. Just "both pedals pressed." I'll try it again tomorrow.

I did make a CAN log of the car self-parking with summon...