Tesla Parameters Thread

[Jack Bauer]

Setting up a thread to record parameter files used on Tesla powered vehicles. Please post as below.

Vehicle : BMW E31 1996 840ci
Drive unit : Large rear non performance
Board version : LDU V2
Battery : Volt/Ampera gen1
Nominal battery voltage : 360v

[Marco]

Hi Guys !

Attached my parameters for the Porsche 944.
Power is limited for the homologation in Romania but still very powerfull !

Vehicle : Porsche 944, 1985
Drive unit : Small Rear Drive Unit
Board version : SDU V4
Battery : Volt/Ampera gen1
Nominal battery voltage : 360v

[jon volk]

Figured Id share my current tune.
Vehicle : BMW E30 325i
Drive unit : Large rear non performance
Board version : LDU V2
Battery : 2x Volt/Ampera gen1
Nominal battery voltage : 360v

Coming from standalone efi tuning Im pretty nitpicky about drivability. The only hiccup In still trying to fix is some very low speed on/off throttle jerkiness. Think slowly creeping about a parking lot scenarios




*SEE UPDATED PARAMETERS IN POST FURTHER BELOW.

[johu]

0-60 in 3.5s. Another example that slip control really sucks
If only I had a vehicle to test this I could fix the low speed jerkiness. It just doesn't happen with a 40kW ACIM Also I'm using single channel encoder which works a bit different.

[jon volk]

0-60 in 3.5s. Another example that slip control really sucks
If only I had a vehicle to test this I could fix the low speed jerkiness. It just doesn't happen with a 40kW ACIM Also I'm using single channel encoder which works a bit different.

It’s pretty mild and if I back all the way out of the throttle and ease back on it goes away. It’s this perfect spot of trying to modulate speed below say 5 mph where it occurs. I’ve lived with it but I’ve been trying to tune it out.

[johu]

I added Jons config to the wiki since it seems very refined
https://openinverter.org/wiki/Configuration_Files

I also added Marcos SDU parameters

[jon volk]

I should also mention the launch in that video is foot on the brake while holding WOT, then just letting off the brake and letting the throtramp do the rest. Its fairly mild to protect stock BMW axles with 250k miles on them.

[P.S.Mangelsdorf]

Figured Id share my current tune.
Vehicle : BMW E30 325i
Drive unit : Large rear non performance
Board version : LDU V2
Battery : 2x Volt/Ampera gen1
Nominal battery voltage : 360v

Coming from standalone efi tuning Im pretty nitpicky about drivability. The only hiccup In still trying to fix is some very low speed on/off throttle jerkiness. Think slowly creeping about a parking lot scenarios

Jon, what are you seeing for range with that setup?

[jon volk]

Updated parameter set. Same car/setup, just improved performance and driveability.

Further details on the what/why behind the changes can be found here.



viewtopic.php?f=10&t=126&start=40

updated tesla LDU parameters.json

(1.06 KiB) Downloaded 245 times

[Roadstercycle]

Jon, I wanted to say thank you for the parameters insight and the explanations. I played with a few settings and now I can not believe the difference in performance. I'm sure myself and many others where wishing or hoping that someone along the way would do something like this as I think we are all pretty nervous about making changes since hearing all the nightmares about wiping inverter's. I have not done 0 to 60 times yet but It's insane now. Thanks again.

Here's my updated gear selector

[Roadstercycle]

I decided to take some more time and do some more playing with settings. I noticed that I was getting some jerkiness at low speed that I never had before when trying a some of Jon's parameters so I dove in deeper because now I knew what I had changed to make it start becoming jerky. I have a P90 performance motor with a 32 Kilowatt pack. 2 six packs of LG chem Chrysler Pacifica batteries. This could be the major difference as moving to 1800 boost and tweaking a couple other settings it smoothed right out. The top charge voltage I use is 387 volts. With the parameters I uploaded here it has been the best it has ever been. No jerkiness, no incline issues, no over current faults, can smoke the tires on demand. The car is just at 3200 lbs with driver. Next I will try to see about the idle mode setup to see if that even makes it more enjoyable to drive. Again a thank you to Jon for giving us some more info about setting up ours motors.

Please note that in these settings I am on single encoder and it is supposedly be on A/B Quad. Mode.

[jon volk]

Interesting as I’ve gotten worse jerkiness with similar parameters. I noticed you’re still on encoder mode 0 which I believe is single. I’m curious what happens if you change only that to quadrature/ab.

[Roadstercycle]

You know I never even realized that it was on single. For some reason I thought I never changed it from what is was when I got the board. Holy Moly if I change that now do you think my settings will go haywire?

[Roadstercycle]

Well I'm going to go try it now.

Did not feel any difference at all. Still really smooth.

[jon volk]

That looks like the only main setting that’s different than anything anyone else has tried

[Roadstercycle]

That looks like the only main setting that’s different than anything anyone else has tried

Maybe, but I really did not "try it" it was just on that setting the whole time. Like I said I did not feel any difference switching from single to A/B. Maybe Damien can shed some light on why it has to be on the A/B setting.

[jon volk]

The battery capability may certainly play into things.

With sending CAN parameters with my VCU, I have the ability to switch between two completely different setups with a button on the dash. I may try doing an A-B comparison with these. I also need to try some dead time parameter changes Arber mentioned in another thread.

It is nice getting multiple data points as well. Its tough to gauge things in a vacuum of one car.

[Roadstercycle]

The battery capability may certainly play into things.

With sending CAN parameters with my VCU, I have the ability to switch between two completely different setups with a button on the dash. I may try doing an A-B comparison with these. I also need to try some dead time parameter changes Arber mentioned in another thread.

It is nice getting multiple data points as well. Its tough to gauge things in a vacuum of one car.

I agree with the vacuum of one car. Mine is getting better thanks to you. On a side note, I did this last single mode and A/B test with a low battery. I'll charge it up and let you know if there is any difference.

[jon volk]

I put a little more slip and boost in it to get rid of the initial startup rumble so to speak. At 395v I was still getting some jerkiness so I put that value into my rpm based slip limit and it worked pretty well. The issue is with that boost I was able to hit overcurrent fault on a steep hill startup. I would say it was a step forward in most driving situations. I need to try Arbers deadtime increase suggestion to try and get rid of that desat fault.

[Roadstercycle]

So this morning I tried the A/B encoder 1 setup with a full battery. I still did not feel any difference in driving or stopping on my steep driveway and starting. No Jerkiness. So I would really like to know the difference in single encoder mode and A/B mode.
Now this afternoon I did a few drives and really slow in parking lots in the A/B encoder mode. It was really faint but I was noticing some very subtle jerkiness taking off really slow. I then returned to the single encoder mode to see if that changed anything. To my surprise the jerkiness went away and throttling was clean again. So now I am really anxious to hear from Damien what the difference is and will it hurt anything in single mode. I've been driving in single mode for weeks without knowing it so whether it is hurting anything is beyond me.

[ManuFromParis]

Hello all,

here is my parameter set, even though we're still working on several improvements.

Car is a 911 with Tesla SDU and 8 x i3 batteries.
Rear tires are MICHELIN Pilot Sport Ps2 255/40R17, so a lot of grip.

The major issue we had was some unwanted power cuts due to excessive inverter temperature when high torque was requested for too long.
We solved that with an improvement of the water cooling system (I'll describe that on another topic) but mostly with the tuning of the slip parameters which I did not understand well in the first place :
- slipstart was set to 40%, which was too high, it's now 15%
- fslipmin was set to 2 or 3, and now reduced to 1.37 while fslipmax remains at 7
- iacmax can now be set to 1000A

The behavior is now much smoother, from complete start to high speeds (> 150km/h), even though we'd like higher torque in very low speed.

I have questions though :
- while the throttle is under "slipstart" value, which fslip is used ? I assume it's the fslipmin value as a constant ? Anyone could confirm that ?
- above slipstart, what is the method to set the fslip value ? Is it depending on speed or only on throttle position ? (ratio from fslipmin-to-max)

Code: Select all

{
  "boost": 2200,
  "fweak": 225,
  "fconst": 400,
  "udcnom": 0,
  "fslipmin": 1.37,
  "fslipmax": 7,
  "fslipconstmax": 8,
  "polepairs": 2,
  "respolepairs": 1,
  "encmode": 1,
  "fmax": 600,
  "numimp": 36,
  "dirchrpm": 100,
  "dirmode": 1,
  "snsm": 12,
  "pwmfrq": 0,
  "pwmpol": 0,
  "deadtime": 63,
  "ocurlim": -1250,
  "il1gain": -1.5,
  "il2gain": -1.5,
  "udcgain": 7.5,
  "udcofs": 0,
  "udclim": 540,
  "snshs": 2,
  "bmslimhigh": 90,
  "bmslimlow": -100,
  "udcmin": 300,
  "udcmax": 520,
  "idcmax": 5000,
  "idcmin": -5000,
  "tmphsmax": 85,
  "tmpmmax": 300,
  "throtmax": 100,
  "throtmin": -100,
  "iacmax": 1000,
  "ifltrise": 10,
  "ifltfall": 3,
  "chargemode": 0,
  "chargecur": 0,
  "chargekp": 80,
  "chargeki": 10,
  "chargeflt": 8,
  "chargemax": 90,
  "potmin": 900,
  "potmax": 4095,
  "pot2min": 4095,
  "pot2max": 4095,
  "potmode": 0,
  "throtramp": 100,
  "throtramprpm": 20000,
  "ampmin": 10,
  "slipstart": 15,
  "brknompedal": -50,
  "regenramp": 60,
  "brknom": 30,
  "brkmax": -50,
  "brkcruise": -50,
  "brkrampstr": 100,
  "brkhistr": 1000,
  "brkhistp": 1000,
  "brkout": -50,
  "idlespeed": -100,
  "idlethrotlim": 50,
  "idlemode": 0,
  "speedkp": 0.21,
  "speedflt": 5,
  "cruisemode": 0,
  "udcsw": 250,
  "udcswbuck": 540,
  "tripmode": 2,
  "pwmfunc": 0,
  "pwmgain": 100,
  "pwmofs": 0,
  "canspeed": 0,
  "canperiod": 0,
  "fslipspnt": -1.37,
  "ampnom": 0
}

[StefSonic7EV]

Hi Manu,

Nice post, lots of details. What is your car's weight, probably around 1,5 tonnes, right?

I'm asking because I was trying to figure out parameters for a good few months now, first on the old engine and v4 Board, last two weeks I went for a new SDU (burned inverter in the first one) and v6 Board.
My car weights 750kg, Chevy Bolt battery 15kwh pack, semi slicks 245/40/17 and limited slip diff.

I got to the same conclusion on fslipmin, 1.37 is enough for a not jerky start. Increasing it makes no difference.
Totally different for fslipmax, with fweak 240 anything above 3,8 stalls the engine. It makes strange noise, does not rev anymore, but taking the foot off the accelerator and putting it back on, makes engine work. Also, problems with larger fslipmax are only on speeds below 60km/h.
I was rationalizing that with a light car the motor accelerates quickly already so does not need a large slip to speed up. Above 60km/h the drag is probably not smaller than in your Porsche so that's why large fslipmax is still ok.

iacmax makes no difference to the performance, for me. or throtle ramp parameter.

I did not play with the slipstart yet. But maybe I should since with the v6 Board I can't do a burnout anymore. Somehow it is slow off the line and really takes on around 20-30km/h.

Anyway, I'm a fan of v6 Board, accelerator pedal is much smoother and engine never shuts down. A bit slow start from 0km/h but I sense I need to play with the parameters more.

Interesting thing you said about inverter temperature, I had the same problem when racing on a hot day. Coolant was staying at 40 degrees and since inverter cuts power at 80 degrees, not much room to play.

How did you fix it?

My params:

Code: Select all

{
  "boost": 2200,
  "fweak": 240,
  "fconst": 800.09,
  "udcnom": 360,
  "fslipmin": 1.37,
  "fslipmax": 3.7,
  "fslipconstmax": 6.09,
  "fmax": 600,
  "dirchrpm": 100,
  "dirmode": 1,
  "ocurlim": 1200,
  "bmslimhigh": 90,
  "bmslimlow": -100,
  "udcmin": 300,
  "udcmax": 400,
  "idcmax": 500,
  "idcmin": -500,
  "tmphsmax": 85,
  "tmpmmax": 300,
  "throtmax": 100,
  "throtmin": -100,
  "iacmax": 800,
  "chargemode": 0,
  "chargecur": 0,
  "chargekp": 80,
  "chargeki": 10,
  "chargeflt": 8,
  "chargemax": 90,
  "potmin": 300,
  "potmax": 4095,
  "pot2min": 4095,
  "pot2max": 4095,
  "potmode": 0,
  "throtramp": 10,
  "throtramprpm": 20000,
  "ampmin": 5,
  "slipstart": 40,
  "brknompedal": -25,
  "regenramp": 1,
  "brknom": 30,
  "brkmax": -30,
  "brkcruise": -30,
  "brkrampstr": 50,
  "brkout": -100,
  "idlespeed": -100,
  "idlethrotlim": 50,
  "idlemode": 0,
  "speedkp": 1,
  "speedflt": 5,
  "cruisemode": 0,
  "udcsw": 320,
  "udcswbuck": 540,
  "tripmode": 2,
  "pwmfunc": 0,
  "pwmgain": 100,
  "pwmofs": 0,
  "canspeed": 0,
  "canperiod": 0,
  "fslipspnt": -1.37,
  "ampnom": 0
}

Hello all,

here is my parameter set, even though we're still working on several improvements.

Car is a 911 with Tesla SDU and 8 x i3 batteries.
Rear tires are MICHELIN Pilot Sport Ps2 255/40R17, so a lot of grip.

The major issue we had was some unwanted power cuts due to excessive inverter temperature when high torque was requested for too long.
We solved that with an improvement of the water cooling system (I'll describe that on another topic) but mostly with the tuning of the slip parameters which I did not understand well in the first place :
- slipstart was set to 40%, which was too high, it's now 15%
- fslipmin was set to 2 or 3, and now reduced to 1.37 while fslipmax remains at 7
- iacmax can now be set to 1000A

The behavior is now much smoother, from complete start to high speeds (> 150km/h), even though we'd like higher torque in very low speed.

I have questions though :
- while the throttle is under "slipstart" value, which fslip is used ? I assume it's the fslipmin value as a constant ? Anyone could confirm that ?
- above slipstart, what is the method to set the fslip value ? Is it depending on speed or only on throttle position ? (ratio from fslipmin-to-max)

Code: Select all

{
  "boost": 2200,
  "fweak": 225,
  "fconst": 400,
  "udcnom": 0,
  "fslipmin": 1.37,
  "fslipmax": 7,
  "fslipconstmax": 8,
  "polepairs": 2,
  "respolepairs": 1,
  "encmode": 1,
  "fmax": 600,
  "numimp": 36,
  "dirchrpm": 100,
  "dirmode": 1,
  "snsm": 12,
  "pwmfrq": 0,
  "pwmpol": 0,
  "deadtime": 63,
  "ocurlim": -1250,
  "il1gain": -1.5,
  "il2gain": -1.5,
  "udcgain": 7.5,
  "udcofs": 0,
  "udclim": 540,
  "snshs": 2,
  "bmslimhigh": 90,
  "bmslimlow": -100,
  "udcmin": 300,
  "udcmax": 520,
  "idcmax": 5000,
  "idcmin": -5000,
  "tmphsmax": 85,
  "tmpmmax": 300,
  "throtmax": 100,
  "throtmin": -100,
  "iacmax": 1000,
  "ifltrise": 10,
  "ifltfall": 3,
  "chargemode": 0,
  "chargecur": 0,
  "chargekp": 80,
  "chargeki": 10,
  "chargeflt": 8,
  "chargemax": 90,
  "potmin": 900,
  "potmax": 4095,
  "pot2min": 4095,
  "pot2max": 4095,
  "potmode": 0,
  "throtramp": 100,
  "throtramprpm": 20000,
  "ampmin": 10,
  "slipstart": 15,
  "brknompedal": -50,
  "regenramp": 60,
  "brknom": 30,
  "brkmax": -50,
  "brkcruise": -50,
  "brkrampstr": 100,
  "brkhistr": 1000,
  "brkhistp": 1000,
  "brkout": -50,
  "idlespeed": -100,
  "idlethrotlim": 50,
  "idlemode": 0,
  "speedkp": 0.21,
  "speedflt": 5,
  "cruisemode": 0,
  "udcsw": 250,
  "udcswbuck": 540,
  "tripmode": 2,
  "pwmfunc": 0,
  "pwmgain": 100,
  "pwmofs": 0,
  "canspeed": 0,
  "canperiod": 0,
  "fslipspnt": -1.37,
  "ampnom": 0
}

[ManuFromParis]

Hi Manu,

Nice post, lots of details. What is your car's weight, probably around 1,5 tonnes, right?

Nope, fortunately not !
It's now around 1150kg because
- it has been striped to bare metal before the tesla fitting, and soundproof material is very heavy on these cars
- total weight off from a 911 is quite a lot : around 200 for engine, 80 for gearbox with clutch, pump, filters and ignition parts goes quickly above 10kg and the front fuel tank is around 15kg alone.
I need to have it weighted again, but I'm confident, regarding the weight of each part added for the conversion

I'm asking because I was trying to figure out parameters for a good few months now, first on the old engine and v4 Board, last two weeks I went for a new SDU (burned inverter in the first one) and v6 Board.
My car weights 750kg, Chevy Bolt battery 15kwh pack, semi slicks 245/40/17 and limited slip diff.

I got to the same conclusion on fslipmin, 1.37 is enough for a not jerky start. Increasing it makes no difference.
Totally different for fslipmax, with fweak 240 anything above 3,8 stalls the engine. It makes strange noise, does not rev anymore, but taking the foot off the accelerator and putting it back on, makes engine work. Also, problems with larger fslipmax are only on speeds below 60km/h.
I was rationalizing that with a light car the motor accelerates quickly already so does not need a large slip to speed up. Above 60km/h the drag is probably not smaller than in your Porsche so that's why large fslipmax is still ok.

iacmax makes no difference to the performance, for me. or throtle ramp parameter.

It made a difference on the 911 because what we wanted to get rid of was the Inverter Temp Cut off, and these were generated by Inverter temp rises, on long and intense accelerations.

I did not play with the slipstart yet. But maybe I should since with the v6 Board I can't do a burnout anymore. Somehow it is slow off the line and really takes on around 20-30km/h.

Anyway, I'm a fan of v6 Board, accelerator pedal is much smoother and engine never shuts down. A bit slow start from 0km/h but I sense I need to play with the parameters more.

Interesting thing you said about inverter temperature, I had the same problem when racing on a hot day. Coolant was staying at 40 degrees and since inverter cuts power at 80 degrees, not much room to play.

How did you fix it?

Basically with the delete of oil/water heat exchanger.
We added a small oil cooler with a constant fan and kept the OEM oil pump.

We added a big water cooler just under the rear spoiler, which let a lot of air in, as soon as you drive above 40km/h.
The water bleeding was and issue because this cooler is higher than anything else, but we managed to trap the air with a side water tank, fitted in parallel with the main water circuit, and a long water hose on the filling fitting.

We suspect the water pump to be too powerfull and too quick though (around 50L/min) because we still have some cut-off on warm days, when going up a hill for instance (lot of torque involved) or with intense start from stop (0-60pmh)

[Gregski]

Captain Dump Question here, I ask because no one else will on behalf of us Newbs, ha ha. Can someone please tell me what the "i" in idclim, iaclim, and idcspnt stands for? I understand that the conventional symbol for current is I, which as per the Wiki Wiki originates from the French phrase intensité du courant, (current intensity). However it would be silly if those were to be read idclim aka Current Direct Current Limit, iaclim aka Current Alternating Current Limit, and idcspnt aka Current Direct Current Setpoint

Does it mean Interrupted Direct Current?

[Isaac96]

Pretty sure it is Current Direct Current Limit. Maybe more along the lines of "Current (Direct Current) Limit." But it does help to make the difference between AC and DC very clear.
How would you write it without redundancy? I can't think of anything obvious.