Tesla Large Drive Unit Support Thread

[pm_dawn]

Hi !

I'm struggling to get my p85 motor running since more than a year back.
Lately I have regained some momentum and tried to wire the motor up again.
I have made a small control panel with swithces for ignition, start, forward and reverse.
Throttle is also connected but no resolver cable yet.

The Board is a V2 early assembled boards that I got from Damien, with a esp Wifi board and a bare rs232 board.
I have used the RaspberryPi Zero in my first tries. So I also have that on hand.
But the interface of the espwifi looks much newer and has more functions I guess.

So I tried to get the esp-wifi board to work.
Board comes up but I cannot get it to communicate to the controller.

I tried to connect the rx/tx from CONN8 on the controller board.
And feeding vcc/gnd from Jtag pin2 and pin4.
The esp starts and I can login in to the wifi and brows the page, but there seems to be no
communicate with the board over the serial.

I was under the impression that CONN8 on the board was TTL before the RS232 chip.
But I get no com.
I have tried switching the rx/tx lines also no difference.

Does anyone have an idea?

Best REgards
/Per Eklund

[Sutho]

Got my Large Drive Unit running late last week, but thought it best that I update the firmware on the board before going any further....and now it won't run. It also has different parameters.

I am confident that I have loaded the wrong firmware (new FOC version) for the Tesla board, but would appreciate clarification of what version I should be running? Have been searching, but can't be confident that I will load the corrent non-FOC version.

A few quick points about my integration:

1. I believe this has been mentioned a few times, but the encoder wiring is reversed on these boards in relation to the Factory Tesla wiring. I am using a Tesla loom (and chassis) and needed to swap the A and B to get the motor to run.

2. The Main contactor is not being activated by Pin 6 on my board. The Precharge clicks in and will drop-out if HV is not there and will stay in (until Start is pushed) if HV is there. I needed to hard switch the Main contactor after Precharge.

------------------------------
More information
------------------------------

From what I can see, I should have been loading the STM32_sine version of firmware in lieu of the STM32_FOC version.

I have now loaded the last STM32_sine.bin version that I can find, but the Parameters are not displaying on the Web Interface. I have loaded a suitable .JSON file, but that doesn't seem to help.

Are there other files that I need to load/reload to get the inverter back to "normal"?

Thanks in advance.

[Jack Bauer]

So in relation to point 1 yes this has been address before on thread and elsewhere. It seems Tesla like to change the encoder wiring between years. I am running a 2016 European sourced drive unit with the original harness and works out of the box.

Regards main contactor activation this has also been addressed but most likely culprit is UDCSW parameter set too high.

Suggest reading up about firmware updates but correct in that you do not use the foc version on Tesla boards. If the parameters are not being displayed it means the mcu is not communicating and most likely not programmed correctly.

[Roadstercycle]

OK, I'm back at it. I'm finding abrupt starting when giving throttle. Once the wheels start going then I can feather the throttle way below the area that I have to push it to to get the initial start. I've tried boost in small increments but a bit afraid of damaging something. I just don't get a smooth acceleration. It's like going from off to on then I'm able to control throttle inputs smoothly. I have it off the ground as to not break anything. I plotted it and it's coming on at about 2800 of the pedal which has a range of 956 to 4096. I've tried regen variables too.
I really hate to be a PIA here. Fortunately and unfortunately we have a Model 3 so I know how smooth it can be. I am running a 360 volt nominal pack x 2. Here's my latest Para. file.

params.json

Para 12-02-19

(1.29 KiB) Downloaded 521 times

[tom91]

Pedal ramp is way high. We run a max of 3-4 even in our drift car.

Also we have our cars running a boost of 2000 on quite stiff 96s packs.

So give a higher boost and lower pedal ramp a try.

[Sutho]

Suggest reading up about firmware updates but correct in that you do not use the foc version on Tesla boards. If the parameters are not being displayed it means the mcu is not communicating and most likely not programmed correctly.

Thanks for the response. I have tried reverting back to the latest STM32_sine.bin that I can find which appears to be v4.64.R.

From the information I could find, loading the firmware from the Web Interface is performed by using the Update function and loading the .bin file. Is that correct? Is there a suggested version/date that I should be loading?

[Jack Bauer]

The wiki is a good place to start : https://openinverter.org/wiki

Here is a basic check. On the board there are two leds. One should be on constantly and one should blink rapidly when the mcu is running. If this is not the case then the firmware has not loaded correctly and should be power cycled and update run again.

As a general note all boards bought from the evbmw webshop are loaded with the latest firmware at time of shipping.

[Sutho]

Maybe someone has seen this symptom previously?

Having inadvertently loaded the wrong firmware (FOC) and then attempting to revert back to the correct version (Sine), I can no longer see the Parameters table in the Web Interface. I did transfer some files to the inverter when trying to work out why the FOC version wasn't working, so maybe I have filled up the Flashdrive memory?

A brief video of what I am seeing is here: https://youtu.be/LjtKgMCRQCI

We are about to remove the drive unit from the test chassis so we can see the LEDs, but please let me know if you have any ideas on how to correct this. Thanks!

[jon volk]

I dont think I ever posted this on the forum, but if you need a harness, the center harness for an MS3Pro EFI box uses the same 23pin connector as the LDU. This also comes with all positions populated with 8 foot leads of TXL wire that can be re-arranged and youll have plenty of leftover for other connections.

https://www.diyautotune.com/product/ms3 ... r-harness/

[Roadstercycle]

I dont think I ever posted this on the forum, but if you need a harness, the center harness for an MS3Pro EFI box uses the same 23pin connector as the LDU. This also comes with all positions populated with 8 foot leads of TXL wire that can be re-arranged and youll have plenty of leftover for other connections.

https://www.diyautotune.com/product/ms3 ... r-harness/

Jon, This is a great find. I actually used these before when building many Micro squirt setups but never realized they are the same . Great quality too!

[Roadstercycle]

Guy's wondering about push buttons. I see we can make the drive controller work with momentary push button forward and reverse but wondering how we get neutral? Or is it un-needed? I would like to have a push button forward, reverse, and neutral or is neutral unnecessary???

[Roadstercycle]

Pedal ramp is way high. We run a max of 3-4 even in our drift car.

Also we have our cars running a boost of 2000 on quite stiff 96s packs.

So give a higher boost and lower pedal ramp a try.

Still having an issue with a smooth acceleration even after lowering the pedalramp and increasing boost.

Also

Guy's wondering about push buttons. I see we can make the drive controller work with momentary push button forward and reverse but wondering how we get neutral? Or is it un-needed? I would like to have a push button forward, reverse, and neutral or is neutral unnecessary???

[jon volk]

Are you on the latest firmware? Try lowering the fslipmin to 1 or .9. I found higher values would cause jerkiness on my setup.

I also found using brake off idle mode made things much smoother for startup situations. This way it starts the inverter as soon as your foot comes off the brake pedal. For me, this was much smoother than having it start just with the throttle pedal input. Try uploading the .json of my settings from the Tesla parameters thread to see how that works. You just need to have the brake pedeal switch hooked up for it to work properly.

Every setup will be slightly different based on vehicle weight and battery capability, but those settings work for me. Ive got around 2,000 miles on my swap so far.

[Roadstercycle]

Are you on the latest firmware? Try lowering the fslipmin to 1 or .9. I found higher values would cause jerkiness on my setup.

I also found using brake off idle mode made things much smoother for startup situations. This way it starts the inverter as soon as your foot comes off the brake pedal. For me, this was much smoother than having it start just with the throttle pedal input. Try uploading the .json of my settings from the Tesla parameters thread to see how that works. You just need to have the brake pedeal switch hooked up for it to work properly.

Every setup will be slightly different based on vehicle weight and battery capability, but those settings work for me. Ive got around 2,000 miles on my swap so far.

Jon, Thank you, I believe I am on the latest firmware unless it's been in the last few days. I'll try your settings again. I tried them the other day and it did not seem to help but I will try again. On the rack as I push the pedal down about about a 1/8 pedal travel it starts like and on off switch but long as I don't let the wheels stop I can feather the throttle perfectly and smoothly. On the ground it just takes off again like an on and off switch. I'm afraid of braking something so I have it back on the rack.

[Roadstercycle]

Are you on the latest firmware? Try lowering the fslipmin to 1 or .9. I found higher values would cause jerkiness on my setup.

I also found using brake off idle mode made things much smoother for startup situations. This way it starts the inverter as soon as your foot comes off the brake pedal. For me, this was much smoother than having it start just with the throttle pedal input. Try uploading the .json of my settings from the Tesla parameters thread to see how that works. You just need to have the brake pedeal switch hooked up for it to work properly.

Every setup will be slightly different based on vehicle weight and battery capability, but those settings work for me. Ive got around 2,000 miles on my swap so far.

Hi Jon, tried your setup and it seemed smooth on the rack but had a little gear clunking, but I believe that was because the tires where unloaded. Put it on the ground to try it and it still jump starts or jerks. I'm thinking maybe the P90 motor may be a different animal to setup unless yours is a sport motor too. Most of my hair is on the floor now but I'm still at it. Can you tell me why your "udcnom": is at 390 when I believe all others are at 0? I've read the parameters sheet many times and I'm missing something. Again Thank you for the help. If I was a EE I may be better on this. I've done lots of electronics and I'm actually kind of embarrassed that I'm having so much trouble with this.

[Jack Bauer]

This tuning guide written by Arber333 over on Diyev a few years back may be some help :

Now comes the tuning part.
First you want to find a flat surface - a parking lot etc. so you can drive and stop without checking traffic. Change only one parameter at a time and save settings that work!

1. set fslipmin so that you feel car taking off smoothly and try to change it by +/-0,1Hz and check differences in starting. Save when satisfied.

2. lower boost value in 100pts untill motor jitters at start. Then return it to last good value.

3. try lowering ampmin in 0,1 increments and observe throttle travel. When throttle is not just smooth but becomes sluggish return some previous increments untill throttle reaction is acceptable.

4. change fweak value in +/-10Hz increments from starting point and observe torque in starting. This value is very dependent on battery voltage and is very subjective.

Now you find a hill or ramp and set car on it. You want to hold car in position on slope just using throttle pedal. If there parameters are not good motor will jump or will feel sluggish

1. add boost if motor is oscillating if it is smooth reduce it in 100pt increments untill you get oscillation. Then return to last good value
2. reduce/increase ampmin in 0,25 increments untill you get oscilation in motor and return last good value

Now set the car into a hill to set fslipmax. Warning full throttle will be used. Be sure there is no other traffic!
Set fslipmax to chosen value (guess it at 2xfslipmin if you have no other way) and try to take off with full throttle.
If car feels sluggish with full throttle you have to add more slip.
If motor starts to jitter there is too much slip. Try to reduce it in 0.1Hz increments.
When you feel satisfied with settings save them and go on setting regen and braking effect.

[jon volk]

Hi Jon, tried your setup and it seemed smooth on the rack but had a little gear clunking, but I believe that was because the tires where unloaded. Put it on the ground to try it and it still jump starts or jerks. I'm thinking maybe the P90 motor may be a different animal to setup unless yours is a sport motor too. Most of my hair is on the floor now but I'm still at it. Can you tell me why your "udcnom": is at 390 when I believe all others are at 0? I've read the parameters sheet many times and I'm missing something. Again Thank you for the help. If I was a EE I may be better on this. I've done lots of electronics and I'm actually kind of embarrassed that I'm having so much trouble with this.

What battery are you running? That initial "jump" feeling is primarily driven by the boost value. You can try reducing that. This stuff is all really quite subjective as far as feel. I can go as low as 1170 or so before getting a jittery motor startup with the two Chevy Volt packs.
*EDIT* boost had a tangible affect on the "jump" feeling in conjunction with the rest of the parameter settings Im running. It is not the only parameter in play for this.

udcnom makes boost/slip variable in relation to voltage. The spreadsheet I posted in an earlier reply demonstrates how its all linked. Its a relatively new addition so not everyone is using it yet.


Damien, I dont think Ive seen that tuning guide before. I added it to the Wiki for future reference.

[arber333]

Yep, I tuned my Mazda that way and it worked for my friends with different motors without much data.

It did not remove the starting jolt though. I remedied it by raising fweak point when johannes implemented motor voltage adaptation. My motor works with high minslip value of 3hz and 7.2hz of maxslip.

I must say boost wasn't any good until 7000 value for me and I use 400v system. Motor felt like rubber spring. But 7000 boost required longer dead time on igbts.

Everything is subjective on testing...

[jon volk]

Its interesting how different motors react. I only have experience with the LDU on this board. I played in the garage just rocking back and forth at lower boost and it was very tame on startup.

With the comparison made earlier to driveability of a Model3, I dont have any expectation that this setup will deliver OE level of control in its current state. If it can evolve to vary parameters in real time against multiple tables like an EFI standalone, I think we'll get close to that level, eventually. Just having udcnom and fslipconst have been great improvements thus far.

For the boards intended purpose of turning a drive unit on a budget, it does it very well.

[Roadstercycle]

Thank you, For all the help. I will do more testing today and see if I can get an easier smooth start. Unfortunately I won't be able to get into a parking lot until after Thursday as that's the day I go to the smog referee to hopefully get a pass on doing smog because of the switching to electric. Then I can get plates and insurance. California really is a pain in the butt for custom vehicles and smog.

Again thank you all.

Side note: Running two 6 packs of Chrysler Pacifica batteries, 387 volts top charge at the moment.

[jon volk]

Give this set of parameters a shot. I was playing around with my car today to see how smooth I could get the startup. I also found that raising the slipstart seemed to get rid of some very low speed jerkiness I had in select parking lot scenarios.

[Roadstercycle]

Give this set of parameters a shot. I was playing around with my car today to see how smooth I could get the startup. I also found that raising the slipstart seemed to get rid of some very low speed jerkiness I had in select parking lot scenarios.

I'll give it a try. Thank you.

[Rx7FD]

After a short drive yesterday, my main contactor shut off, and then my precharge resistor and relay overheated/ caught on fire. The cause: my tesla motor appears to be shorted out, as in the resistance between the 2 power leads is 0 ohms. Any ideas on what this might be? I'm hoping it's not catastrophic.

[Jack Bauer]

Sadly this sounds like an inverter failure. Is the short in both directions?

[Jack Bauer]

Just an fyi for everyone once the current stock of Tesla boards is exhausted that will be the end of complete boards being available from the evbmw.com webshop for the forseable future. The logistical overheads are just too great and I need the bandwidth to focus on new projects. Bare pcbs will continue to be available and as always designs both present and future will remain open and free.