Tesla Large Drive Unit Support Thread

[Jack Bauer]

Your encoder is connected backwards.

[Roadstercycle]

Jack, I upgraded this morning right after your new video which answered my charger question, Thank you.

Also I want to say thank you for doing this for all the guys that wanted very badly to run a Tesla motor. I am not electrically inclined when it comes to can bus and circuit board design but if I was I would be helping you all I could with these projects. I have an idea for the model 3 motor. I'll post over in the Model 3 page.

[muehlpower]

Hello, does anyone know if there is a harware difference between base and sport LDUs?

[Lee Malo]

Hello, does anyone know if there is a hardware difference between base and sport LDUs?

Around 7/2017 and up , all of the drive units are the same . said In another way , from 6-2017 and down the LDU are different between the performance unit and the NON performance units . The windings and inverter are built different, I was told by a Tesla technician .

[Fierobsessed]

Hello, does anyone know if there is a harware difference between base and sport LDUs?

Definitely a different inverter. The IGBT's in the base unit are rated at 75A, while the sport has 120A rated IGBT's that can peak even higher than that, the leads will melt though if they are hit harder than that for any real length of time. I have no info as to whether or not there is any difference in the motor at all. But I would be willing to bet that the Sport LDU has lower resistance across the phase windings.

[Rx7FD]

I got the motor working on 50v, but now with 350v, it no longer works. It is getting the forward input and a correct throttle input, but the motor does not spin or make any sounds at all. When the throttle position goes above 60%, the main contactor is opened and there is an error saying overcurrent. When graphing dc current, it remains at 0. Also, the dc voltage reading is incorrect (421v instead of the true 350v).

Any ideas on what the issue could be?

[Jack Bauer]

Have you reloaded the high voltage parameter file?

[Rx7FD]

How would I reload that file?

I have reset the parameters over wifi, it didn't work. I have also changed the max voltage parameters to no avail.

[Jack Bauer]

I would suggest learning a little more about the system before proceeding further. Start with the wiki here :
https://openinverter.org/wiki/Main_Page

Parameter files are available in this thread :
viewtopic.php?f=10&t=195

[Rx7FD]

Thanks for the info, I'll let you know how it goes

[Roadstercycle]

Brake switch input. I am curious if the brake switch input is just for shutting off cruise control or is there also another function. The reason I ask is because if one where to use a brake light wire for the input and the 4 ways are turned on will it just effect the cruise control or will it do something in the motor also. Just wondering.

[Rx7FD]

Thank you Damien, got it working after uploading your parameter file. One more question, I think reverse and forward are switched. I have the motor in the same orientation as the model s, power leads come upwards on passenger side. However, in forward mode, the car moves in reverse. Is this normal? Just a slight error in the programming? Thanks for all the help!

[jon volk]

Yes this is normal. F and R are reversed on the tesla setup. Just wire it accordingly for your switches.

[jon volk]

Brake switch input. I am curious if the brake switch input is just for shutting off cruise control or is there also another function. The reason I ask is because if one where to use a brake light wire for the input and the 4 ways are turned on will it just effect the cruise control or will it do something in the motor also. Just wondering.

The brake input also triggers the brake pedal regen as well as ignore throttle input as a safety. Use the switch at the pedal. If you take of off the emergency flashers, it’ll drop throttle input on every blink.

[dougyip]

I'm getting strange Input values. Board is fully connected except for HV and running V4.57. Digital inputs - start/forward/reverse read okay. Pot value reads okay, but potnom reads 0 regardless of potmin/potmax values. Voltage (uaux) reads a constant 16.43 volts which is incorrect - actual voltage is 12.7. Tmphs reads okay, but tmpm alternately reads the correct value and 0. Interesting thing is digital input din_BMS alternately reads 0/1 at exactly the same frequency as tmpm.

[Jack Bauer]

Uaux is not implemented on the Tesla boards.
One of the motor temp sensors is not implemented hence the jumping value.
The bms pin is used to switch between the motor temp sensors hence the pattern you observed.
Potnom will not change until you are in run mode.

[dougyip]

Thanks Damien. I've now got the motor spinning at full voltage. Had to swap encoder wires and change "ocurlim" to something < -500 ( I see you use -2500). The temperature sensors that are not implemented are the stator temp sensors. They are very important to us for road racing. Stator temps reach 170 degC within about 5-6 laps of a 16 lap race. The stock Tesla controller cuts back on max power to keep temps from rising further. Can we use an external controller to send CAN messages to modify iacmax, or idcmax, in real time to accomplish the same thing?

[Jack Bauer]

Why not plot the temp vs resistance curve of the missing sensor and Johannes can add it to the firmware.

[dougyip]

I see that it will be an issue to plot the curves for the stator temperature sensor Mot1 and Mot2. Since the sensors are buried inside the stator casing, I guess I will have to find a large oven! Nominal value at 15 degC is 148k and value lowers with temp, so they are probably some type of 100k NTC thermistor.

[Jack Bauer]

Some good news on board supply. I have now acquired a much more reliable pcb manufacturer here in Ireland and boards will be in stock for immediate dispatch. No more silly delays and boards are being built on an automated smd line so quality is much improved.

[Kevin Sharpe]

I see that it will be an issue to plot the curves for the stator temperature sensor Mot1 and Mot2. Since the sensors are buried inside the stator casing, I guess I will have to find a large oven!

You could try asking Zero EV if they have any redundant LDU parts they can sell you... then you would only need to put a small part in the oven

[dougyip]

I actually have a LDU with a blown inverter, so will strip it down and use that for testing.

[arber333]

I see that it will be an issue to plot the curves for the stator temperature sensor Mot1 and Mot2. Since the sensors are buried inside the stator casing, I guess I will have to find a large oven!

You could try asking Zero EV if they have any redundant LDU parts they can sell you... then you would only need to put a small part in the oven

I dont think that is neccessary. I plotted various temp sensors for Volt. Prius and Leaf components by running hot coolant through them.
You need one 2kW dishwasher heater, 12V coolant pump and laser temp reader or other reference. You do have to bypass heaters thermostat though. Maybe to control temperature you can use digital thermostat relay. They are not expensive. Then temperature is maintained in hoses and stator heats up.
Then you just have to maintain temperature untill stator sensor stops changing value and record that.
Then increase temperature on thermostat by 5deg and wait for stator value to stabilize again.

For temperature over 100deg use 100% ethylene glycol (clear liquid) which has boiling point at 200deg.

[dougyip]

The problem with running water through the coolant channels is that there will be a significant temperature gradient across the stator. One side will be at the coolant temperature, the other will be at the air gap temperature. The actual sensor temperature will be somewhere between. I have actually located an industrial oven I can use.

[Kevin Sharpe]

I actually have a LDU with a blown inverter, so will strip it down and use that for testing.

Great, it will be good to have this final temperature sensor supported