Leaf Gen 1 Inverter Board

[KarmannLeafBlower]

That is very cool.
I am desperate for a converter plate and coupler solution for the Karmann Ghia/Leaf project I am working on. How did you do yours? Or did you change the transmission also?
Best wishes,
Edd

[Kevin Sharpe]

I am desperate for a converter plate and coupler solution for the Karmann Ghia/Leaf project I am working on. How did you do yours?

Have a look at this thread;

viewtopic.php?f=12&t=569

Also note comments re KISS;

viewtopic.php?f=12&t=569&start=20#p10676

[KarmannLeafBlower]

Thanks Kevin, that's helpful. Digested both threads already.
What's your reference to KISS? Keep it simple, stupid?

[Kevin Sharpe]

What's your reference to KISS? Keep it simple, stupid?

Yeah... sometimes we make this stuff way to complex... so much depends on your skills and the approvals regime you're operating in

[langoo]

That is very cool.
I am desperate for a converter plate and coupler solution for the Karmann Ghia/Leaf project I am working on. How did you do yours? Or did you change the transmission also?
Best wishes,
Edd

See PM

[mackoffgrid]

Hi guys

I have my Gen1 motor at last.

I have the harness all wired up and the CAN bus is communicating correctly so far !!!

Without taking the lid off the inverter I have not been able to determine the the polarity of the Inverters High Volt DC input??

If someone can help by drawing the +ve and -ve on this photo I'd appreciate it.
I'm not sure if the photo I've tried to attache has worked or not?

[mackoffgrid]

I don't seem to able to add an attachment so I'll try a link to github.

https://github.com/mackelec/SolarUte/bl ... ctions.jpg

URL Gen1 Inverter DC connections

[RetroZero]

If I remember from my school lesson at 4 degrees in March this year....Teacher taught us the light bulb method.
1. Take BLACK wire.
2. Take RED wire and connect car light bulb in series in the cable.
3. Connect BLACK wire to NEG of power supply (12V car battery) AND ANY of terminals on Inverter.
4. Connect RED wire (with light bulb) to POS of power supply and the OTHER terminal on Inverter.
5. If light comes ON , it means the polarity is INCORRECT. If not - BLACK wire is to NEG and RED wire is to POS.
I had trouble with this basic yet VITAL test.
I think of the way you connect two batteries together in parallel, where you connect + with + , and - with -. If you put a light bulb in series into iether of your cables, it should NOT light up - no change in polarity.

Hope the teacher managed to get this ONE important point into my thick skull. If not, I'm sure there will be a comment or 3000 in a few seconds on how I blew up your inverter.

Please ask for a second opinion! I would not want to be liable for any inverter blow-ups.

[mackoffgrid]

Thanks Retro, an old technicians trick

I think I've worked out the polarity, it is in the TMS manual - just had to study it again.

[Cookie6000]

Great success the last few days in getting the evbmw gen 1 vcu working in the car. Works really well when you have the correct inv_volts and presets in there for the pack voltages from the prius pack I am using for testing and tweaking the throttle values to suit the pedal. Next up is just to wire up the shunt and get that feeding info back to the web interface.

The one problem I have having is delivery of power. I can crawl around no issue but try give even a LITTLE bit of welly and the inverter craps out. Not sure if this is a parameter issue or interference to the messaging from the resolver and or the CAN messaging back to the vcu. The board has a quirk where I cannot connect the usb and read the serial monitor when it is connected to the vehicle. As soon as I do that, I get usb connection errors. So, when I make updates, I have to disconnect the board from the car. So, any ideas on the current draw and following inverter fart?

[zippy500]

Just a shot, May or may not be the same problem,I know yours is a Gen1, What is your pack voltage ?, I was running a can based controller on a gen2 invertor on 200v( fully charged) SLA batteries , it ran fine at low throttle so as you gave it some beans, the invertor would cut out,

Turns out that the leaf invertor cuts out at 152v from load, battery sag, can you check your voltage under load ?

[Cookie6000]

Sounds like a good prognosis zippy
Its one possibke issue I thought might be hitting me alright. The prius pack I have was up around 209, now down around 198v. I'll try measure it tomorrow and report back. Thanks

[james@N52E01]

We’ll find out soon enough when you’ve got the CAN shunt wired in

[mackoffgrid]

I bench tested my Leaf Gen1 / VCU using rectified mains power. I determined that the Inverter throws an error when the voltage drops at about 175 - 185 volts.

It's not hard to imagine a prius battery plus voltage drop drooping enough to trigger an error when you called for extra current.
Add an extra 12 volt battery (or 2) in series to help if necessary.

[Cookie6000]

I bench tested my Leaf Gen1 / VCU using rectified mains power. I determined that the Inverter throws an error when the voltage drops at about 175 - 185 volts.

It's not hard to imagine a prius battery plus voltage drop drooping enough to trigger an error when you called for extra current.
Add an extra 12 volt battery (or 2) in series to help if necessary.

Fantastic mackoffgrid. Thanks for confirming that for us I was pretty close to the mark at 195v left in the pack to trigger the error. I'm focussing on getting the full Leaf pack in. If I get a chance, I'll add a few more 12v in the system just for a test spin.

[james@N52E01]

Hi all, I just finished configuring the Isabellenhütte IVT-S CAN shunt to work with the Gen 1 Leaf VCU board. I made the following modifications to v1.5 of the Leaf VCU code to get this to work. See the changes below

BEFORE

Code: Select all

//////timers//////////////////////////////
Metro timer_Frames10 = Metro(10);
Metro timer_Frames100 = Metro(100);
Metro timer_wifi = Metro(1100);
Metro timer_hv = Metro(1000);


int inv_volts_local;
int16_t final_torque_request = 0;
#define INVERTER_BITS_PER_VOLT 2
#define INVERTER_BITS_PER_RPM 2

I added the following lines in the middle

AFTER

Code: Select all

//////timers//////////////////////////////
Metro timer_Frames10 = Metro(10);
Metro timer_Frames100 = Metro(100);
Metro timer_wifi = Metro(1100);
Metro timer_hv = Metro(1000);

float Version=2.00;           
uint16_t loopcount=0;         
unsigned long startime=0;     
unsigned long elapsedtime=0;  
uint port=0;               
uint16_t datarate=500;        

int inv_volts_local;
int16_t final_torque_request = 0;
#define INVERTER_BITS_PER_VOLT 2
#define INVERTER_BITS_PER_RPM 2

Also you will need to change

BEFORE

Code: Select all

void setup() 
  {
  Can0.begin(CAN_BPS_500K);   // Inverter CAN
  Can1.begin(CAN_BPS_500K);   // Vehicle CAN
  //Can0.watchFor(0x1da); //set message filter for inverter can. Note sure if I can use two seperate values here. it might just pick 1!
  Can0.watchFor();
  Can1.watchFor(0x1ff); //just a blank message to block receive from e46 messages.
    
    Serial.begin(115200);  //Initialize our USB port which will always be redefined as SerialUSB to use the Native USB port tied directly to the SAM3X processor.
    Serial2.begin(19200);  //Serial comms with ESP32 WiFi module on serial 2
   // Timer3.attachInterrupt(Msgs10ms).start(10000); // 10ms CAN Message Timer
   // Timer4.attachInterrupt(Msgs100ms).start(100000); //100ms CAN Message Timer
    
  pinMode(led, OUTPUT);
  pinMode(Brake, INPUT);
  pinMode(IN1, INPUT);  //T15 input from ign on switch
  pinMode(IN2, INPUT);
  pinMode(OUT1, OUTPUT);
  pinMode(OUT2, OUTPUT);
  pinMode(OUT3, OUTPUT);
  

I added the following line

AFTER

Code: Select all

void setup() 
  {
  Can0.begin(CAN_BPS_500K);   // Inverter CAN
  Can1.begin(CAN_BPS_500K);   // Vehicle CAN
  //Can0.watchFor(0x1da); //set message filter for inverter can. Note sure if I can use two seperate values here. it might just pick 1!
  Can0.watchFor();
  Can1.watchFor(0x1ff); //just a blank message to block receive from e46 messages.
    
    Serial.begin(115200);  //Initialize our USB port which will always be redefined as SerialUSB to use the Native USB port tied directly to the SAM3X processor.
    Serial2.begin(19200);  //Serial comms with ESP32 WiFi module on serial 2
    Sensor.begin(port,datarate);  //Start ISA object on CAN 0 at 500 kbps
   // Timer3.attachInterrupt(Msgs10ms).start(10000); // 10ms CAN Message Timer
   // Timer4.attachInterrupt(Msgs100ms).start(100000); //100ms CAN Message Timer
    
  pinMode(led, OUTPUT);
  pinMode(Brake, INPUT);
  pinMode(IN1, INPUT);  //T15 input from ign on switch
  pinMode(IN2, INPUT);
  pinMode(OUT1, OUTPUT);
  pinMode(OUT2, OUTPUT);
  pinMode(OUT3, OUTPUT);
  

and finally

BEFORE

Code: Select all

Serial2.print("v");//dc bus voltage
Serial2.print(Sensor.Voltage);//voltage derived from ISA shunt
Serial2.print(",i");//dc current
Serial2.print(Sensor.Amperes);//current derived from ISA shunt
Serial2.print(",p");//total motor power
Serial2.print(Sensor.KW);//Power value derived from ISA Shunt
Serial2.print(",m");//motor rpm
Serial2.print(inverter_status.speed);
Serial2.print(",o");//motor temp
Serial2.print(inverter_status.motor_temperature);
Serial2.print(",r");//inverter temp
Serial2.print(inverter_status.inverter_temperature);
Serial2.print("*");// end of data indicator

Sensor.Voltage needed to be changed to Sensor.Voltage3

AFTER

Code: Select all

Serial2.print("v");//dc bus voltage
Serial2.print(Sensor.Voltage3);//voltage derived from ISA shunt
Serial2.print(",i");//dc current
Serial2.print(Sensor.Amperes);//current derived from ISA shunt
Serial2.print(",p");//total motor power
Serial2.print(Sensor.KW);//Power value derived from ISA Shunt
Serial2.print(",m");//motor rpm
Serial2.print(inverter_status.speed);
Serial2.print(",o");//motor temp
Serial2.print(inverter_status.motor_temperature);
Serial2.print(",r");//inverter temp
Serial2.print(inverter_status.inverter_temperature);
Serial2.print("*");// end of data indicator

Hope this helps.

[Cookie6000]

Great work James! Going to try this all out once I get the full 24kWh battery in the car. Starting battery box making tomorrow...

[tom91]

How does the direction control work on the canbus?

Is this like the Outlander control where negative torque causes the motor to decelerate and then spin the other way? Or is there a byte/bit that toggles direction of motor rotation?

I have been digging though the code but not really been able to find a true decisive answer.

[james@N52E01]

Code: Select all

void readPedals()
{
ThrotVal = analogRead(Throttle1); //read throttle channel 1 directly
ThrotVal = constrain(ThrotVal, 145, 620);
ThrotVal = map(ThrotVal, 145, 620, 0, MaxTrq); //will need to work here for cal.
if(ThrotVal<0) ThrotVal=0;  //no negative numbers for now.
if(digitalRead(Brake)) ThrotVal=0;  //if brake is pressed we zero the throttle value.
//Serial.println(ThrotVal);   //print for calibration. 
}

Currently it doesn’t look like the code accepts negative torque commands.

[tom91]

Currently it doesn’t look like the code accepts negative torque commands.

Can someone confirm why this is? Does it have anything to do with direction control?

[james@N52E01]

I could be way off but best guess is V1.5 doesn’t employ regen yet as this just needs to be written into the code based off the brake sensor input which isn’t being used yet.

For most conversions it‘s not necessary to have a command feature to run the motor in reverse as you have a gearbox for that (unless like me you’re using the leaf transaxle).
Edit: or a system like Tesla motor or 450h gearbox

In the early versions of the code, the motor would only accept reverse torque commands so it might be worth having look at the differences between the versions. It’s somewhere on my to do list but C coding isn’t my strong suit

[james@N52E01]

One feature I am trying to add as well is adaptive regen using an analogue brake sensor input in place of throttle input 2.

[Cookie6000]

As is the way of electric conversions, a few steps forward lead to a few backward. She refuses to move
Yesterday morning (Friday) I got in, turned the key and initiated the usual three clicks as the pre ch/neg/main contactors did their business and back we went with no issue. I had planned to weigh the car so set up the scales, got back in and turned the key, 'click', click',...... nothing. And that was the theme for the rest of the day....

For whatever reason, the main contractor is not triggering when it is supposed to.
Things checked:
Yes, service plug is in We have 378v at the contactors.
When the ignition is switched, both Neg and PreCh trigger fine feeding the full 378v to the inverter via the 100 Ohm Arcol precharge resistor. I am able to hard trigger the contactors by connecting to 12v separately so they all work ok. Pch resistor too looks ok. Using a household filiment bulb for all these tests too show we are getting up to where we want to be for the inverter to go "yep! I see the volts, I approve of the main contactor being triggered!"
Inverter is powering up and all the voltages on the board look as they should be.
On the Leaf VCU - I was still running on the firmware revision v1.5 (with shunt and contactor functionality) I had the car running on back with the Prius pack which was set to >190v and >190v and the start-up procedure worked perfectly on Thursday with no changes made. HVPreset is set to 190v which is the voltage to enable main contactor. Here is the main triggering code though which worked on Thursday:

Code: Select all

  inv_volts_local = (inverter_status.voltage / INVERTER_BITS_PER_VOLT);


  if (T15Status && !Pch_Flag)  //if terminal 15 is on and precharge not enabled
  {
    digitalWrite(OUT3, HIGH);  //inverter power on
    if (inv_volts_local <190)
    {
      digitalWrite(OUT1, HIGH);  //precharge on
      Pch_Flag = true;
    }
  }
  if (T15Status && !HV_Flag && Pch_Flag)  //using inverter measured hv for initial tests. Will use ISA derived voltage in final version.
  {
    if (inv_volts_local>190)
    {
      digitalWrite(OUT2, HIGH);  //main contactor on
      HV_Flag = true; //hv on flag
    }
  }

I began changing the voltage settings to closer to the new pack values. Multiple changes resulted in nothing apart from one or two occasions when the main contactor triggered after 30 seconds but by then, there was no response from the throttle. CAN comms? When I log into the WiFi module, no pack voltage or motor/inv temp.... hmmm

When I put the meter across the 12v trigger wires in the HVJB for precharge, neg and main, I can see the pch and neg have 12v triggering them but 0.6v on the Main+ showing the command hasn't been initiated on the VCU. After hours of checking everything from VCU to inverter, all connectors, wiring, I began to figure that there was some CAN handshake not occurring for the Main contactor. On another note, I have yet to wire up the Shunt so voltage readings are being taken from the inverter but I am still on the previous firmware rev without those updates, again, which all worked fine on Thursday. A presumed duff USB cable is currently not allowing me to view the serial monitor while the VCU is connected to the car which is one PITA and halting my investigation a touch to add to the confusion.

Any ideas, questions, thoughts? Everything worked fine till 2nd ignition turn Friday yesterday morn...

[Kevin Sharpe]

When I put the meter across the 12v trigger wires in the HVJB for precharge, neg and main, I can see the pch and neg have 12v triggering them but 0.6v on the Main+ showing the command hasn't been initiated on the VCU.

Where precisely are you measuring these control signals? The contactor drivers in the VCU are open collector, active low, so you'd expect them to be 12V when inactive and ~0V when active;

viewtopic.php?f=4&t=179&start=110#p10163

Are you sure the main contactor signal wiring/connector is ok? Maybe the driving the car has dislodged a connection.

Is the Leaf BMS connected to the cars CAN network? Could it be shutting down the inverter somehow?

[Cookie6000]

When I put the meter across the 12v trigger wires in the HVJB for precharge, neg and main, I can see the pch and neg have 12v triggering them but 0.6v on the Main+ showing the command hasn't been initiated on the VCU.

Where precisely are you measuring these control signals? The contactor drivers in the VCU are open collector, active low, so you'd expect them to be 12V when inactive and ~0V when active;

viewtopic.php?f=4&t=179&start=110#p10163

Yep, I remember that from when I was trying to get it up and running the last time. I need to double check again at the VCU board instead of at the HVJB but pretty sure I was taking measurements on the + side of the contactors. WIll be going at it again tomorrow to check the functions on the board.

Are you sure the main contactor signal wiring/connector is ok? Maybe the driving the car has dislodged a connection.

Is the Leaf BMS connected to the cars CAN network? Could it be shutting down the inverter somehow?

Checked it over and over, top to bottom, even reterminated some connectors. BMS not hooked up yet so, that it out of the equation at the mo. I did notice some small strands of copper from when I was making a connector that were sitting on the board. Not sure if they would have shorted something. I'll just work my way back through the board again and see if there is anything doing what it shouldn't be doing.