User avatar
JeremyW
Posts: 1522
Joined: Sun Nov 13, 2011 12:53 am
Delivery Date: 23 Jun 2012
Leaf Number: 19136
Location: San Gabriel, CA

Re: LEAF CANbus decoding. (Open discussion)

Fri Sep 26, 2014 1:40 am

In the google doc we have a few "is this speed/volts/torque?" for a few of the high priority (10ms) messages. I thought of two little tricks to find some of the inverter messages: while in reverse at say a few MPH, throw the car into drive. With the parking brake on, put the car into drive and put a little throttle on to make the motor push against the brakes. From those two tests (in a secluded area with nobody around!) I found the following on the EV CAN bus:

Message 1D4 bytes 2-3 2's complement is in fact commanded torque. Units are unknown but I've confirmed the value is positive while the car was still going backwards but "accelerating" forward (after I put it into drive from reverse while moving).

Message 1DA bytes 4-5 2's complement is motor RPM. Units are again unknown but this one should be easy to figure out since we know wheel speed from other values as well as the reduction gear ratio. Probably 0.5 rpm/bit.

Message 1DA byte 8 with the mux identifier as byte 7 shows the individual phase currents for the motor, including neutral. Byte 7 00-02 are the 3 phases and 03 is the sum or neutral. Even at 10ms per message there is significant undersampling.

Anyway I have about 90% confidence on these values. I encourage anyone with a bit of space around to try the above tricks and see what's what. If you have a 2011 or 2012 with the electronic parking brake note that if you hammer on it too much, it'll automatically disengage!
Former 2012 SL leasee 6/23/12 - 9/23/15
Now driving Honda Fit EV, License plate: CHADEMO
2000 Honda Insight for long trips

User avatar
TickTock
Posts: 1701
Joined: Sat Jun 04, 2011 10:30 pm
Delivery Date: 31 May 2011
Leaf Number: 3626
Location: Queen Creek, Arizona
Contact: Website

Re: LEAF CANbus decoding. (Open discussion)

Fri Sep 26, 2014 7:26 am

JeremyW wrote:In the google doc we have a few "is this speed/volts/torque?" for a few of the high priority (10ms) messages. I thought of two little tricks to find some of the inverter messages: while in reverse at say a few MPH, throw the car into drive. With the parking brake on, put the car into drive and put a little throttle on to make the motor push against the brakes. From those two tests (in a secluded area with nobody around!) I found the following on the EV CAN bus:

Message 1D4 bytes 2-3 2's complement is in fact commanded torque. Units are unknown but I've confirmed the value is positive while the car was still going backwards but "accelerating" forward (after I put it into drive from reverse while moving).

Message 1DA bytes 4-5 2's complement is motor RPM. Units are again unknown but this one should be easy to figure out since we know wheel speed from other values as well as the reduction gear ratio. Probably 0.5 rpm/bit.

Message 1DA byte 8 with the mux identifier as byte 7 shows the individual phase currents for the motor, including neutral. Byte 7 00-02 are the 3 phases and 03 is the sum or neutral. Even at 10ms per message there is significant undersampling.

Anyway I have about 90% confidence on these values. I encourage anyone with a bit of space around to try the above tricks and see what's what. If you have a 2011 or 2012 with the electronic parking brake note that if you hammer on it too much, it'll automatically disengage!


Interesting idea but I am not convinced it really provides insight. Both commanded torque and motor amps will be positive if imparting positive torque - regardless of the motor rotation direction. This occurs when accelerating forward (top right scenario) or decelerating reverse (bottom right) which is the mode you entered when you threw it into D from R. Similarly, during forward regen (top left), 1D4:23 goes negative when regen braking although you are still rotating the same direction. So I don't think we can make any conclusion on whether it is commanded torque or actual current or torque.

Image

Also, how did you conclude that 1D4:45 is motor rpm vs. motor Volts (Volts is directly proportional to motor speed)?

User avatar
JeremyW
Posts: 1522
Joined: Sun Nov 13, 2011 12:53 am
Delivery Date: 23 Jun 2012
Leaf Number: 19136
Location: San Gabriel, CA

Re: LEAF CANbus decoding. (Open discussion)

Fri Sep 26, 2014 10:23 am

TickTock wrote:Also, how did you conclude that 1D4:45 is motor rpm vs. motor Volts (Volts is directly proportional to motor speed)?

Because the value was 0 even when applying a torque with the parking brake on. There should be some voltage present, because when you look at pack current or even the phase currents there is a current flowing through the motor with the rotor locked.*

The value I believe is commanded torque has no noise, where as everything else I've seen from the inverter/battery that is a real world value has some noise on it. Further, it always leads the motor rpm message and motor amps. It has a no inertia vs the rpm message which seems "realistic" in it's inertial qualities. Motor amps has more noise and is "softer" than the commanded torque.

*Note if you try this don't do it for very long! This is a good way to cook the rotor.
Last edited by JeremyW on Fri Sep 26, 2014 10:45 am, edited 1 time in total.
Former 2012 SL leasee 6/23/12 - 9/23/15
Now driving Honda Fit EV, License plate: CHADEMO
2000 Honda Insight for long trips

User avatar
JeremyW
Posts: 1522
Joined: Sun Nov 13, 2011 12:53 am
Delivery Date: 23 Jun 2012
Leaf Number: 19136
Location: San Gabriel, CA

Re: LEAF CANbus decoding. (Open discussion)

Fri Sep 26, 2014 10:29 am

TickTock wrote:decelerating reverse (bottom right) which is the mode you entered when you threw it into D from R.
Note that the leaf does not regen when going in reverse if you are using the regular brakes. In this case the leaf will use the friction brakes to stop, and the "commanded torque" value should be near zero.
Former 2012 SL leasee 6/23/12 - 9/23/15
Now driving Honda Fit EV, License plate: CHADEMO
2000 Honda Insight for long trips

User avatar
TickTock
Posts: 1701
Joined: Sat Jun 04, 2011 10:30 pm
Delivery Date: 31 May 2011
Leaf Number: 3626
Location: Queen Creek, Arizona
Contact: Website

Re: LEAF CANbus decoding. (Open discussion)

Fri Sep 26, 2014 11:13 am

JeremyW wrote:
TickTock wrote:Also, how did you conclude that 1D4:45 is motor rpm vs. motor Volts (Volts is directly proportional to motor speed)?

Because the value was 0 even when applying a torque with the parking brake on. There should be some voltage present, because when you look at pack current or even the phase currents there is a current flowing through the motor with the rotor locked.*

The value I believe is commanded torque has no noise, where as everything else I've seen from the inverter/battery that is a real world value has some noise on it. Further, it always leads the motor rpm message and motor amps. It has a no inertia vs the rpm message which seems "realistic" in it's inertial qualities. Motor amps has more noise and is "softer" than the commanded torque.

*Note if you try this don't do it for very long! This is a good way to cook the rotor.

OK. I buy that - there should be some voltage from series losses.

User avatar
TickTock
Posts: 1701
Joined: Sat Jun 04, 2011 10:30 pm
Delivery Date: 31 May 2011
Leaf Number: 3626
Location: Queen Creek, Arizona
Contact: Website

Re: LEAF CANbus decoding. (Open discussion)

Fri Sep 26, 2014 11:15 am

JeremyW wrote:
TickTock wrote:decelerating reverse (bottom right) which is the mode you entered when you threw it into D from R.
Note that the leaf does not regen when going in reverse if you are using the regular brakes. In this case the leaf will use the friction brakes to stop, and the "commanded torque" value should be near zero.

Did you see regen when you threw it into D while moving in reverse? I would assume so - otherwise, where is the energy going?

User avatar
JeremyW
Posts: 1522
Joined: Sun Nov 13, 2011 12:53 am
Delivery Date: 23 Jun 2012
Leaf Number: 19136
Location: San Gabriel, CA

Re: LEAF CANbus decoding. (Open discussion)

Fri Sep 26, 2014 11:29 pm

Yes, it'll regen in reverse if you are already moving backwards and throw it in drive. But not if you touch the brakes.
Former 2012 SL leasee 6/23/12 - 9/23/15
Now driving Honda Fit EV, License plate: CHADEMO
2000 Honda Insight for long trips

User avatar
TickTock
Posts: 1701
Joined: Sat Jun 04, 2011 10:30 pm
Delivery Date: 31 May 2011
Leaf Number: 3626
Location: Queen Creek, Arizona
Contact: Website

Re: LEAF CANbus decoding. (Open discussion)

Sat Sep 27, 2014 6:08 am

Thanks. Nice sleuthing, Jeremy. I updated the 1DA entry. I suspect you are also right on 1D4:23 since it is a less noisy signal than the motor amps but will leave the ? until we devise a more positive proof like your torque while stationary voltage vs speed test. Maybe lift the wheels off the ground so there is minimal inertia to fight and floor it, then do the same on the road, and compare the results. I would assume the same requested torque (at least briefly) but should have considerably less current for the first case.

User avatar
JeremyW
Posts: 1522
Joined: Sun Nov 13, 2011 12:53 am
Delivery Date: 23 Jun 2012
Leaf Number: 19136
Location: San Gabriel, CA

Re: LEAF CANbus decoding. (Open discussion)

Sat Sep 27, 2014 6:27 am

Thanks! :) I suppose I could disconnect the inverter communication connector (the grey plug on the passenger side of the inverter) and then boot the car up and see which CAN messages disappear.
Former 2012 SL leasee 6/23/12 - 9/23/15
Now driving Honda Fit EV, License plate: CHADEMO
2000 Honda Insight for long trips

User avatar
TickTock
Posts: 1701
Joined: Sat Jun 04, 2011 10:30 pm
Delivery Date: 31 May 2011
Leaf Number: 3626
Location: Queen Creek, Arizona
Contact: Website

Re: LEAF CANbus decoding. (Open discussion)

Sat Sep 27, 2014 7:37 am

JeremyW wrote:Thanks! :) I suppose I could disconnect the inverter communication connector (the grey plug on the passenger side of the inverter) and then boot the car up and see which CAN messages disappear.

LOL! and how much the dealer wants to charge you to reset the DTCs. :-)

Return to “LEAF CANBus”