Mitsubishi outlander charger and DC:DC

[royhen99]

The Zombieveter code also provides the CAN messages to control the charger. It ramps up to 12A if not contrained by an external PWM on CP.

[AdrianB]

The Zombieveter code also provides the CAN messages to control the charger. It ramps up to 12A if not contrained by an external PWM on CP.

Thank you royhen99. Is this documented somewhere? I didn’t know the Zombieverter supported this onboard charger.

[royhen99]

It's not documented as far as I know. The hardware side is wire every thing up, HC, AC, 12V supply, 12V o/p and sense for dc-dc converter , wire charger cp to charge port or to pwm from oil pump output. Select CAN bus 1 or 2 for ChargerCan, chargemodes - Out_lander, Chgctrl - enable, depending on which version of code it will either try to start charging immediately or wait for valid cp duty cycle to be read from the charger before starting. If using the oil pump pwm the charge current will be maximum the charger can do and is too much for a granny cable. If charger cp is conected to the charge socket then the charger will limit the current based on cp duty cycle.

[Alibro]

Hi guys
I'm getting some odd behaviour from my Outlander charger and wondered if anyone else has seen the same issue.
I got the car on the road a couple of months ago and have done around 600 miles so have charged it maybe 30 or 40 times give or take.
Initially it charged without any issues every time I plugged it in but for the last week or two I have found the charge has stopped after 30 minutes or so. I have been beside it when the charge stops and I noticed the coolant pump changed in tone for a second or so which indicates the 12V output either spiked or dipped, not sure which yet but I hope to figure it out soon. I also took a CAN dump of the charge but am still looking through it. If I spot anything I'll report back here.
When it first happened I was using a Podpoint EVSE I'd bought off Ebay and was charging at 3.4kW. I then tried charging with a Nissan Leaf Granny charger which charges at 2.2kW but the same thing happened.

I'm using Arber333's Due code to control the charger but when it happens if I reset the Due there is no change, I need to reset the charger for it to start charging again.

When I gather more data I'll get back but wanted to ask here if it has happened before.

I'm not sure if it is related but if I drive the car from cold it is mostly reliable but yesterday after a couple of attempts at charging I took it for a drive and the car cut out three times within a couple of miles. One of the times it cut out was exactly the same time I turned on the lights.

[Bigpie]

My outlander charger has been faultless for last 2 years and 10k miles. Stick up a CAN log and I'll see if anything jumps out.

[arber333]

I'm using Arber333's Due code to control the charger but when it happens if I reset the Due there is no change, I need to reset the charger for it to start charging again.

No need to worry about it, i noticed the same.
Symptoms; It seems that when charger is allways connected to HVDC and pin8 is connected to 12V power supply for a longer time charger sometimes goes out. Also i noticed that when i keep charger connected to 12V and remove the EVSE plug sometimes charger requires reset.

Solution; I simply connect pin8 of charger control connector to a relay which i control by ignition or by PP pin from EVSE. This relay must come on before DCDC enable pin4 is issued as DCDC part is also supplied from the charger logic supply.

In addition now i connected Charger behind main contactor and i require to perform precharge everytime i charge and drive. If done properly it doesnt mean any difference...

[Alibro]

No need to worry about it, i noticed the same.
Symptoms; It seems that when charger is allways connected to HVDC and pin8 is connected to 12V power supply for a longer time charger sometimes goes out. Also i noticed that when i keep charger connected to 12V and remove the EVSE plug sometimes charger requires reset.

Solution; I simply connect pin8 of charger control connector to a relay which i control by ignition or by PP pin from EVSE. This relay must come on before DCDC enable pin4 is issued as DCDC part is also supplied from the charger logic supply.

In addition now i connected Charger behind main contactor and i require to perform precharge everytime i charge and drive. If done properly it doesnt mean any difference...

My outlander charger has been faultless for last 2 years and 10k miles. Stick up a CAN log and I'll see if anything jumps out.

Thanks for getting back guys.
Up until now I had to watch the charging like a hawk to avoid over voltage but the last week or two it will only charge for 30 minutes or less. If I reset it I might get another 30 minutes but often not.

It never occurred to me that having CAN going to the charger all the time while driving the car might cause an issue so I can address that.

The charger only gets HVDC when ign is on via the precharge and a 30A fuse.

I have PIN 8 only coming on when the ign is on. In fact 12V ign, 12V power and 12V sense are all in the same connector and switched on every time I turn the ign and it has been working fine up until now. Is this an issue?

When I was trying to figure out the startup sequence I tried connecting the charger/dc to dc and inverter multiple different ways and found it difficult to find a sequence that worked. I never figured out why connecting the charger one way could prevent the inverter from turning on but I suspected there may be a back feed voltage from the one of the charger connections screwing with the inverter.

I'm still going through the CAN dump so will post shortly

[Alibro]

This is the CAN dump from the charger/Due

It was over 29Mb so I removed the frame ID's that did not change the entire time listed below and reduced the time frame
0285, 0286, 0563, 0568, 0569, 0663, 0664

038A was climbing up until 12:55:49 then starts to fall.

Another odd thing I noticed after the charging stopped was the Inverter reporting the wrong voltage on the WiFi display it was indicating 360V when the pack was at 380V and only corrected itself when I turned the ign off and on again

[Alibro]

Here is the full log from around 20 minutes before it failed.

[Bigpie]

What was that logged with? doesn't open with savvycan

[arber333]

When I was trying to figure out the startup sequence I tried connecting the charger/dc to dc and inverter multiple different ways and found it difficult to find a sequence that worked. I never figured out why connecting the charger one way could prevent the inverter from turning on but I suspected there may be a back feed voltage from the one of the charger connections screwing with the inverter.

Hm... can you elaborate on that "interfeering with the inverter" thing. I am still working on my mazda convertion to full Outlander system and i havent noticed any problem. I didnt go too far with RPM yet though.

I reconfigured 0x285 msg to operate from a fixed 10ms timer
I set EVSE function to transmitt 0x286 traffic with charging content only at charging occasion
I set Inverter function to transmitt 0x286 traffic with driving content only when FWD, REV, or N!

[Alibro]

What was that logged with? doesn't open with savvycan

I bought a CanAlyst-2 a couple of years ago and as it was installed on my laptop I used it. I read the file using LibreCalc and used auto filter to focus on certain frame ID.
This is it on Ebay
https://ebay.us/Fa7Ean

Hm... can you elaborate on that "interfeering with the inverter" thing. I am still working on my mazda convertion to full Outlander system and i havent noticed any problem. I didnt go too far with RPM yet though.

I reconfigured 0x285 msg to operate from a fixed 10ms timer
I set EVSE function to transmitt 0x286 traffic with charging content only at charging occasion
I set Inverter function to transmitt 0x286 traffic with driving content only when FWD, REV, or N!

Like I said I tried multiple ways to connect the inverter and charger and sometimes the inverter would not enable drive at all and other times I had to turn the ign off for ten seconds or more before turning back on or I got no drive. I think maybe one of the charger connections (not sure which) has a few volts on it when the charger is powered up with 12V and this might have been enough to turn on the inverter which is a problem if it is on too long with no input. TBH I would need to go back through all the configurations to be sure.
My inverter is a Nissan Leaf Gen1 and VCU is Damiens Gen1 version which was the predecessor to the Zombieverter.

[Bigpie]

Damn, Savvycan doesn't seem to support that format. I tend to like using it to decode with a DBC and plotting various values.
Can it it export into a Savvycan supported format?

[Alibro]

The software supplied with this tool doesn't allow opening files so I'll have a go at capturing using Savvycan.

Been a while since I touched it.

[royhen99]

The data being sent on 0x285 does not look correct, A4 40 AA 00 DF C0 xx xx
I thought it should be 00 00 B6 00 00 00 00 00 for charging.
From the log it looks like the charge may have terminated at 375V. Hard to tell exactly as only 1V resolution.

[arber333]

You can see the messages here
https://mazdamx3ev.wordpress.com/2023/1 ... -commands/

Or what i used to control teensy
viewtopic.php?p=65840#p65840

I found out it is very important to use timer to keep 0x285 strict 10ms timing.

[tom91]

On my projects I broadcast the 0x285 messages at 100ms

with the content of 0x00 0x00 0xB6 0x10 0x78 0x00 0x00 0x10

this all works fine for me and does not have any issues of hanging or stopping charge without intent.

[Alibro]

The data being sent on 0x285 does not look correct, A4 40 AA 00 DF C0 xx xx
I thought it should be 00 00 B6 00 00 00 00 00 for charging.
From the log it looks like the charge may have terminated at 375V. Hard to tell exactly as only 1V resolution.

Do you mind if I ask how to extract the voltage from the CAN?

Damn, Savvycan doesn't seem to support that format. I tend to like using it to decode with a DBC and plotting various values.
Can it it export into a Savvycan supported format?

I spent most of the evening figuring out how to setup a Due for SavvyCan so hopefully tomorrow I'll be able to do some more testing and get more CAN logs.

[arber333]

On my projects I broadcast the 0x285 messages at 100ms

Its strange but charger works with 100ms and inverter works with 100ms, but heater seem to work with less than 30ms. I seem to remember AC compressor needs 30ms as well. Due to easier application i simply setup a single timer for one 0x285 telegram at 10ms.

[royhen99]

Do you mind if I ask how to extract the voltage from the CAN?

The voltage reading comes from ID 0x389 byte 0. From your data, when charging, this reached a maximum of 374V. Byte 0 data 0xBB = 187, 187 x 2 = 374V.

[Alibro]

Its strange but charger works with 100ms and inverter works with 100ms, but heater seem to work with less than 30ms. I seem to remember AC compressor needs 30ms as well. Due to easier application i simply setup a single timer for one 0x285 telegram at 10ms.

I'm using your Due sketch to control my charger and it was working perfectly for the first 40 odd charges but only recently started giving trouble. Maybe it's just faulty.
I'm also wondering now if the inverter cutting out has been caused by the dc to dc glitching while I'm driving. I'll do a bit more testing but I'm considering removing the Outlander charger, digging out my Leaf PDU and cutting it down so only the DC to Dc is left then installing it to keep the 12V happy.
I always charge in the garage anyway so I could rig up a HV connection to plug in the Outlander charger for ac charging only.
If this works then the Outlander will have been causing issues all along. If not then at least I'll have learned a few more things and can rule some things out.

[Alibro]

The voltage reading comes from ID 0x389 byte 0. From your data, when charging, this reached a maximum of 374V. Byte 0 data 0xBB = 187, 187 x 2 = 374V.

Thanks, that very helpful.

[arber333]

I'm using your Due sketch to control my charger and it was working perfectly for the first 40 odd charges but only recently started giving trouble. Maybe it's just faulty.

Hm... what mechanism do you think is there to disturb the inverter? Disturbances usually happen because faulty 0x285 heartbeat. It happend to me recently and i solved it by clearly determining timer for heartbeat.

What kind of control do you use with charge? You have a voltage control loop in place? See here a simple loop reading from 0x389

Code: Select all

0x389h 8bytes Charger status
  B0 = HV-Batteryvoltage    (hAB=342V -> 2V/bit)
  B1 = AC-Mains voltage     (hE2=226V -> 1V/bit)
  B2 = DC-charge current (1?) (h5B=9,1A -> 0,1A/bit
  B3 = Temperature 1    (starts at -40degC, +1degC/bit)
  B4 = Temperature 2    (starts at -40degC, +1degC/bit)
  B5 = Statusbyte     (CA while charging)
    bit0(LSB) = 
`   bit1    = Mains voltage present
    bit2      = 
    bit3      = Charging
    bit4      = Error (no CAN messages received)
    bit5      =
    bit6    = DC-DC converter request active
    bit7(MSB) = 1KHz pilot present    
  B6 = AC-Mains current     (h91=14,5A -> 0,1A/bit)
  B7 = DC-charge current (2?) (same as B3, sometimes differs by 0,1A)
0x38ah 8bytes Charger status

Code: Select all

void sendCANframeA() {
outframe.id = 0x286; // Set our transmission address ID
outframe.length = 8; // Data payload 8 bytes
outframe.extended = 0; // Extended addresses - 0=11-bit 1=29bit
outframe.rtr=1; //No request
outframe.data.bytes[0]=0x3C;
outframe.data.bytes[1]=0x0F; // 0F3C=3900, 0DDE=3550, 0,1V/bit
if(voltage < 195) { // if Charger senses less than 390V
outframe.data.bytes[2]=0x78; // 78=120 12A, 50=80 8A, 32=50 5A, 1E=30, 3A 14=20 2A at 0,1A/bit
}
else if(voltage < 196) { // if Charger senses less than 392V
outframe.data.bytes[2]=0x1E;
}
else {  //any other case
outframe.data.bytes[2]=0x00;
}
outframe.data.bytes[3]=0x37; // why 37?
outframe.data.bytes[4]=0x00;
outframe.data.bytes[5]=0x00;
outframe.data.bytes[6]=0x0A;
outframe.data.bytes[7]=0x00;
if(debug) {printFrame(&outframe,1); } //If the debug variable is set, show our transmitted frame
if(myVars.CANport==0) Can0.sendFrame(outframe); //Mail it
else Can1.sendFrame(outframe);
}

[Alibro]

Hm... what mechanism do you think is there to disturb the inverter? Disturbances usually happen because faulty 0x285 heartbeat. It happend to me recently and i solved it by clearly determining timer for heartbeat.

What kind of control do you use with charge? You have a voltage control loop in place? See here a simple loop reading from 0x389

Code: Select all

0x389h 8bytes Charger status
  B0 = HV-Batteryvoltage    (hAB=342V -> 2V/bit)
  B1 = AC-Mains voltage     (hE2=226V -> 1V/bit)
  B2 = DC-charge current (1?) (h5B=9,1A -> 0,1A/bit
  B3 = Temperature 1    (starts at -40degC, +1degC/bit)
  B4 = Temperature 2    (starts at -40degC, +1degC/bit)
  B5 = Statusbyte     (CA while charging)
    bit0(LSB) = 
`   bit1    = Mains voltage present
    bit2      = 
    bit3      = Charging
    bit4      = Error (no CAN messages received)
    bit5      =
    bit6    = DC-DC converter request active
    bit7(MSB) = 1KHz pilot present    
  B6 = AC-Mains current     (h91=14,5A -> 0,1A/bit)
  B7 = DC-charge current (2?) (same as B3, sometimes differs by 0,1A)
0x38ah 8bytes Charger status

Code: Select all

void sendCANframeA() {
outframe.id = 0x286; // Set our transmission address ID
outframe.length = 8; // Data payload 8 bytes
outframe.extended = 0; // Extended addresses - 0=11-bit 1=29bit
outframe.rtr=1; //No request
outframe.data.bytes[0]=0x3C;
outframe.data.bytes[1]=0x0F; // 0F3C=3900, 0DDE=3550, 0,1V/bit
if(voltage < 195) { // if Charger senses less than 390V
outframe.data.bytes[2]=0x78; // 78=120 12A, 50=80 8A, 32=50 5A, 1E=30, 3A 14=20 2A at 0,1A/bit
}
else if(voltage < 196) { // if Charger senses less than 392V
outframe.data.bytes[2]=0x1E;
}
else {  //any other case
outframe.data.bytes[2]=0x00;
}
outframe.data.bytes[3]=0x37; // why 37?
outframe.data.bytes[4]=0x00;
outframe.data.bytes[5]=0x00;
outframe.data.bytes[6]=0x0A;
outframe.data.bytes[7]=0x00;
if(debug) {printFrame(&outframe,1); } //If the debug variable is set, show our transmitted frame
if(myVars.CANport==0) Can0.sendFrame(outframe); //Mail it
else Can1.sendFrame(outframe);
}

At the very moment the charging stopped the sound from the coolant pump briefly changed. This would indicate something happened to the dc to dc converter which I didn't think was effected by CAN so either the contactors opened for some reason (I didn't hear them but this is still possible) or the charger glitched. I need to figure out which it is.

Automating turn off of the charger is on the to do list but I've been trying to sort a cutting out issue while driving. The charger now cutting out is a new issue but may well be related. It would be nice if it is as I can replicate this issue in my garage.

[arber333]

At the very moment the charging stopped the sound from the coolant pump briefly changed.

Is it possible that EVSE somehow disconnected PP signal? That would cause the change you speak of. Or maybe overheating....
There is B5 = Statusbyte within 0x389 which can tell you the status for charger and there is one for DCDC in 0x377 B7 = Statusbyte