Identifying CanBus parameters

Mitsubishi Outlander PHEV Forum

Help Support Mitsubishi Outlander PHEV Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

Roy13

Member
Joined
Sep 23, 2017
Messages
13
Hi everyone,
I have made a GPS tracker for my PHEV that reports back to my raspberry pi with database (emoncms & HASS) at home.
I would love to have State Of Charge available so i can see it from my phone any where anytime similar to a Leaf.
Also i don't want to see the fake SOC meter on the dash but the real SOC, hopefully you all understand what i mean.
I have an MCP2515 connected to an Arduino UNO and have been taking various logs using Hyperterminal.
Please feel free to look over them and see if you can see any patterns, or better so if you can direct me to a program that could help me start a list on the page for future tinkerer's.
Currently ID 353 Data 0 & 1; show an incremental pattern counting up. Time between Data 0 incrementing can be measured giving a charge current.
ID 346 Data 2 appears to be dashboard SOC as i get FE when full and 06 when battery depleted and generator switching on.
Soon as i work out how to upload my logs i will.
I also have a Nissan leaf if anyone wants logs from it.
 
Finally, somebody who is into monitoring, rather than polling :lol: . How about 375, (D*256+E)/100-5

Gives you SOC with two decimal digits. No crap with 30 is empty or anything the like. True WYSIWYG. :mrgreen:

-- edit --

You may have a problem there, as 375 is only available on the secondary CANBUS. You will not see it on the primary CANBUS, which is available via the OBD port that sits under the steering wheel. If you really want it, you will have to tap into the secondary bus, for example at the connector the hooks up the rear motor ECU. That is how I did it.

BTW: To my best knowledge, 346 will give you this:

BARS SOC GAUGE: (C*17/100)
PCT EV DRIVING: H
POWER (METER) (A*256+B-10000)/100
 
I knew there would be someone out there playing with this stuff.
Well conveniently my GPS tracker is located in the boot with the battery right near the Rear motor controller.
I have thrown a picture in the link above please point out the connector and if you remember the colour of the high and low CanBus wires.
I'll post a log of the data for others to see in the same folder. Also i'll make note of when it's active so i can read it anytime or only if ignition on?

You must have spent many hours reading over the numbers looking for patterns....?

I'll have access to a written off PHEV soon so radio out and connect to stereo CanBus to see if i can get GPS directly from the car.
I don't hold much hope on this though too many number in a coordinate.

So what have you made up that's connected to these CanBus's?
 
Go here: http://mmc-manuals.ru/manuals/outlander_iii/online/Service_Manual_2013/2014/index_M1.htm

Select Workshop manual and the correct model year, in the top left corner.

Section 80 -CONFIGURATION DIAGRAMS => OVERALL CONFIGURATION DIAGRAM<PHEV> => OVERALL CONFIGURATION DIAGRAM <LOW VOLTAGE (LHD)> provides an overview of the CANBUS harness.

Section 80 -CONFIGURATION DIAGRAMS => FLOOR<PHEV> => FLOOR<LHD> details out the boot area.

Section 90 -CIRCUIT DIAGRAMS<PHEV> => CONTROLLER AREA NETWORK (CAN) => CONTROLLER AREA NETWORK (CAN)<LHD> provides details on various connectors and wire colours.

From the top of my head, you should be looking at Connector D211 in the third schema. It has also +12v and ground in case you need it. You should be able to find details about that in same source.
 
Roy13 said:
So what have you made up that's connected to these CanBus's?
I have three OBD Link WiFi adapters hooked up (yes, I am broke). One in ST MA mode to monitor front (primary) CANBUS, one in ST MA mode to monitor secondary (rear) CANBUS and one that I use for request/response communication. The latter is basically needed to get whatever I was not able to identify in the existing traffic. All three adapters are controlled from a single Java app on my phone (phone sets up Access Point, adapters run in Infrastructure Mode, so the app can talk to all three adapters simultaneously).

The app captures existing traffic (via first two adapters) and retrieves additional info via third adapter. All this info is buffered in the app. The app itself can be talked to via local port 35000 using the ELM327 protocol. This allows me to use Torque Pro (or any other standard OBD monitoring tool) to extract data from the app and display it on screen.

You will not believe how fast Torque Pro can be when it does not have to wait for request/response delays :mrgreen:
 
Roy13 said:
You must have spent many hours reading over the numbers looking for patterns....?
Uhhh .... a few, yes :lol:

I did three phases:

first phase: prepare to extract as much information as possible via normal request / response
second phase: prepare to capture as much raw data as possible using ST MA command
third phase: combine first two phases and map out data from two sources in graphs and look for similarities
 
So much useful information thankyou, I'm taking the family to the snow then ripping into the car as soon as I return.
I'm currently monitoring using Bluetooth obd front & rear torque, display and controller SOC (real) & cell 01 voltage so now I know why mine is painfully slow!
I would be keen to set up an old phone on the dash with exactly your setup but permanent!
If I put a display on my GPS tracker I have to run wires all the way up front and those cheap 16 X 2 displays aren't very flash looking.
 
Roy13 said:
I'm currently monitoring using Bluetooth obd front & rear torque, display and controller SOC (real) & cell 01 voltage so now I know why mine is painfully slow!
You mean, using request / response requests, rather than monitoring with STM or ATM commands? Then, switching to ATM or STM + buffering will make a huge difference. With me, Torque Pro refreshes several times per second, basically regardless of how many PIDs I display. Really funny.

(Of course, the buffered data that you obtain via request / response requests will be updated as often, but will most of the time be a little bit outdated)

Enjoy the snow. and the family.
 
Ok i connected to the rear canbus via plug 210 which appears to be the canbus link for 2 buses. Logged some data and included in the folder above.
Now to write some code on a Hologram dash that talks to a arduino nano via level translation that talks to a mcp2515 on the can bus. Some masks and filters to only get the can data i want and some maths on the hex data thanks to anko. Should be be in say a year....
Then I'm going to research these STM, ATM commands you speak of and set that up on an old android phone.
 
Nice, I guess you are getting there ;-)

FYI: for example ID 287 shows the central PHEV ECU requesting torque (in Nm) from the Generator ((E*256+F-10000)/10), the Front E-motor ((A*256+B-10000)/10) and the Rear E-motor ((C*256+D-10000)/10):

19056 ID: 287 DLC: 8 0x27 0x10 0x27 0x10 0x27 0x10 0x00 0x00
...
23884 ID: 287 DLC: 8 0x29 0x0B 0x28 0xF5 0x27 0x10 0x03 0x00

First is while stationary (foot on the brake or in N or in P) or while coasting in B0.
Second is 4WD in EV mode.
Both not charing or recuperating.

If only I knew what G and H were ..... ;-)
 
Ok up & running. I have posted a screen shot in the folder of my emoncms database that I log the data to.
First drive returning home flat, yet 36.9% remaining on arrival. If only I could manage the generator.
I'll have access to a statutory write off soon to tinker with if any one has some ideas they'd like me to try on it as not to put fault codes on their car!
And I only just noticed that last post while typing this. G & H I'll look over my logs some more for patterns too.
 
Roy13 said:
First drive returning home flat, yet 36.9% remaining on arrival. If only I could manage the generator.
Please explain. You took off with '0' battery and arrived with 36.9%? That should not happen ever, unless you had selected Charge mode.
 
Sorry no took off with full charge depleted to a point where petrol kicked in & dash displayed --km range. Once backing into the garage range up too 2km then took the 36.9% reading.
 
Yes it's due to the country roads. Just came off a 110km zone then it's a 3km hill climb to my house at 80kph. Both roads force the car into parallel hybrid mode. Not the most efficient mode & every day I have a small range left it won't let me use it due to being over 70kph. I've noticed if I'm flat going into the city the ice remains off well after --km range is displayed.
My first thoughts are to cut & intercept the can lines to the generator one mcp2515 receiving filtering out the generator on signal and the other sending the data to the generator. But obviously both chips being able read and send. Leaving generator torque output at 0 initially to see what happens.
 
To note once when towing the caravan up hill i got down to around 22% & got a warning on the dash about depleting the battery and limited power. Sorry that's non specific but I wasn't logging at the time & didn't expect it so no photo's.
 
Roy13 said:
Yes it's due to the country roads. Just came off a 110km zone then it's a 3km hill climb to my house at 80kph. Both roads force the car into parallel hybrid mode.
Speeds above 125-ish km/h will force your car into parallel hybrid mode. Speeds below that can be done in EV mode, in serial hybrid mode or (as long as you stay above 60-ish km/h) in parallel hybrid mode, depending on SOC and the amount of power needed for driving.
Roy13 said:
Not the most efficient mode
Why not?
Roy13 said:
... & every day I have a small range left it won't let me use it due to being over 70kph.
True. But this is only the range between 25.5% and 32.5%. If you have 36.9% SOC, you should be able to use it, even at speeds above 70 km/h.
Roy13 said:
I've noticed if I'm flat going into the city the ice remains off well after --km range is displayed.
Indeed. At city speeds, the low water mark for startingthe ICE is 25.5%. At higher speeds it is 30.5%. But, if I am not mistaken, only when the ICE is still cold.
Roy13 said:
My first thoughts are to cut & intercept the can lines to the generator one mcp2515 receiving filtering out the generator on signal and the other sending the data to the generator. But obviously both chips being able read and send. Leaving generator torque output at 0 initially to see what happens.
I think the ICE can run with Generator Torque Request being 0. IMHO, there will be one signal that controls starting of the engine and another controlling Generator Torque.
 
Ok I'll pay closer attention on Wednesday when I repeat the same trip. However I think your on to something as at almost full charge I registered 102.5%.
 
Well nearly a year ago i used this bit of Arduino magic to receive the battery capacity.
if(rxId == 0x375) {
result = ((rxBuf[3] * 256 + rxBuf[4]) / 100.0) - 5;
Serial.println();
Serial.print(result);

It has been working well but i have to reset the Hologram Dash every 2 weeks roughly due to some bugs in by project. Which i'm finally trying to iron out now.
As i have a written off vehicle to play with and a WiFi box(remote control) from an Aspire to play with.
I have enabled the remote control in the ECU with a flash OBD device called a Gscan. My radio now has timer charge and climate control delay buttons in settings(once i enabled & plugged remote control in).
Pin 13 Pink/Green wire on D-31 plug that goes to the charger if held low stops the car from charging! Brilliant for my Hologram Dash with a SIM card in it, as soon ill be able to control the charger from any where!
I thought advancing on that i might tap into the canbus 5 on that same remote control module plug. Pins 8/16.....however i cannot read the canbus..... using the exact same setup that reads canbus 6 plug D-210.

I also discovered if you remove Fuse 14 & 15 you can drive in full EV mode as the petrol motor cannot start. It will run for 30sec in it's prime stage but not start or try again. If like me you were thinking of putting a switch in these fuses once there back in you have to cycle the ignition for the petrol engine to start again. Sorry i know there are many question based around this but these were my preliminary findings.
 
Back
Top