Mitsubishi outlander charger and DC:DC

Mitsubishi hybrid drive unit hacking
User avatar
FJ3422
Posts: 113
Joined: Fri Jul 10, 2020 9:55 am
Location: Netherlands
Been thanked: 1 time

Re: Mitsubishi outlander charger and DC:DC

Post by FJ3422 »

FJ3422 wrote: Sat Aug 08, 2020 11:37 am Just got my Outlander DC-DC & Charger fully working, thanks for the information in this thread !

A tone-generator app & headphone-out + small amplifier gave me the possibility to generate the various 1KHz PWM-width's for simulating the EVSE.

However, it would be nice to be able to set a maximum charge voltage. Does anyone know how to activate the CV mode ? Or does someone have a (long) CAN-log of the last part of a charging process ? (CC to CV and following shutoff)
Figured out some more scalings then were already mentioned earlier:

Code: Select all

0x377h 8bytes DC-DC converter status
	B0+B1 = 12V Battery voltage	(h04DC=12,45V -> 0,01V/bit)
	B2+B3 = 12V Supply current	(H53=8,3A -> 0,1A/bit)
	B4 = 	Temperature 1		(starts at -40degC, +1degC/bit)
	B5 = 	Temperature 2		(starts at -40degC, +1degC/bit)
	B6 = 	Temperature 3		(starts at -40degC, +1degC/bit)
	B7 = 	Statusbyte 		(h20=standby, h21=error, h22=in operation)
		bit0(LSB) = Error 
`		bit1	  = In Operation
		bit3      =
		bit4      =
		bit5      = Ready
		bit6	  = 
		bit7(MSB) = 
	
	
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
	B0 = Temperature		Raises together with other temperatures, but with +5deg offset
	B1 = Temperature		(starts at -40degC, +1degC/bit)
	B2 = DC-bus voltage 		(9B=310V -> 2V/bit)
	B3 = PWM-signal from EVSE 	(h19=25% -> 1%/bit)
	B4 = Status
		bit0(LSB) = 
`		bit1	  =
		bit2	  = Wait for mains voltage 
		bit3      = Ready for charging
		bit4      =
		bit5      =
		bit6	  = 
		bit7(MSB) = 
	B5 = 00
	B6 = 00
	B7 = 00


SEND
0x285	B3 	= hB6 enables charging, other bytes can stay 00

0x286	B1+B2 	= voltage setpoint 		(0E74=370.0V, 0,1V/bit) -> How to enable CV mode ??
	B3 	= current setpoint DC-side 	(78=12A -> 0,1A/bit)
	
Dilbert
Posts: 410
Joined: Mon Aug 12, 2019 7:21 pm
Location: Dublin, Ireland
Been thanked: 4 times

Re: Mitsubishi outlander charger and DC:DC

Post by Dilbert »

FJ3422 wrote: Sat Aug 08, 2020 11:37 am Just got my Outlander DC-DC & Charger fully working, thanks for the information in this thread !

A tone-generator app & headphone-out + small amplifier gave me the possibility to generate the various 1KHz PWM-width's for simulating the EVSE.

However, it would be nice to be able to set a maximum charge voltage. Does anyone know how to activate the CV mode ? Or does someone have a (long) CAN-log of the last part of a charging process ? (CC to CV and following shutoff)
Yes it would be nice to get cv mode working on the charger, I will try capture a log if I turn on the charger when the battery is almost full, it should go into constant voltage mode (and possibly balance the pack)...
PatrykS
Posts: 38
Joined: Mon Jan 06, 2020 11:22 pm
Been thanked: 1 time

Re: Mitsubishi outlander charger and DC:DC

Post by PatrykS »

arber333 wrote: Sat Apr 25, 2020 2:46 pm Also my friend just got Mitsubishi Outlander pack 13kWh 96S
All Outlander packs (12 and 13kWh) are 80S, not 96S.
arber333
Posts: 3241
Joined: Mon Dec 24, 2018 1:37 pm
Location: Slovenia
Has thanked: 74 times
Been thanked: 223 times
Contact:

Re: Mitsubishi outlander charger and DC:DC

Post by arber333 »

PatrykS wrote: Thu Sep 10, 2020 11:22 pm All Outlander packs (12 and 13kWh) are 80S, not 96S.
True, he got additional 2 modules and so his capacity must be larger than 13kW.
arber333
Posts: 3241
Joined: Mon Dec 24, 2018 1:37 pm
Location: Slovenia
Has thanked: 74 times
Been thanked: 223 times
Contact:

Re: Mitsubishi outlander charger and DC:DC

Post by arber333 »

Dilbert wrote: Tue Aug 11, 2020 7:52 am Yes it would be nice to get cv mode working on the charger, I will try capture a log if I turn on the charger when the battery is almost full, it should go into constant voltage mode (and possibly balance the pack)...
Maybe you could just wait for the charger to report your CV on address 0x389 B1? uC would just wait for 0x389 to report such voltage and it can order charger to reduce amps to 1A and then hold there untill 0x389 reports absolute limit.
I would put absolute limit maybe 2V higher from CV limit.
arber333
Posts: 3241
Joined: Mon Dec 24, 2018 1:37 pm
Location: Slovenia
Has thanked: 74 times
Been thanked: 223 times
Contact:

Re: Mitsubishi outlander charger and DC:DC

Post by arber333 »

Dilbert wrote: Sat Apr 11, 2020 6:12 pm I've successfully got the Outlander charger to charge the prius battery :D

0x285 Frame Sent to the charger, to control EVSE
Byte 2 = 0xb6 pulls in the EVSE

0x286 Frame Sent to charger to control charging:
Byte 0-1 = Voltage setpoint (seems to be ignored, but i'm wondering is there a constant voltage mode which uses this)
Byte 2 = Current in amps x 10
Can you tell me which CAN interval you used for 0x285 message and which for 0x286?
I am preparing a DUE brain to work with charger and activate DCDC and pull EVSE relay in.
I think of using 100ms fast pinging for 0x285 and 800ms for 0x286. Would that be ok?

tnx
Dilbert
Posts: 410
Joined: Mon Aug 12, 2019 7:21 pm
Location: Dublin, Ireland
Been thanked: 4 times

Re: Mitsubishi outlander charger and DC:DC

Post by Dilbert »

I posted my test application a few pages back, this was run on the due. I was sending them at 50ms, but it doesn't seem too picky.
seanyt
Posts: 22
Joined: Tue Oct 29, 2019 9:45 pm
Location: Ireland/Netherlands
Has thanked: 2 times
Been thanked: 3 times

Re: Mitsubishi outlander charger and DC:DC

Post by seanyt »

Dilbert wrote: Tue Aug 11, 2020 7:52 am
FJ3422 wrote: Sat Aug 08, 2020 11:37 am Just got my Outlander DC-DC & Charger fully working, thanks for the information in this thread !

A tone-generator app & headphone-out + small amplifier gave me the possibility to generate the various 1KHz PWM-width's for simulating the EVSE.

However, it would be nice to be able to set a maximum charge voltage. Does anyone know how to activate the CV mode ? Or does someone have a (long) CAN-log of the last part of a charging process ? (CC to CV and following shutoff)
Yes it would be nice to get cv mode working on the charger, I will try capture a log if I turn on the charger when the battery is almost full, it should go into constant voltage mode (and possibly balance the pack)...
I was also curious about this CV/CV mode change and i have been informed that the PHEV ecu monitors voltage and when it gets to the suitable voltage it regulates the current to maintain this voltage setpoint. So it essentially takes over the CV control.
So whatever microcontroller you use will need to keep this in mind and have a watchdog system so you don't exceed your peak voltage.

Much like the PHEV ecu sets the torque limit/rpm limit and power limit and the inverter is the dumb device.
arber333
Posts: 3241
Joined: Mon Dec 24, 2018 1:37 pm
Location: Slovenia
Has thanked: 74 times
Been thanked: 223 times
Contact:

Re: Mitsubishi outlander charger and DC:DC

Post by arber333 »

Dilbert wrote: Tue Oct 20, 2020 7:23 pm I posted my test application a few pages back, this was run on the due. I was sending them at 50ms, but it doesn't seem too picky.
Ok thanks. I will first adapt EVTV CAN code. It seems this will run without a problem.

A
Dilbert
Posts: 410
Joined: Mon Aug 12, 2019 7:21 pm
Location: Dublin, Ireland
Been thanked: 4 times

Re: Mitsubishi outlander charger and DC:DC

Post by Dilbert »

seanyt wrote: Wed Oct 21, 2020 8:14 am
Dilbert wrote: Tue Aug 11, 2020 7:52 am
FJ3422 wrote: Sat Aug 08, 2020 11:37 am Just got my Outlander DC-DC & Charger fully working, thanks for the information in this thread !

A tone-generator app & headphone-out + small amplifier gave me the possibility to generate the various 1KHz PWM-width's for simulating the EVSE.

However, it would be nice to be able to set a maximum charge voltage. Does anyone know how to activate the CV mode ? Or does someone have a (long) CAN-log of the last part of a charging process ? (CC to CV and following shutoff)
Yes it would be nice to get cv mode working on the charger, I will try capture a log if I turn on the charger when the battery is almost full, it should go into constant voltage mode (and possibly balance the pack)...
I was also curious about this CV/CV mode change and i have been informed that the PHEV ecu monitors voltage and when it gets to the suitable voltage it regulates the current to maintain this voltage setpoint. So it essentially takes over the CV control.
So whatever microcontroller you use will need to keep this in mind and have a watchdog system so you don't exceed your peak voltage.

Much like the PHEV ecu sets the torque limit/rpm limit and power limit and the inverter is the dumb device.
Yep that's pretty much it.

I still find it hard to imagine they didn't include a CV mode in the charger, especially for balancing the pack, where CV mode would be useful. So far we have figured out CC mode.

I need setup to log the outlander again and start a charge cycle with a full battery and see if we can get CV mode.
arber333
Posts: 3241
Joined: Mon Dec 24, 2018 1:37 pm
Location: Slovenia
Has thanked: 74 times
Been thanked: 223 times
Contact:

Re: Mitsubishi outlander charger and DC:DC

Post by arber333 »

Dilbert wrote: Tue Oct 20, 2020 7:23 pm I posted my test application a few pages back, this was run on the due. I was sending them at 50ms, but it doesn't seem too picky.
OK... i made a Arduino DUE setup with CAN msg and resistors and relays. I did made some findings.
First of all DCDC works with just adding 12V to pin 4 of the connector. Works without any issues.
Next, charger caused some problems. I couldnt get the DUE to pull the EVSE CP signal in. I have EVSE in my house.
It became clear that my DUE circuit starts to pull CP signal in but couldnt grab the relay because CP signal behind diode would drop below 0.8V and DUE would release relay. When i measured the signal it seemed as if i got only 1/2 signal strength on the DUE side. As charger uses CP line to regulate its output i speculated it must have its own circuit there. So i tried to connect CP signal only to the charger this time. And this time it grabbed EVSE and pulled it in! I also commanded second CAN msg and charger started with 7.5A at 388Vdc.
I didnt have any coolant in the pipes so i left it there for 20mins only...
Also my BMS signal works and commands 0A CAN msg.

Now i need to make a simple sense circuit so DUE would find CP signal and engage CAN signal. Anyone has any ideas?
Dilbert
Posts: 410
Joined: Mon Aug 12, 2019 7:21 pm
Location: Dublin, Ireland
Been thanked: 4 times

Re: Mitsubishi outlander charger and DC:DC

Post by Dilbert »

Yes the pilot goes straight to the charger. On the proximity signal from the charge handle I just used a pull up resistor and an adc on the due. If the button in the handle is pressed, I just command the charger to 0A, I think there is a can signal which will drop out the charge point (obviously alters the pilot resistance ).
arber333
Posts: 3241
Joined: Mon Dec 24, 2018 1:37 pm
Location: Slovenia
Has thanked: 74 times
Been thanked: 223 times
Contact:

Re: Mitsubishi outlander charger and DC:DC

Post by arber333 »

Dilbert wrote: Sun Oct 25, 2020 9:55 pm Yes the pilot goes straight to the charger. On the proximity signal from the charge handle I just used a pull up resistor and an adc on the due. If the button in the handle is pressed, I just command the charger to 0A, I think there is a can signal which will drop out the charge point (obviously alters the pilot resistance ).
What do you mean by "proximity" signal? Do you mean PP signal or CP? Inside my cable i have PP connected by 220R resistor directly to PE (gnd). This will signal to EVSE cable is 32A capable. CP on the other hand goes straight to charger pin 9 now. I can split it and connect to DUE with a diode and resistor divider input, but it wont work because it will create different resistance for the charger input. Charger wont pull the EVSE in. I tried this allready and nothing happens. Also DUE is 3V3 device so i need to be carefull what voltage i show it.
arber333
Posts: 3241
Joined: Mon Dec 24, 2018 1:37 pm
Location: Slovenia
Has thanked: 74 times
Been thanked: 223 times
Contact:

Re: Mitsubishi outlander charger and DC:DC

Post by arber333 »

Hi

I am out of my depth here... i have an option to wire CP signal to Outlander charger and that is all well and good. But i want also to sense this CP signal with my DUE so i can use it to disable the traction on my car and to run my coolant pump.
I take the CP signal and put it through diode to have a clean 12V PWM. Then i use resistor pair to pull thze EVSE in. This is also inside Outlander charger.
I tried to put an optocoupler and a darlington transistor to tthe signal and it just turns my input signal ON even though there is no CP signal on the line yet.

Does anyone know how i would use opto as a repeater to copy the signal after diode and resistors? Then i could route my signal to charger and operate my equipment and inhibit driving...

TNX
Dilbert
Posts: 410
Joined: Mon Aug 12, 2019 7:21 pm
Location: Dublin, Ireland
Been thanked: 4 times

Re: Mitsubishi outlander charger and DC:DC

Post by Dilbert »

You are manually forcing the evse to pull in by loading the control pilot, you don't need to do that.

Yes you can use the PP signal to determine that the charge handle is connected. This PP signal is a pull down resistor in the handle, you need to provide a pull up to say 3v3 on your side.

Once the due detects the PP signal, it can then decide when to start the charge cycle. The due can then turn on the coolant pump and send the CAN message to instruct the charger to pull in the evse charger point.

I done similar to your self trying to get the evse to pull in, tried everything. I could get the evse to pull in based on attaching a resistor to ground but the charger still wouldn't charge. Once you send the correct CAN message the charger will pull in the evse.
arber333
Posts: 3241
Joined: Mon Dec 24, 2018 1:37 pm
Location: Slovenia
Has thanked: 74 times
Been thanked: 223 times
Contact:

Re: Mitsubishi outlander charger and DC:DC

Post by arber333 »

Dilbert wrote: Wed Oct 28, 2020 7:52 pm You are manually forcing the evse to pull in by loading the control pilot, you don't need to do that.

Yes you can use the PP signal to determine that the charge handle is connected. This PP signal is a pull down resistor in the handle, you need to provide a pull up to say 3v3 on your side.

Once the due detects the PP signal, it can then decide when to start the charge cycle. The due can then turn on the coolant pump and send the CAN message to instruct the charger to pull in the evse charger point.

I done similar to your self trying to get the evse to pull in, tried everything. I could get the evse to pull in based on attaching a resistor to ground but the charger still wouldn't charge. Once you send the correct CAN message the charger will pull in the evse.
Tnx, i went and looked at the EVSE wiring schematic again and i see what you mean. True PP signal will pull on a DUE pin pullup and signal my relay to close. What i didnt count on was my cable. I have 6 wires inside. 5x 6mm2 power cables and 1x 0,5mm2 CP wire. There is no PP wire because this is connected with 220R directly to PE inside one of the handles. DOH! I will need to get a propper cable to test this with power. But at least DUE works without problems.

A
Dilbert
Posts: 410
Joined: Mon Aug 12, 2019 7:21 pm
Location: Dublin, Ireland
Been thanked: 4 times

Re: Mitsubishi outlander charger and DC:DC

Post by Dilbert »

Yes the PP wire in the diagrams, goes from the charge port to the phev ecu, so it can detect when the evse is connected.

I tested mine with a granny cable as the evse and the charge port from an outlander can be purchased for €50 on ebay.

One idea I had was to run the PP wire via a simple timer, which could then be used to control the charging cycle for night rate electricity.
arber333
Posts: 3241
Joined: Mon Dec 24, 2018 1:37 pm
Location: Slovenia
Has thanked: 74 times
Been thanked: 223 times
Contact:

Re: Mitsubishi outlander charger and DC:DC

Post by arber333 »

Dilbert wrote: Wed Oct 28, 2020 9:58 pm Yes the PP wire in the diagrams, goes from the charge port to the phev ecu, so it can detect when the evse is connected.

I tested mine with a granny cable as the evse and the charge port from an outlander can be purchased for €50 on ebay.

One idea I had was to run the PP wire via a simple timer, which could then be used to control the charging cycle for night rate electricity.
I was blind the whole time! Well now i see... and i just pass through PP signal to my DUE and it works!
There is a snag however, most of cheaper cables only use 6 wire cables. So you need to put 220R in both handles so that car (DUE) will sense the pulldown and start CAN telegrams.

I tried various CAN speeds and i must say with the first msg ID 285 sent at 500ms charger does not activate. I needed to send it at faster speed. I used 100ms and it works.
Then charger is perfectly happy with ID 286 at 800ms.
I use BMS signal to replace 0A inside byte 2. Very crude version of CV control...

I could CAN read the byte0 of ID 389 report and if pack voltage would go higher than 390Vdc (value C3) command 1A untill pack reaches 392V (value C4), then 0A.
arber333
Posts: 3241
Joined: Mon Dec 24, 2018 1:37 pm
Location: Slovenia
Has thanked: 74 times
Been thanked: 223 times
Contact:

Re: Mitsubishi outlander charger and DC:DC

Post by arber333 »

:twisted:
User avatar
bobby_come_lately
Posts: 465
Joined: Sun May 03, 2020 5:39 am
Location: Manchester, UK
Has thanked: 33 times
Been thanked: 36 times
Contact:

Re: Mitsubishi outlander charger and DC:DC

Post by bobby_come_lately »

Can I check which model number people are working with? Looking through eBay for a charger and they all look quite different, from this one:

https://www.ebay.co.uk/itm/2018-MITSUBI ... Sw7O9e6z-N
Model number TW-90?

To this one: https://www.ebay.co.uk/itm/Mitsubishi-O ... Sw4Ste4~aK

To this one, which I think might actually be an inverter?
https://www.ebay.co.uk/itm/OUTLANDER-II ... Swx-9fFWBf

Also, if anyone has dimensions on theirs, it would be very welcome.
Dilbert
Posts: 410
Joined: Mon Aug 12, 2019 7:21 pm
Location: Dublin, Ireland
Been thanked: 4 times

Re: Mitsubishi outlander charger and DC:DC

Post by Dilbert »

Yep top link is from a newer phev. Middle link is the more common one, also used in a few other evs.

Try get the charge port too makes things a bit easier. Although if you don't have an evse / granny cable you should be able to just fake the control pilot.
User avatar
FJ3422
Posts: 113
Joined: Fri Jul 10, 2020 9:55 am
Location: Netherlands
Been thanked: 1 time

Re: Mitsubishi outlander charger and DC:DC

Post by FJ3422 »

This is the combined charger / DC-DC I tested. The TW90 seems to be an air-cooled DC-DC from the newer generation Outlander.

Be aware that the charger/DC-DC is mounted in the interior and is NOT waterproof. All the electrical connections are sealed, but there is no seal between the cover plates.
Dilbert
Posts: 410
Joined: Mon Aug 12, 2019 7:21 pm
Location: Dublin, Ireland
Been thanked: 4 times

Re: Mitsubishi outlander charger and DC:DC

Post by Dilbert »

Looking at the size of the charger it should easily fit in a spare wheel well of a car, inside the boot.
arber333
Posts: 3241
Joined: Mon Dec 24, 2018 1:37 pm
Location: Slovenia
Has thanked: 74 times
Been thanked: 223 times
Contact:

Re: Mitsubishi outlander charger and DC:DC

Post by arber333 »

Dilbert wrote: Sun Nov 01, 2020 10:06 am Looking at the size of the charger it should easily fit in a spare wheel well of a car, inside the boot.
True, but i cut my spare wheel well out a long time ago. I am using that place for a battery box now. This is why i intend to put 2x Outlander chargers each on its side under the rear seat. I managed to fit one charger there already. Has anyone made those chargers run in paralel?
I guess i will see what happens when those CAN reports colide on a single line :).

About the Outlander charger being non IP67... i found its quite well protected from outside. What i intend to do is put an alu panel under the seat so both chargers will be protected from spray and debris under the car.
User avatar
bobby_come_lately
Posts: 465
Joined: Sun May 03, 2020 5:39 am
Location: Manchester, UK
Has thanked: 33 times
Been thanked: 36 times
Contact:

Re: Mitsubishi outlander charger and DC:DC

Post by bobby_come_lately »

FJ3422 wrote: Sun Nov 01, 2020 10:03 am
This is the combined charger / DC-DC I tested. The TW90 seems to be an air-cooled DC-DC from the newer generation Outlander.

Be aware that the charger/DC-DC is mounted in the interior and is NOT waterproof. All the electrical connections are sealed, but there is no seal between the cover plates.
Thank you.

Would anyone be able to stick a tape measure on it? I have a space under my bonnet where it may fit bit it's going to be tight height-wise.
Post Reply