MG Charger Hacking

[LRBen]

Had another play with the charger today.

Managed to cut down the messages required for charging to the following:
image.png

I have yet to move this to the zombie test module to confirm. But it's looking promising so far.

I also think I have found the current limit message in 0x29C, the last two bytes.

Turns out I was very close all those months ago. The last two bytes are a voltage setpoint times by 50. Currently have a zombie build that now ramps down to this setpoint according to the voltspnt value. Still need to test that it shuts off as it ramps very slowly to the voltspnt. Will try that next time I'm at the workshop. Might need to be adjusted by a slight amount to allow zombie to shut down the charge.
Otherwise it's looking good.

[Jack Bauer]

Nice work!

[AZD]

Alright Gents,
Help is needed and will be much appreciated.
I was hoping to replace a Brusa charger on my AZD frankenstein and to wire it up as per Damien's video, but the Molex pinout on the old connector appears to be completely different with some pins just blanked off as shown in the attached photo.
Need your second opinion on what the true story is considering MG ZS schematics are impossible to find.

It is understood Damien is out of reach for mere mortals, so the only source of information, whether free or well paid, is now gone.

[LRBen]

Alright Gents,
Help is needed and will be much appreciated.
I was hoping to replace a Brusa charger on my AZD frankenstein and to wire it up as per Damien's video, but the Molex pinout on the old connector appears to be completely different with some pins just blanked off as shown in the attached photo.
Need your second opinion on what the true story is considering MG ZS schematics are impossible to find.

It is understood Damien is out of reach for mere mortals, so the only source of information, whether free or well paid, is now gone.

MG tend to iterate component quite rapidly it seems. What is the part number for the charger you have?

[tom91]

MG ZS schematics are impossible to find.

Not true you can get them from MG for a small fee. Provide the partnumber and I am sure someone could figure out what year of ZS it is from and then we can dig through the diagrams to find the right one.

[LRBen]

In reverse engineering news, my zombie test module is getting very close to completion. Now that we have voltage cut off it will ramp down and stop charge via zombie. Been using it for the last few weeks without issue.
Today I tested a DC-DC fix for a bug where it would sometimes take quite a while for the DC-DC to kick in when going into run mode. Now it works in a couple of seconds every time.

Last thing to resolve is the V2L, although I still can't rule out physical wiring errors on my part there.

[m.art.y]

Now that we have voltage cut off it will ramp down and stop charge via zombie.

Hi, were you able to test what is the max voltage it will charge to? Will it charge 108s pack fully?

[Landsi]

AZD, i cannot open the images you added to your post.
But did you see the callout from LRBen about a slight difference in pinout for his charger?
Just in case that is the same as yours ... But as mentioned, a part number would be very good information along with your question!

I've got a Gen 2 charger from an MG5, part number EP2CCU6625A in my current project.

A slight difference in the pinout compared to the V2Lchargerpinout.png on the github, actually only figured it thanks to Damien's video. Just the pins of the hybrid can are difference. The charging wake up input also seems to have no effect. The pinout I have working is as follows:

A3: CP
A4: PT can H

B2: charge port temp +
B3: PP
B4: PT can L

D2: Charge port temp -
D3: 12v wake up

G2: hybrid Can H
G4: Gnd

H1: 12v
H2: hybrid Can L

[Bigpie]

I have corrected the missing images, please use the Full Editor button and upload directly

[LRBen]

Hi, were you able to test what is the max voltage it will charge to? Will it charge 108s pack fully?

Just have a 96s pack so can't test higher than that I'm afraid.

[AZD]

Yes I had to connect PP to the charge socket to get it to work.

Makes me wonder if Gen1 also requires a PP wire. A bit of grey area. According to the video it either needs a clean CP which I have or some CAN messages for DC to kick off.
Can't start mine on the bench for some reason. Not before Arduino Due arrives.

[AZD]

MG tend to iterate component quite rapidly it seems. What is the part number for the charger you have?

That issue is now sorted. The seller was under impression all Molex connectors are keyed and supplied the opne that fits from a different harness
The part number is ZS10BC6600A, the same exact charger seen on Damien's bench.

I am now a little stuck with trying to make the charger, erm, charge.
Once CP is grounded via 1kOhm, which is odd by itself, I do get AC and CP shown. Removing resistor send the CP voltage back from circa 3 to 6 volts and AC load dissapears. And that's how far I managed to get so far.

CAN messages

[manny]

Did you look at the info on the wiki https://openinverter.org/wiki/MG_ZS_Charger

I don't know what your setup is like. But the charger needs a CP signal and the 0x29C can message to work. And the other connections in the wiki.

Here is a DBC file to see the values sent by the charger. Just change the .txt to .dbc

[AZD]

Did you look at the info on the wiki https://openinverter.org/wiki/MG_ZS_Charger

I don't know what your setup is like. But the charger needs a CP signal and the 0x29C can message to work. And the other connections in the wiki.

Here is a DBC file to see the values sent by the charger. Just change the .txt to .dbc

Thank you for your input. Is it now accepted the top 1mm pin in AC Aptiv connector is another HV interlock? Damien is referring to that pin as a weird one, very likely some sort of PP wire.
My setup is exactly the same as Damien's.
I am not a CAN savvy to my great shame, and I was able to send some attached trace commands manually from Kangaroo soft no to avail.
I'm now trying to sort out a SavvyCan to try sending some logs to see what happens.

[manny]

The HV interlock (might also be a PP passthrough) on the AC connector is connected to the LV connector (pin 11). it's not important. I will fix the wiki.
You have the HV interlock on the DC connector shorted?

your test file look good.
AC max current = 32A
DC max current = 2A
DC max voltage = 462V

[AZD]

Got it, thank you. I will make that PP/HVIL redundant. My original wake up wire seems to be not needed too.
Yes, I have a jumper on DC connector. I am just trying desperately to see the load on my DC circuit made of two bulbs and a capacitor from my dead DMOC before bolt the chrger to the car.

I just have unpacked a Canable dongle that works with SavvyCan and even managed to get on the bus somehow. But this is where my knowledge ends. Will need a few month to figure out how it works.
I probably need someone on paid basis to get on my PC via remote desktop to run all CAN voodoo stuff

[manny]

Not sure it will work without the charger seeing voltage on the DC connector

[AZD]

I was hoping some CAN spoofing can help with that, although I am under impression from Damien's description this is very easy going charger that doesn't ask for too much to make it work

[AZD]

Just wondering if it's worth playing some other files but the 29C?

https://github.com/damienmaguire/MG-EV- ... c_can3.csv

[AZD]

The HV interlock (might also be a PP passthrough) on the AC connector is connected to the LV connector (pin 11). it's not important. I will fix the wiki.
You have the HV interlock on the DC connector shorted?

your test file look good.
AC max current = 32A
DC max current = 2A
DC max voltage = 462V

Well, that would be too good to be true.
A real picture looks like this after feeding Damien's 29C CSV to CAN.
Perhaps you were not wrong saying it wants some DC unlike my old dead Brusa charger

Savvy traces