Help for peugeot Ion 2012 that doesn't work.

[MickeyS70]

At the moment the car not charge because the charge sotps in few minutes without energy transfer (100W in the last charge by public charger) . I want to try if possible to use an EVSE settled with 6 Amps of charging current,

The car won't charge if cell voltages are fluctuating as indicated..

but i must move the car for reach a 230V power outlet...

You can tow it with all 4 wheels on the ground, key to position 'ACC' and gear in anything but P (ignore the waring in the manual)

yesturday i make some other test and had captured some data using CarScanner and CanIon.
Canion show all cells in a single screen but has a refresh time too much slow for see what happen when the error triggers

The data you see has already been processed by the BMU and is re-transmitted on the OBD bus. The data transfer on the internal BAT CAN between the BMU & CMU boards is faster, but not accessible without a CAN bridge.

.
But I noticed that sometime, a single refresh cycle shows one or two cell with 2.1V (in yellow) an then the value returns immediatly at 3.8V.

That is a typical CMU fault, a reported 2.1V actually means no data received.

And sometime the level gauge on the dashboard of the car has level bars flashing, they sometimes disappears and reappear again. I was make a video , but in the time of record the gauge was regular....

Typical symptom of a CMU issue.

Now i want to make some graphs with data captured and chek if there are some useful information.

The most useful information is which cells voltages are fluctuating.

But the new 12V battery is coming and think to have in my hand only tomorrow, so i made last test with the poor battery with in parallel a good battery with cables. So not in an optimal environment.

Not ideal, if the car isn't in READY, all power is supplied by the 12V battery only, (low voltage can affect CMU boards).

 

[Franco Vecchi]

So maybe some problems on one CMU? is it already happened? I ordered also hardware for a canbridge based on arduino due, when it will be functional i will try to read can message on the BMU/CMU bus.
Actually i do not had an adeguate space for works inside battery, so i want make a good diagnostics first of dismount it.. ;-)

"That is a typical CMU fault, a reported 2.1V actually means no data received." It means that error P1A52 maybe caused by this random erratic value?

 

[MickeyS70]

So maybe some problems on one CMU?

Yes, most likely one (two) channels on CMU01

is it already appened?

Yes, this is a known issue, e.g. @alvinseven has a large thread about this topic
https://myimiev.com/threads/battery-pack-change-plans.5525/page-9

I ordered also hardware for a canbridge based on arduino due, when it will be functional i will try to read can message on the BMU/CMU bus.
Actually i do not had an adeguate space for works inside battery, so i want make a good diagnostics ;-)

That’s the main advantage of a CAN bridge, no need to remove the HV battery.

 

[Franco Vecchi]

Sorry for the many messages....
I see something maay be intersting in data captured yesturday:
in the first we have erratic temprature, 10 degrees is not true (2 -3-4 degrees is the real temprature)
in the second we have erratic voltage of some cells (2.1V and then 3.8 again) with no current (0A)
and see also the cell voltages on the current peak
in the third we can see the total voltage of battery vs. ca peak of current. From about 330 to 220V on a current peak of about 17A

All of this indicate some erratic voltage cell values (the 2.1 random with zero current) and that all cells are poor (also max cell voltage is lowering a lot on a current peak of < 20A).
I'm right?

 

[MickeyS70]

Sorry for the many messages....
I see something maay be intersting in data captured yesturday:
in the first we have erratic temprature, 10 degrees is not true (2 -3-4 degrees is the real temprature)

Strange, but this won’t stop the car from driving or charging, this ‘hotspot’ is reported by CMU04 and could also be faulty, keep an eye on it to see if these values remain constant @10deg or are always 6C higher than the rest.

I the second we have erratic voltage of some cells (2.1V and then 3.8 again) with no current (0A)

That’s a known issue (CMU problem) although I cannot see that on your video

and see also the cell voltages on the current peak
in the third we can see the total voltage of battery vs. ca peak of current. From about 330 to 220V on a current peak of about 17A
All of this indicate some erratic voltage cell values (the 2.1 random with zero current) and that all cells are poor (also max cell voltage is lowering a lot on a current peak of < 20A).
I'm right?

Yes, @ a mere 17A load min/max cell voltage drops to about 2.5V, which ties in with the total voltage of 220V DC i.e. all 88 cells seem affected.

A CAN bridge can take care of the second issue but will be no good if the cells themselves are weak, but I have never seen behaviour like that, so I’m not convinced that this is actually real.

I would still go ahead and install the bridge which should then allow to charge the battery fully and run more tests.

 

[Franco]

Strange, but this won’t stop the car from driving or charging, this ‘hotspot’ is reported by CMU04 and could also be faulty, keep an eye on it to see if these values remain constant @10deg or are always 6C higher than the rest.

That’s a known issue (CMU problem) although I cannot see that on your video

Yes, @ a mere 17A load min/max cell voltage drops to about 2.5V, which ties in with the total voltage of 220V DC i.e. all 88 cells seem affected.

A CAN bridge can take care of the second issue but will be no good if the cells themselves are weak, but I have never seen behaviour like that, so I’m not convinced that this is actually real.

I would still go ahead and install the bridge which should then allow to charge the battery fully and run more tests.

From experience a fluctuating value in the cell voltage is typical of a missing connection, sometimes even a defect in the CMU.

 

[Franco]

[moderator edit: below information relates to Peugeot Partner EV, not a triplet]

Here is the traffic on the CAN line between the batteries and the master BMS (under the driver's seat).
The second digit on the addresses is that of the various VCUs, from 1 to A (I have a 80s set), then from the first line of each from b2 the temperature values +50, finally the third and fourth lines are for the voltage values (here is the sniffing from my original battery).
A question: does anyone know the meaning and origin of the data on 3C2 and 3C3?

 

[MickeyS70]

A question: does anyone know the meaning and origin of the data on 3C2 and 3C3?

Found this:

CMUs boards listen, from startup, for 0x3C3 frames only if these frames are sent regularly. If the transmission of these frames stops for a moment before resuming, the balancing requests are never considered again, until the next restart of the CMUs boards.

Further information:
https://www.diyelectriccar.com/threads/mitsubishi-miev-can-data-snooping.179577/page-2

 

[Franco]

the CAN messages screen is from my Peugeot Partner, the addresses and byte values correspond to those I found in the AEVA forum on Mitsubishi iMiev
The master BMS or BMU in my car is under the driver's seat, next to it there is also the EVECU the unit that controls the electric traction, therefore the direct command of the power relays inside the battery and other functions.
Both BMU and EVECU are Mitsubishi branded

 

[MickeyS70]

the CAN messages screen is from my Peugeot Partner, the addresses and byte values correspond to those I found in the AEVA forum on Mitsubishi iMiev

Mystery solved.

Once you look at the BAT CAN messages on your iON you will see, that while there are similarities, they are not identical.

See below: (second link in Post 3)
https://myimiev.com/threads/cmu-can-‘spoofing’.5505/post-48271

 

[Franco Vecchi]

Some little bit news: mouted a new 12V battery,
Now the monitoring is a little more stable.
Things are not changed, but made some more data logging using CarScanner.
I also measured the true battery voltage with a voltmeter, on the inverter contacts, and I saw that the true battery voltage value (338V) is the higher one of the two that the app shows, (battery total voltage) and it is also the one which varies less. "Target voltage battery" (about 320V) has huge variations and I don't understand where it comes from.
I also tried to recharge with a charger set at 6A: the charging made several attempts but always with little current, the charger indicated peaks of just over 1A then stopped, started again on its own and so on.
Only once it seemed to have started, but unfortunately my energy source was a 1000W inverter connected to the Leaf's service battery and it went into overload....
For now I don't have any power sources from the mains where I have the car unfortunately.

Another thing I noticed, which has always happened in the various test sessions I have done, is that at the beginning the machine remains in READY for several minutes, (if there is low current absorption), but as the time passes, becomes more and more sensitive and exits READY more and more often until it no longer goes into READY, even if the service battery is perfectly charged.
This was happend EVERY session of work on the car.
What could cause this strange behavior? something that heats up too much in a control unit such that after a maximum of half an hour the machine no longer goes into Ready mode?

Next thing I need to check, but I need someone to help me on the car, is to measure the voltage of the battery as the absorption increases so as to verify that the diagnostic indication is correct.

 

[MickeyS70]

"Target voltage battery" (about 320V) has huge variations and I don't understand where it comes from.

It’s a calculated voltage for balancing purposes only, it will fluctuate, if your reported cell voltages change due to faulty CMU data, best to ignore for now..

I also tried to recharge with a charger set at 6A: the charging made several attempts but always with little current, the charger indicated peaks of just over 1A then stopped, started again on its own and so on.

Same issue again, fluctuating cell voltages will stop the charging process, then recover and so on…

Only once it seemed to have started, but unfortunately my energy source was a 1000W inverter connected to the Leaf's service battery and it went into overload....

No surprise there, 220V x 6A will overload your 1kW inverter eventually.

What could cause this strange behavior? something that heats up too much in a control unit such that after a maximum of half an hour the machine no longer goes into Ready mode?

Do you get any fault codes/lights on the dash when the car no longer goes READY?

 

[Franco Vecchi]

It’s a calculated voltage for balancing purposes only, it will fluctuate, if your reported cell voltages change due to faulty CMU data, best to ignore for now..

Same issue again, fluctuating cell voltages will stop the charging process, then recover and so on…

ok

No surprise there, 220V x 6A will overload your 1kW inverter eventually.

yes, i'm not surprised, if i find a more powerful inverter i will try again

Do you get any fault codes/lights on the dash when the car no longer goes READY?

only the same P1A52 and occasionally the U1111 when the dasboard is off.

As soon as possible i want to check the true voltage and current on battery with a clamp Amps and a voltmeter ,applying a progressive load (i.e. by the car heather) and see if the big drop on voltage is real.
If it is real i suppose that the battery cells are very bad.

Also , if it is impossible to charge the car, i wish to put current into it by inverter contacts , by using a transformer and a rectifier, so to put into battery a small curret (1 to 3Amps) in DC and see what happens at cells voltage.
I know tha it is a bit dangerous! but i'm experienced on works with these voltages on industrial devices.
I must create a dedicated device having a minimum of safety.....!

 

[kiev]

Only once it seemed to have started, but unfortunately my energy source was a 1000W inverter connected to the Leaf's service battery and it went into overload....

Unfortunately without access to the Mains it will be nearly impossible to charge.

i estimate that you might get 5 minutes of use with a 1000W inverter connected to the 12V battery, after which the battery will be depleted.

 

[Franco Vecchi]

Unfortunately without access to the Mains it will be nearly impossible to charge.

i estimate that you might get 5 minutes of use with a 1000W inverter connected to the 12V battery, after which the battery will be depleted.

My inverter was connected to a 12V battery of my Nissan Leaf in ON state, so it is recharged continuosly at 1800W from traction battery... it is a V2L homemade
not a grater efficiciency but it works for hours....

 

[MickeyS70]

only the same P1A52 and occasionally the U1111 when the dasboard is off.

Ignore U1111, the P1A52 problem can be ‘fixed’ with a CAN bridge and the car should then be able to charge afterwards

Also , if it is impossible to charge the car, i wish to put current into it by inverter contacts , by using a transformer and a rectifier, so to put into battery a small curret (1 to 3Amps) in DC and see what happens at cells voltage.
I know tha it is a bit dangerous! but i'm experienced on works with these voltages on industrial devices.
I must create a dedicated device having a minimum of safety.....!

Let’s ignore the “safer” CAN bridge solution for a minute:

The car needs to be READY (i.e the HV contactors closed) for this to work, so you have a 30 min window max. It needs about 0.3kWh to run all systems, so charging will be slow, to say the least. It may even drop out of READY if it detects current flowing the ‘wrong way’ when stationary.

Hopefully you won’t have to go down this road, but if you do, the CHAdeMon DC circuit is far more suitable for this kind of ‘hack’ as the car is off and the quick charge relays are operated externally.

My inverter was connected to a 12V battery of my Nissan Leaf in ON state, so it is recharged continuosly at 1800W from traction battery... it is a V2L homemade

Your V2L is powered by the Leaf’s DC/DC, not the 12V aux, in ON stage

 

[Franco Vecchi]

Hopefully you won’t have to go down this road, but if you do, the CHAdeMon DC circuit is far more suitable for this kind of ‘hack’ as the car is off and the quick charge relays are operated externally.

Ah! great idea!
I was thinking that use chademo is very difficult, but under your input, i find this thread
https://myimiev.com/threads/12-kw-chademo-charging.2001/#post-35097
that explain how to do, and it is very doable!
I will try in this manner!