not charging my iON, cell or CMU or BMU problems

[czalbert]

Hi all,

I have a Peugeot iON - 2011 (66.000 km). I bought with 40.000km and I used 26.000 km without any problems. One day I could'nt charge my car, started the charging and flashed the ! icon on the dashboard and stopped the charge. My mechanic read out the OBDII and we got an error code in the 10th block F cell has low voltage. I checked the Volts and I saw every 2-3 seconds drops to under 2.5 V and goes back to the normal level and the charging is stopped. I can't charge the car. I got a new cell and we replaced it.

But now the problem is still. I used the canION app and I saw a some intresting data. The cell what we changed after charging in 3-6 seconds drop down suddenly the Volts number from 3,7 to 3,2 or sometimes 0,7 V and suddenly up to the normal Volts and after this repeat 3 times and the system block the charging. I attached the picture about the cells Volts under charging https://photos.app.goo.gl/yU6hRXbaEBlaJ6Ch1 and the other picture about the drops https://photos.app.goo.gl/08wmYf5UFC3qOQ2x2. I can’t charge. I tried start the car and under pressure also drop down the Volts in random on the number 74 cell.

And after I checked the Volts History and I found there sometimes the Volt up to 6,400 Volts and after goes back to 340 Volts. (picture about that: https://photos.app.goo.gl/hzSIrO3dZh3SzEKZ2) And the battery icon on the right is red and sometimes under charging change to full (green) and suddenly goes back to red. (actually charging is 14%)

First I thought this cell is wrong but we replaced it and still the problem. After the cell replacing the OBD error code is P1A5B
I think that is not cell problem, do you have any ideas? What is that?

Thank you.

 

[kiev]

It sounds like the cell that you changed was not bad. The problem is likely on the circuit board rather than the cell itself being bad. A single cell on a workbench can be tested by charging and discharging manually to determine it's capacity and whether or not it is defective. This could be done on the cell that was removed to verify that the cell is good or bad. A defective electrical component, or a bad solder joint, could cause the voltage readings to randomly and intermittently vary in the manner in which you described. A complete 8-cell module swap, or just the 2 circuit boards in an 8-cell module, might fix your problem. Or a re-flow of the solder joints using a hot-air blower on your current circuit board might fix it also.

Here is the Reference Info from the Service Manual on troubleshooting this code.

Code P1A5B = CMU10 battery cell low voltage abnormal

Trouble Diagnosis:

Check Conditions
This code appears after 3 seconds elapse after the electric motor switch is turned ON, or 3 seconds elapse after the main battery starts to be charged, AND
None of the following diagnosis codes occur at the same time.
P1AA8: Local CAN (for main battery) signal time-out
U1082: Local CAN (for main battery) bus off
[note: This just means that the CAN buss is working okay and is not the source of the issue.]

Judgement Criterion
The battery cell voltage is less than 2.3 V.

FAIL-SAFE AND BACKUP FUNCTION
None available

PROBABLE CAUSE
The module 10 (consisting of battery cell, CMU board and battery cell voltage sensor) in the main battery is failed.

DIAGNOSIS PROCEDURE
STEP 1. M.U.T.-III data list
Check the data list reference table items corresponding to the diagnosis code, i.e. check these items:
Item 102: CMU ID10 battery cell A voltage
Item 103: CMU ID10 battery cell B voltage
Item 104: CMU ID10 battery cell C voltage
Item 105: CMU ID10 battery cell D voltage
Item 106: CMU ID10 battery cell E voltage
Item 107: CMU ID10 battery cell F voltage
Item 108: CMU ID10 battery cell G voltage
Item 109: CMU ID10 battery cell H voltage

An OK data condition is a reading of 2.3 V or more.

Q. Is the check result normal?
If 2.3 V - 2.5V, then charge the main battery.
If 2.5 V or more, then it is an intermittent malfunction (Refer to GROUP 00).

 

[czalbert]

Dear kiev.

Thanks your response.

Yes maybe it is a sensor problem. But I don't understand If it is a sensor why only the problem occurs when the car charging or running, when I checked the data under ignition the problem didn't occur.

Now I got a tip, have to change two blocks. For example change the 10th blocks with the 11th blocks and if the problem moved to 11th, it has the problem with the CMU board.

Thanks your advice.

 

[kiev]

Swapping two modules is a great idea, it may be a lot of work, but maybe it doesn't take too long to open the pack after you have done it a time or two...

When the key in at ON but not to START or in READY mode, then the main contactors in the pack are not energized and no pack current can flow.

During charging and running (READY mode), then the contactors are energized and pack voltage is exposed to outside the pack and current can flow either in or out of the pack.

Cell "F" is in the section of a module with the simple circuit board without many components and connects with a little jumper cable to the main cell monitoring board.

When you do open the pack, use a voltmeter to measure the individual voltages of each cell in the two modules before you swap them. Write this down and compare it to the values that Canion reports.

If you find a low cell then it can be manually charged back up to equal the other cell using a small variable power supply. Set the voltage equal to the cell voltage and then connect to the cell, then slightly raise the voltage and current will flow. continue slowly raising the voltage until it is equal to the other cells and the current has diminished.

 

[czalbert]

Dear kiev.

We swapped the 2 blocks and the problem still on the original (#74 cell) so it's not sensor board problem. Do you have idea what is that? It's very intresting the Volt dropping and the error code still on the #74 cell and the 10th block after we changed the 10th block to 11th block.

Thank you.

 

[kiev]

Howdy albert,

Sorry I didn't think thru this module swapping clearly-- the cell numbers are determined by the BMU and then programmed and stored in an eprom chip on the CMU (cell monitoring) boards located on each module. The BMU doesn't care where in the pack the module is physically located, and the CMU will still respond over the CAN buss using the module 10 CAN ID. That's why swapping the modules had no effect.

If you are sure that the replacement cell is a good cell, then it is pointing a finger at the CMU board.
You will need to either repair the board, or replace it, or replace the entire module to solve this issue.

[edit: the CMU number of the board would need to be written into the eprom of a replacement board or module, unless it already has the same CMU number, no VIN coding required]

good luck, keep us posted with your progress,
kenny

ps did you measure the cell voltages with a meter while it was open? how did they compare?

 

[priusfan]

Bonjour,
The cell used by Albert for replacement was a good one. I sent it to him and I checked before...
I fully agree with kiev's comments.
I think also the trouble is in the CMU (BMS part is deficient for one cell).
Xavier

 

[czalbert]

Bonjour Xavier.

Yes your cell was great and I was very sad because we replaced the cell and the problem was still. I hoped that is a cell problem, the electronic problem is harder.

Tomorrow we will check the board and thanks for the comment for kiev.

I don't know what can I do if we won't able to fix the board. I think impossible found CMU board. I will contact with the Peugeot, because now the official answer, have to change the whole main battery, but why? For the cheap Volt sensor?

So now I have to find solution.

Thanks your helping.
Albert

 

[kiev]

i can troubleshoot and repair the board if you can't find someone nearby to do it.

 

[czalbert]

Dear kiev.

We checked the board yesterday and there is no problem in the analogue section. The signal is OK to the digital component. Can the digital part also be improved?

Thanks
Albert

 

[kiev]

By "digital component" do you mean the LTC6802 chip shown in this photo, or are you talking about the microcontroller or CAN buss chips? The LTC is a generic part and could be de-soldered and replaced with a new one.

Inspect all those little yellow chip capacitors for cracked solder joints or cracked body--they are prone to cracking and they short out when failed. Try heating with hot air, or cooling with cold spray, while monitoring to find a defective part.

 

[czalbert]

Dear kiev. Thanks your advices, I try find somebody who can replace the LTC and fix the board.

But now I have a new solution. I found a used board. Just I don't know, is it programming for 10th block. If don't how can I do that?

You wrote in the previous comment "replace the entire module to solve this issue. Replacing will require a visit to the dealer to re-program the cell numbers and the VIN into the pack." - (In Hungary I think the dealer can't do that. How should they do it? Do they work with OBD Peugeot software?)

Is it necessary if I change just the board? Or that have to do when change the entire module?


Thanks and sorry for the lot of stupid question.

Albert

 

[kiev]

The LTC6802 is a purpose made Lithium cell monitoring chip with analog voltage inputs and transistor drivers, but it communicates with digital data over a serial port to the big square (quad flat pack) microcontroller chip. Then the microcontroller communicates with digital data to the outside world (BMU box under the back seat) thru the CAN buss. So there is a defect somewhere between measuring the cell voltage and the reporting of that voltage to the BMU over the CAN buss.

The BMU is responsible to number the cells and modules--it sends commands which cause that identity information (the PID number) to be stored into an eprom memory of each module. This is an initialization routine that only has to be done once when the pack is built, but must be repeated if a module board (CMU) is swapped or replaced. The CMU microcontroller chip doesn't know which PID number it should respond with until it has been assigned by the BMU.

Tesla modules are on a daisy-chain serial line to a main BMS board that connects to the CAN buss. They use Texas Instrument chips that will determine module and cell numbering automatically without an external command, but for our cars we just don't know yet how it is done, although member coulomb has been reversing the code and has found out the numbering sequence.

But for sure that task can be done by a dealer service department which has the Mitsubishi Universal Tester, also know as the MUT. Peugeot may have a different name, but same functioning device. The MUT communicates data and sends commands over the CAN buss to all the electronic boxes in the car. There are clones of the MUT sold on ebay, but most of them don't have the programming feature--they can only read the data and trouble codes.

It may be that the BMU is smart enough to know that a cell module board has been replaced and can re-program it automatically, but i doubt it. Nobody knows for sure, but you could test this theory if you wanted to swap it out and give it a try. Please share your findings and results if you do swap boards or complete modules.

 

[sandange]

I am no expert in this field.
I have some information from the facebook I Miev en Francaise group that may be of relevance.

Part of a posting there by Gerald Fiola– in French
Savez vous que les 4 pieces principales mcu moteurs chargeur et contrôleur communiques ensembles via une programmations Mais que la piece a un numéro unique et qu'il est pas possible de la remplacer par une autre venant d'un autre miev !!! Vraiment dommage Vous devez donc payer pour une piece neuve si vous devez changer une piece des composantes principales facture salée pour un Mcu $8000 plus installation de la folie pour l'environnement . Vraiment décevant

Google translation------

Do you know that the 4 main parts
mcu , motor, charger, and controller
communicate together via a programming But that the piece has a unique number and
it is not possible to replace it with another coming from another miev !!!
Really a pity So you have to pay for a new piece if you have to change one piece of the main components
A salted bill for a Mcu $ 8000 plus installation
All of this is craziness for the environment.
Really disappointing.
--------------------------------------------------
Here is a video ( in French) of an attempt to program a salvaged MCU to replace a faulty unit on a 2012 I Miev ,
Testing done at Mitsubishi with their equipment and technician the speaker in the video!

Quick rough translation is -- unable to program vin number not recognized.

https://www.facebook.com/geradinepl...f_id=1516078747872731&notif_t=group_activity.

------ He Now Drives a Bolt--------

More Google translated text from Gerald - not perfect but you get the general translation

So before buying the BOlt I also checked with Simon André's Auto Electrical and all the parts can be reprogrammed several times on the Volt and Bolt models.
But not on the Miev models and all those gas also from Japan ,Korea.
So the goal here is to inform you so that you are informed before buying or before changing.

I also learned that changing the 4 units together is the only way to have used replacement parts in a miev ...

 

[kiev]

Merci beaucoup!

We have to figure out how to re-program the VIN to re-use salvage parts--but it is a tightly coupled system that will be difficult to crack...

 

[czalbert]

Or here is the hardware's way, if the software crack doesn't work. Replace the board but find the part which stored the VIN number and can communicate with the BMS and replace from the original board:

https://www.youtube.com/watch?v=g-buef3J50g

remondin Mitsubishi I-MiEVi ETACS juhtplokki eesmärgiga võtta auto originaalploki pealt maha vajalikud mikroskeemid ja panna need teise auto (Citroen C- zero) pealt kasutatuna ostetud plokile. Selgus, et ka tehase MUT3 arvutiga ei saa seda plokki teist korda üle kirjutada.

I translated with google: repairing the Mitsubishi I-MiEV ETACS control unit with the aim of removing the necessary microcircuits from the car's original block and placing them on the block purchased on the other car (Citroen C-zero). It turned out that the factory MUT3 computer can not overwrite this block for the second time.

 

[tinoale]

Hi all,

I think that I may have the same issue.

My wife took the car (fully charged) for a 35km ride, no heater, arrived with one bar below 50%. She was surprised by such a poor performance but found the car with 2 more bars when she took it back home few hours later.

On the next day, when she told me that I immediately thought something was fishy. The cas was fully charged again during the night. I pulled Canion and found one cell (cell #71) was randomly droping to a lower voltage and bounding back to the normal voltage.

https://www.youtube.com/watch?v=VFCCpODjg-s

This video was taken with the car started in "ready" mode.
As you can see the voltage to which it drops is not always the same. It doesn't seem to stay there either, juste a blink.

I can still charge the car. But I am afraid that if a component or solder is starting to fail, this is only going to get worse until the car can't be used anymore.

My understanding from this thread is that the best bet would be to swap one (or more) of the components on the CMU board WITHOUT changing the one chip that holds the VIN configuration.

@czalbert, have you found a solution yet ? have you tried to fix the CMU board by swaping some components ?

Physically I have no idea how a "module" looks like. Could someone point me some ressource on how to dismantle the battery down to the module ? I hope I won't have to use it but I'd like to be prepared...

Thanks a lot !

 

[kiev]

Here's a good video on replacement of a cell

http://myimiev.com/forum/viewtopic.php?p=23601#p23605

 

[tinoale]

Hi,

My issue has repeated during several days and I noticed this :
- in the morning, my home > work trip was using as much as 10 bars of the battery gauge, which was grossly too much considering the distance. Car was cold and battery was full at the start.
- on the evening, my return trip would use only 2 bars, which is grossly too low of usage. Car was in a garage (around 10°C). And battery not full.

I deducted from this behavior that the cell voltage was only playing tricks in the morning. That could be caused by 1/ low temp and/or 2/ a full battery.

That was during a very cold week we had here. Unfortunately I could not charge at work to cross-check the behavior with a warmer but full battery at the start.

Nonetheless, the cold has passed and the gauge has stopped acting up. No more issue thus far.

So untill this happens again, I can only conclude this was due to below freezing temp. What do you think could explain this issue ?
Do I likely have a cracked solder join somewhere in this #77 cell CAN line ?
Could it be the LTC6802 that malfunction only in below breezing cold ? Does that sound like a likely failure for an electronic component ?

 

[JoeS]

Tinoale, glad the problem has gone away (at least, for the time being).

Very disconcerting about the dramatic loss in range - especially with the heater OFF. More confusing is the slow drop on the evening return trip. And, they average out to normal for the round trip?

From my perspective, your video did not demonstrate an unusual situation, as the the bars hop around as the car discharges and is no longer sitting at 100%SoC. I would be tempted to simply plug the car in and let it fully fully charge up and go through the balancing cycle, perhaps a couple of times.

At this point in time, there could be many sources of the problem and your speculations are but some of the ideas. I would simply continue using the car normally until it starts to repeatedly demonstrate something - for this, CaniOn is invaluable.

Sorry, I can't offer any concrete suggestions.