Replace CMU?

[sm7tjc]

Hi.
This is my first post here.
My I-miev from 2011 will not charge more than for about a minute or so.
It started with interrupted charging now and then, but now it is almost impossible to charge.
I get error P1A81 from the BMU, and the yellow light comes on.
When I checked the voltages from CMU12 one cell was jumping all around, from 0V up to 4.05V (the same value as the other cells) very fast.
The car runs just fine, so the cell must be working OK.

I guess that this means that the CMU is faulty.
Is this a common fault that can easily be fixed or is it better to just get a new CMU?
If I get a new CMU it needs to be programmed with the right identity, can I do it from Diagbox or is it more complicated?
Perhaps I just can move the EEPROM from the old board to the new?

Lots of questions, but I'm a fast learner.

 

[kiev]

Not sure that this is a common fault, but there is an IC on the board that measures the cell voltages and drives the cell balancers. There may be several cases in which this chip was found to be at fault, and it could be that replacing the chip would restore your CMU.

Other folks have successfully replaced the CMU board from a salvage pack and either reprogrammed or replaced the eeprom to code in the VIN. CMU 6 and 12 are on the 4-cell modules but the board is the same as used on the 8-cell modules except not populated with parts for the additional cells.

Simon and coulomb know more about the automatic renumbering feature of the BMU.

 

[coulomb]

Simon and coulomb know more about the automatic renumbering feature of the BMU.

I only know the details of how it works, at the firmware level, but unfortunately not any practical method to trigger it.

 

[sm7tjc]

Thank you for the answers.
I found a monitor function in Diagbox, and the cell voltages on CMU12 looked like this

As the dropouts are completely random it looks like the IC that reads the cell voltage is failing.
The IC is from Analog Devices and I just happened to have an account there, so I ordered a couple as samples (FREE ).
But can it perhaps be the power supply section that is causing this?
I guess that it is a lot of work to remove and replace the battery (?) just to test this, so it would be better to be on the safe side.

 

[kiev]

Are you seeing any dropouts on the other cells in that module? It seems that a low voltage power supply issue would affect all the cells, not just one..?

i think there is a forum thread with pictures and schematics of the CMU board, i remember seeing this

 

[coulomb]

i think there is a forum thread with pictures and schematics of the CMU board, i remember seeing this

Yes, one by that helpful Kiev chappie

http://myimiev.com/forum/viewtopic.php?f=23&t=4302

Edit: only a couple of partial schematics are traced, but one happens to involve the auto-numbering circuit.

 

[sm7tjc]

Yes, I have dropouts on all the cells. One cell is more severe, but I can see fluctuation on all the cells (This pack has 4 cells).
Today I found a guy from the Netherlands, who is selling these boards on ebay, and I ordered one.
I am wating for his respond about the numbering. I think that he can configure the board with the right number, but else I gues that it would be easy to just swap the eeprom??
I will read the thread about the CMU boards. It is really interesting to learn about this stuff, but not so funny to be stuck with a car that won't charge. Right now I'm using an old Skoda to get to work, and the cost for fuel is terrible (I had forgot about that part....).

 

[Gary12345]

What country are you in? There are some specialist independent garages that can fix this in the UK, the ones that are part of the HEVRA network.

Cheers.

 

[sm7tjc]

A final report.
The guy who sold CMU boards on ebay had the right board for me.
I have no garage so it was a bit problematic to change the board, but this weekend I could do it at work.
It was easier than I thought to remove the battery and change the board.
Now the car is running fine again. But in the process of dealing with this problem I managed to reset the BMU.
I guess that I now have to cycle the battery a few times before I get a correct reading on the battery status.
I tried to run the battery low today but at about 20% I suddenly got turtle mode. After just a kilometer the power was reduced to less than 2 kW.
I managed to crawl in to the parking lot at the supermarket and charge the car.
After 10 minutes of charging I got two dots on the SOC meter and could drive home without any turtle mode. I guess that the BMU is a bit confused at the moment.

 

[kiev]

Glad to hear you were successful. Have you done a full charge yet and seen the BMU catch back up to the cells?

 

[sm7tjc]

I have charged a couple of cycles now. The battery indicator seems to be reasonably accurate but the guessometer is still a bit optimistic. It show 145 km range at full charge while reality is less than 100.