Another i-MiEV Battery Pack Failing...? Help!

[martinwinlow]

Hi All,

I recently acquired a Mitsubishi i-MiEV, first UK-registered in 2009 and one of the first to arrive here (apparently). It was originally red and white but was wrapped in SSE's green wind turbine logo thing (SSE is a large electricity generator/distributor in the UK). It also has a 'dumb' charging port i.e. no EVSE required, just charges straight from a standard UK 13A socket. VIN number is WEGTPHE4XBXXX01--, maker's plate on F/N/S Door frame shows:
'Model: ZAA HA3W LDD
HA3W - 0000917
Engine: Y4F1 Ext: R4WA'

I bought the car earlier this year for relatively little money and we have another 2 here so, push-come-to-shove I can always use it for spares but I'd really rather like to get it working.

Basically, it won't start. As soon as the ignition is turned on the yellow exclamation mark appears and doesn't go out and the hazard lights come on. Also, no double-ding or 'ready' light when the key is turned to 'start'. The previous keeper (now residing in Australia) says that the errors initially would clear by simply cycling the ignition but eventually he took it to a Mitsu dealer who stated they could not fix it without replacing the battery pack. Hence the low sale price.

An early look with caniOn showed this (taken 09Apr2019):-

... suggesting cells 38 and 42 are dying/aging.

I couldn't get my 2014 eBay-bought MUT3 working at the time as I has lost the CD but today, I finally managed to source some working sofware and got my MUT3 working. I see that cell 38's voltage is now 2.1V.

I offer the following screenshots of an amateur's initial use of it in the hope that they will be useful in determining a positive course of action:-

These 3 EV-ECU errors would not clear...

All other errors cleared OK...

So, Q's...

1/ Is there a way of using the MUT3 to overide the non-clearable errors to get the car into a 'Ready' state to see if it basically works ie goes forwards and backwards etc?

2/ Is there a way of using the MUT3 to overide the errors to get the car to charge?

3/ From this, does it appear that I'll need to replace cells 38 and maybe 42 to stop the errors from coming up and preventing he car from working/charging?

4/ What else can I check to confirm I need to replace the aforementioned cells and/or fix anything else?

Any help gratefully received.

 

[kiev]

CMU05 cell F is Canion cell #38, so both devices show agreement of your lowest cell.

Canion cell #42 will be in cell B of CMU06 cell , the 4-cell module on the left side of the pack.

Some folks have had bad cells, others have had bad chips on the CMU board that caused bad cell voltage data and throw the DTCs.

i assume you are using a known-to-be-fresh 12V Aux battery--the car won't function well with a weak or old Aux.

From what i've read there is no MUT function to override the fault codes to get the car to go to READY. Any critical DTCs have to be fixed and cleared.

There is a member Rupert on the speakev forum going thru a similar problem on his Czero.
https://www.speakev.com/threads/c-zero-battery-pack-repair.137750/

The good news is that other folks have fixed their cars by either replacing single cells or CMU boards or ICs on the CMU boards, so it can be done.

 

[martinwinlow]

Hi Kiev,

Thank you for taking the time to reply.

Robert's thread has made for a most interesting read - a bit like an electric detective novel!

Yes, I installed a brand new, fully charged 12V aux battery just before performing this MUT3 analysis - the old one was definitely dead.


Qs...
1/ Will the voltage sampling on the MUT3 be rapid enough to ID the sort of voltage fluctuations that would likely indicate a PCB/IC issue rather than a duff cell (where I assume the voltage would be pretty steady)?

2/ Would you agree that I am at the point where I will have to drop and open up the pack to make any further progress with rectifying the fault (whatever it may turn out to be)?

I'm away for a week now but I'll get back on ot on my return. Thanks again.

Regards, Martin.

 

[Gary12345]

Hi there,

I've fixed the fluctuating voltage issue (caused by failed chip in the battery pack) on my i-miev and from looking at your pictures it is pretty certain that you have a cell which has a low capacity compared to the others.

The next task would be to remove the main drive battery and find the cell with the low voltage, then test/swap that cell with a good one. You can't over-ride the error because the car knows the cell looks bad so wont charge or use it. You may be able to charge the cell so its voltage is within normal range (3-4.1v) and then put it back in the car, do a full charge and then see if the car works, however its quite a bit of work getting the main battery pack in and out of the car.

Thanks,

Gary.

 

[martinwinlow]

Thanks, Gary.

50Ah cells, then?

 

[Gary12345]

Hi yes they are 50AH cells.

Seems like you've got a dead cell, although the screenshots from Canion and MUT dont really match up, plus you have the fault code for Voltage Sensor.

Drop the pack out, test each cell with a multimeter, see if any are actually low. If they are, you'll need replacement cells. If not, it is the LTC chip on one of the CMU boards (on top of each battery module) that has become faulty. I got a new LTC chip fitted by a mobile phone repair guy!

Thanks.

 

[DBMandrake]

Any progress on this ?

Some very inconsistent results between Canion and MUT there - and even within that Canion screen shot, which claims the pack is at a SoC of 6% yet the pack voltage is 358 volts (full charge is 360 volts, discharged is about 320 volts) and 3.98 volts for a cell would ordinarily be approximately 80% SoC.

Because the car won't go into ready mode (due to an excessive voltage differential at a voltage that should be a high SoC, which will set a fault code) you can't try to discharge the pack to re-charge it and attempt to re-balance the cells.

The 2.1v reported by the MUT as stated above means either you have the intermittently faulty LTC chip which is giving a false reading, this would fit your symptoms, or if the cell really is at 2.1v and has sat at at that low voltage for any length of time it will be stuffed, even if it was OK before...

Before pulling the pack apart I would check to see whether you can reproduce that 2.1v cell reading - check in both MUT and Canion.

Also assuming the voltage is back to 3.98v and not 2.1v, before pulling the pack apart it would be worth trying to see if the "cell smoothing" diagnostic function can be activated - if it can, leave it active for many hours (at least 6 hours) and see if it will equalise the cell voltage imbalance.

It will turn on the bleeders for all high cells and attempt to bleed them down to the voltage of the lowest cell. If this is able to equalise the voltages of the cells it will then probably let you put the car into ready mode or charge it - from there you may be able to get a better idea of whether it is two faulty cells you have (which is what the Canion screenshot looks like) or an intermittently faulty LTC chip - which is what the MUT screenshot suggests... :?

I suppose if you're really unlucky you could have a faulty LTC chip and two dud cells!

If you need two cells and you're having difficulty sourcing cells give these UK based guys a try:

https://www.secondlife-evbatteries.com/

I've just ordered 4 cells from them and have just started capacity testing them to verify their performance, and assuming their capacity is as promised I'll be fitting them to my car in the not too distant future.

If it turns out you need a CMU board let me know and I can send you the contact details for an EV breaker in the Netherlands who can supply CMU units - the same guy who supplied Rupert Burbridge his CMU boards that he used for a successful repair of his pack.

 

[martinwinlow]

Hi Simon,

Thanks for all that. I have already bought 2 cells and a BMS PCB from 2ndlife and they are sitting on the floor staring at me (and have been for the last 7 weeks... ).

A coule of questions...
1/ How is the 'cell smoothing' thing done (via the MUT, I assume?)?
2/ Is it likely that leaving the car sitting for much longer will cause any further pack problems (I suppose I am worried about the healthy cells discharging dangerously)? I'm just really struggling to find the time to take on the job of dropping the pack etc, right now...
3/ If I do use the 'cell smoothing' facility, isn't there a danger that *all* the cells will be drawn down to a dangerously low level?

I'll have another look with the MUT shortly and post the results here.

MW

 

[DBMandrake]

Hi Simon,

Thanks for all that. I have already bought 2 cells and a BMS PCB from 2ndlife and they are sitting on the floor staring at me (and have been for the last 7 weeks... ).

A coule of questions...
1/ How is the 'cell smoothing' thing done (via the MUT, I assume?)?

It's a feature of MUT-III for i-Miev or Diagbox for Ion/C-Zero - yes. In Diagbox it's one of the repair functions in the BMU ECU, The latest beta version of the android app Hobdrive (discussed elsewhere on this forum) also claims to be able to activate cell smoothing but I have not tried this myself. Since you already have a MUT I would use that.

2/ Is it likely that leaving the car sitting for much longer will cause any further pack problems (I suppose I am worried about the healthy cells discharging dangerously)? I'm just really struggling to find the time to take on the job of dropping the pack etc, right now...

Depends what voltage the cells are at now - can you plug both Canion and MUT back in and get a voltage reading of all the cells on both to check on them ?

Any cells that are over about 3.3 volts will be fine however if any are below 3 volts I would be concerned. If one of the cells really is sitting at 2.1 volts and it isn't a faulty measurement due to the LTC chip it will be damaged now and will need replacing so any further time spent at that low voltage doesn't really matter as it's already toast.

Keep in mind that the 12 CMU boards are powered directly from the cells themselves. They go into a very low power sleep mode when the car is off, but they draw a tiny amount of power when the car is not in use, likewise the cells themselves will have a tiny amount of self discharge. So the pack will gradually discharge when left sitting but we're talking about maybe 6 months to lose 10-20% charge as the self discharge rate of Lithium Ion is very low compared to most other kinds of cells like Lead Acid, NiMH etc...

3/ If I do use the 'cell smoothing' facility, isn't there a danger that *all* the cells will be drawn down to a dangerously low level?

I doubt the smoothing function would activate if any cells were too low in voltage. I would only try the cell smoothing feature if the lowest cell is above 3.7 volts.

 

[martinwinlow]

Hi All,

I hope this isn't going to put anyone's nose out of joint but I'm moving my tale of woe to speakev due to the very obvious parallels with another thread there... https://www.speakev.com/threads/c-zero-battery-pack-repair.137750/page-1. Hopefully see you all there.

 

[martinwinlow]

For those who are interested, this is a link to the SpeakEV thread I am now updating re my repair...

https://www.speakev.com/threads/i-miev-with-failing-battery-another-cmu-ltc-chip-failure.144620/page-2

MW