Solved: P1A46 on 2012 with 18 recently swapped stock cells. Drives but won't charge.

[Jiminy]

Howdy,

I just completed my first stock cell swap on a friend's US spec 2012 i-MiEV with 82K miles. Swapped in 18 cells after carefully matching voltages. Got codes P1A46 in BMU and P1A45 in EV-ECU upon first wake up. No other codes. Cleared codes with Autel scanner. Drove car 7 miles and it drove fine. Not as peppy as my car with 4 year old pack but drivable. Got home after test drive, plugged it into Level 2 evse but charge light flashed and car icon with exclamation mark appeared again and charge indicator light went off again.

Particulars:

1. Car has new Lifepo4 20Ah battery with 13.4 resting voltage. Been using these on my car and on others with no issues. This is not the issue.
2. Unplugged AC compressor HV, no change
3. Unplugged AC compressor LV, no change
4. Unplugged Heater HV, no change
5. Unplugged Heater LV, no change
6. Checked continuity of black ground wires at C-21 and C-22, good
7. Checked 12V voltage on center pin of C22, good
8. Made sure battery pack ground strap was attached well.
9. Performed continuity checks on wiring from BMU to C21 and C22, good except one wire color different from troubleshooting chart.
10. Tried different evse, no go
11. DC-DC is charging at 14.5v
12. Checked fuses under hood, all good
13. All CMUs report no issues. All 66 temp sensors normal at 24c ish.
14. During my test drive the newly installed, used but new-to-me 2012 cells started showing as high spots in Canion bar chart. Now, there is .2V difference between new and original cells. The replacement stock cells tested at 25-30Ah at 8A discharge from 4.1V-3.5V. Old cells tested at ~12Ah. Yes, they are done but this experiment is important to me.

I'm at the point of dropping the pack again and seeing if I loosened something that I forgot to retighten but there was only one 8mm bolt that I considered loosening to make it easier to remove module 1 but I distinctly remembering deciding not to loosen it and just manipulated the module under that bus bar so that I wouldn't loosen and then forget to retighten it.

Any ideas would be appreciated.

Jim

 

[Jiminy]

Can you spot the newer cells? These were initially matched well to the others before test drive but I was champing at the bit for a drive so...

 

[kiev]

Those 2 DTCs are just one fault; the BMU is the primary detector and threw a code, and the EV-ECU throws its code to point to the BMU.

Are you saying that you made all the checks as found in the steps of the FSM for this code (see attached pdf)? Link to corrected MSB pdf,
http://mmc-manuals.ru/msb/BASE/003/2012/MSB-12EXML54-003/E/MSB-12EXML54-003.pdf

Also are you saying that you might have bent or put a torque on one of the module connecting bus bars? The longest module 1 bus bar connects the lowest voltage cell to the HV(-), negative terminal buss of the pack.

 

[Jiminy]

Hi Kiev,

I did as many of the checks I could do and it pointed to a loose connection or a bad ground fault detector module. I went ahead and pulled the pack this morning and found that I had forgotten to plug the HV cable connector back into the ground fault detector module:



Put Humpty Dumpty back together again and it is charging fine right now. So, that issue is fixed. Next is whether the mismatch in cell quality will allow the car to function somewhat normally as a 30 mile car.

Jim

 

[MickeyS70]

Next is whether the mismatch in cell quality will allow the car to function somewhat normally as a 30 mile car.

Well done.

It’s not the new cells that you need to worry about; they will always outperform the remaining ones and therefore remain imbalanced.

Unfortunately the experience of others that have replaced individual cells into a degraded pack isn’t promising, most agreed that they should have swapped (a lot) more to start with and that benefits were only temporary.

But then beggars can’t be choosers…

 

[Jiminy]

Thanks Mickey.

This swap was just me trying to beat the system. I thought if I spent a bunch of time watching scanning equipment for the laggards that I could replace them all and have a reliable 30 mile car with AC. The car's owner is not without resources. It wasn't about the money. She drives a Model 3 now so the I-MiEV would be a dog park and grocery getter local errand car. We just decided to start the process by spending less to see what would happen.

The car charged up OK yesterday afternoon and I set out in the evening on the shakedown run with 39 miles on GOM. Drove 20 miles on the first half of the battery gauge, all 40 mph or less. It was looking encouraging. Lo and behold the turtle showed up less than 3 miles later and I limped it to a Chargepoint charger and spent 30 minutes charging to make it the 4.4 miles home. Got a battery Each Cell Voltage difference error (P1A4B) in BMU that I cleared while charging. Car made it home fine. This is under light load at the 20+ mile range:




Can you spot the 18 newer cells? I knew you could

SO, I learned the same lesson for myself which is the only way I seem to learn anything. Back to the drawing board. At least I can get the pack out in an hour now and by myself which is good.

 

[piev]

The reason this occurs is the new cells produce higher currents that the weak cells cannot handle and the Voltage sags so much you get a turtle. I have seen worse things happen than this… on a homemade pack a guy I know replaced half of the pack with new cells and melted the wires between the pack because the new cells were carrying too much of the load. It’s best to match all cells in capacity not voltage. I also had a series cell go to zero in my first conversion. I had 00 AWG cables to the controller and they were hot to the touch. The healthy batteries seem to eat the weak ones. Found a loose connection so I suppose the high resistance caused my issue.