Battery only charging halfway

[PV1]

Thanks, I'll keep you guys posted. I've since parked Bear at home with 5 bars, 19 miles RR and will drive Koorz until I'm able to get the battery replaced. I don't want to stress that cell any more in case it's bloated and risk a total short.

 

[BarryP]

Yep, every day that I used the car got a full charge the night before. I tried to time my charging to finish between 2-5 AM followed by a 7:30 departure.

So did you do the 120V 13A charge or the 240V 13A charge every day?
Thanx,
-Barry

 

[PV1]

In the winter, I use 120 volts, 12 amps. In the summer, 240 volts at 13 amps.

I almost exclusively used 120 volts at 8 amps for the first 8 months of ownership, then I got the cord upgraded. After that, it was mostly 120 volts unless I needed faster charging, but this winter I used 120/12 for charging, then switched to 240 volts for preheating. I used the slower charge rate to keep the batteries warm.

In the last month, though, I've been quick charging, running the battery temp up into the 90's F, mainly in preparation for a road trip (which doesn't look like that's going to happen now for other reasons). When it's 90 F outside, it's hard to keep the pack cool.

 

[Aerowhatt]

Thanks for the charging history . . . interesting. Somewhere around 900 cycles total on the pack and mostly shallow ones at that. I wish we knew what level of failures they are having statistically.

Aerowhatt

 

[PV1]

I'd like to know overall failure rates for the i-MiEV and others as well.

I think I'll open a new thread for a statistics question. It'll be called "Battery Failure Statistics". That way this thread doesn't get bloated with stats not pertaining to active warranty claims.

 

[tigger19687]

PV1 thank you for starting another thread on this. I was also going to do that so that all the INFO on Who/what when the batteries needed to be replaced or what have you.
I was trying to see how far my new/used i would go for me on a full charge (scan tool in the mail should arrive today but I have to work this weekend).
Unfortunately I had to go to a further building on Friday so I charged the night before to 15 bars (8amp plug )

Hoping to get the CaniOn going and see what my battery is.

 

[kiev]

... In the last month, though, I've been quick charging, running the battery temp up into the 90's F, mainly in preparation for a road ...

Have you driven the car much since the onset of this reduced cell voltage and noticed any degradation in full-power performance? The reason i ask is to determine if the reduced cell can still pass the full current of the pack or does it limit the current for some reason?

Would you be able to make a Wide Open Throttle (WOT) test run before turning in the car for some comparison testing? This might be used as an early-warning indicator test for a failed cell to get some feel to determine how badly this affects performance and how urgent a problem this might be, i.e. you can still charge and drive it for a month, or it must go to the dealer immediately.

For example, i have seen a cell failure in a DIY conversion with large LFP prismatics in which one cell was reading 0 volts, yet was still able to pass full current. The pack voltage was about 3.2 v lower than expected, but was still accepting charge and being driven. The pack had been run to empty twice about 6 months prior during it's initial commissioning runs due to driver error. Upon dissection of the cell, all the copper had been dissolved from the collector foils and plated over onto the separator sheets.

 

[PV1]

For that DIY pack, it sounds like the cell was fully shorted internally.

I've done some full throttle trying to get up to speed, and it has held a steady voltage though still lower than the others. No lack in acceleration power. It acts like that balancer was stuck on when it was sitting. The cell is just at a lower charge than the rest.

If it wasn't for the warranty, I'd like to just drop the pack and manually charge that cell back up.

 

[Aerowhatt]

I've done some full throttle trying to get up to speed, and it has held a steady voltage though still lower than the others. No lack in acceleration power. It acts like that balancer was stuck on when it was sitting. The cell is just at a lower charge than the rest.

If it wasn't for the warranty, I'd like to just drop the pack and manually charge that cell back up.

I've been watching your data and observations about this recent unfortunate development with great interest (thanks so much for the details).

I have pretty extensive experience with Amita technologies 30ah cells of similar chemistry to the I battery. From the evidence that you have so far it sounds to me like the voltage sense for that cell has gone off spec. It's kicking in the shunt at lower voltage than it should and thus draining it down to where it now appears to be stable. Generally a defective cell will not behave normally (perhaps it will very early on but not for long) under discharge. Cell resistance rises sharply and the voltage on that cell will sag dramatically under big loads. The problem is likely as simple as an out of spec resistor on the BMS board serving that cell. If that's the case you would never get it balanced (it thinks it already is).

Just some thoughts from past experience,

Aerowhatt

 

[Pier]

PV1 wrote: « If it wasn't for the warranty, I'd like to just drop the pack and manually charge that cell back up. »

At the dealer, before changing my pack, they tried to charge the bad cell separately to full charge without any success ( stayed to 75% soc).

Pier

 

[kiev]

.... No lack in acceleration power. It acts like that balancer was stuck on when it was sitting. The cell is just at a lower charge than the rest.
...

'... Generally a defective cell will not behave normally (perhaps it will very early on but not for long) under discharge. Cell resistance rises sharply and the voltage on that cell will sag dramatically under big loads. The problem is likely as simple as an out of spec resistor on the BMS board serving that cell. If that's the case you would never get it balanced (it thinks it already is).

Just some thoughts from past experience,

Aerowhatt

Some excellent clues here, +1 on this post, and more than likely a 2¢ resistor failure rather than an actual bad cell.

[off topic mode/]
When GS Yuasa was being questioned by the NTSB about the Boeing 787 battery incidents they testified to having no cell failures in over 10,000 cells used by other customers. The miev LEV-50 cell is a lower capacity cell of similar jelly-roll design to the 787 cells (LVP-65, CC at 1C max/CV at 4.025 with current taper to C/20).

Boeing claimed that overcharging was not possibly the culprit because the FDR showed that the battery buss voltage was held at 32 volts by the Securaplane charger. i looked at the FDR data also and found that it showed 1-2 amps of current flow into the battery all the time, (such as is done in trickle-charging ?). Securaplane's charger patent is based upon a Ni-Cd cell chemistry charging scheme.
[/mode]http://www.google.com/patents/US5780994

 

[PV1]

So, since it seems that something on the management board went, I think I'll keep driving it and see if it comes into balance.

When other packs failed, I wanted to gather as much info as I could. Now that something happened to my pack, I want to gather every bit of data I can.

 

[Aerowhatt]

So, since it seems that something on the management board went, I think I'll keep driving it and see if it comes into balance.

When other packs failed, I wanted to gather as much info as I could. Now that something happened to my pack, I want to gather every bit of data I can.

I love your attitude! If you want to attempt get into balance the lowest charge rate will do the best. The BMS makes the charger bail out on the charge cycle once a cell reaches the high set point. The slower the charge rate the better the shunts can keep that from happening. L1 of course, and if you have the EVSE upgrade where you can select amperage as low as 1 amp for the finish charge wouldn't be a bad idea (if it goes that low, that is). I had some 7 cell packs that were left abandoned for a long time before I acquired them. Very out of balance and I had to rig a C/65 charge rate and be very patient to bring them into line. Looking forward to the reports.

Aerowhatt

 

[kiev]

That's the spirit--adventure and the quest for knowledge.

You've got the tools to monitor the cell and ensure it doesn't get driven too low, so if you can drive it for a couple of weeks to get some data before swapping out, well then, press on regardless, and endeavor to persevere.

 

[PV1]

Started a slow charge of the pack. It has drifted to .2 volts lower than the rest, lost a bar, and lost 5 miles RR while parked for a day. It is at 10 bars right now and pulling 200 watts from the wall. The cell has come up to .165 lower, so the balancers are definitely at work. There is enough range still to drive it to work (which works out because my brother now needs to borrow my silver car since his car lost a wheel bearing). I can always charge at work as well to double up the balancing effort.

If it is cell self-discharge, it's very slight and the balancers can overcome it. If it's something with the board, I hope it is still accurately measuring voltage.

So far, I've started each driving morning at 13 bars, 12 bars, and either 10 or 11 bars tomorrow morning, depending on how long the car will balance the pack before shutting down. I'm trying to leave CaniOn run as much as I can, but came home tonight and found the phone had shut off.

 

[PV1]

So, it is now hitting lows more than .2 volts lower than the other cells. The EV warning light is now starting to light up.

 

[kiev]

So, it is now hitting lows more than .2 volts lower than the other cells. The EV warning light is now starting to light up.

Is that the electric motor unit warning light? Shaped like a car with an exclamation mark?

 

[JoeS]

So, it is now hitting lows more than .2 volts lower than the other cells. The EV warning light is now starting to light up.

Time to get Bear to the dealer while you can still make it.

I waited too long and had to have it flatbedded.

 

[PV1]

Yes, the Electric Motor Unit warning light.

I'm making the call today and hope they can get the car in soon. I like gathering all of this information, but not if I have to charge twice to go 20 miles. The good thing is that a Mitsubishi i-MiEV dealer is less of a distance from my house than a roundtrip to work, and I can still make round trips without a mid-day charge (for now).

This morning and yesterday morning, I started at 9 bars. I finished yesterday with 2 bars, 7 miles RR. The problem with such a long charge process is that the pack is rather warm.

 

[JoeS]

This morning and yesterday morning, I started at 9 bars. I finished yesterday with 2 bars, 7 miles RR...

CaniOn photos of the cell voltages at that low a SoC?