Battery only charging halfway

Mitsubishi i-MiEV Forum

Help Support Mitsubishi i-MiEV Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

imabubble

New member
Joined
Mar 9, 2015
Messages
4
Hello everyone,

This is my first post, so nice to meet you all. This is also my first EV, and my first real problem....

I usually charge my 2012 MiEV at work at a normal outlet, and then when I get home through an extension chord. For the past few days, I've let it charge for enough time to charge it up most or all the way, but it's stopped at a little over halfway (indicator says 37 miles).

Just today, I noticed the car wasn't charging, even though the green "ready light" was on, so I unplugged and replugged it in. The orange "charging" light came on and I watched it to see what would happen, and sure enough a minute later, it turned off and the car stopped charging. The red "fault" light never showed up. I tried again, and the same sequence of events repeated.

I'm wondering, what should I try next? Could it be a problem with my plug, and so I should try charging at Level 2? Don't tell me the car battery has degraded to half its capacity after only three years. Has anyone else encountered this kind of problem?
 
Try driving it a few miles then plugging in again. See if it takes a charge. It's possible that the battery could have a failed cell, but if this is sudden, then most likely the on board charger failed.
 
imabubble, sadly I believe you are experiencing the symptom of having lost one or more cells. Whereabouts are you located, as maybe a forum member closeby with CaniOn could help verify this? Suggest you get it to a dealer while you can still drive it there. The good news is that the battery is still covered by warranty.

Edit:

imabubble said:
...Just today, I noticed the car wasn't charging, even though the green "ready light" was on, so I unplugged and replugged it in...
Perhaps a minor correction to what you wrote: in order for the "Ready" light to come on, the car must be "Started" using the "ignition key". The vehicle's interlocks will prevent that from happening if the J1772 connector is plugged in, even if it is no longer charging. Conversely, if the car is in "Ready" and you plug in the J1772 connector, it will immediately turn off the car.

I'm pretty sure you didn't intend to say that the "Ready" light was on, as the basic issue I believe you have is that the car stops charging (and now, doesn't even charge) despite the fuel gauge showing only a partially-charged battery.

Please do keep us apprised of your developments.
 
JoeS said:
Perhaps a minor correction to what you wrote: in order for the "Ready" light to come on, the car must be "Started" using the "ignition key"

The ready light I'm talking about is in on the charging pack, not in the car. It comes on when I plug the charger into an outlet - I always confirm that the green light is on before I plug the charging pack into my actual car. (I almost always only use Level 1 charging)

Thanks for your responses - I will try making it to the dealer in San Jose, CA this weekend (it's about 30 miles away, but I have a good AAA membership, so if I don't make it there, I can get towed the rest of the way) and I'll update this thread on what happens.

Meanwhile, I'm wondering how a battery loses a cell? Is there something I can do to prevent this in the future?
 
If the cell has a defect, there isn't much of anything you can do. To minimize risk of damaging the battery, the things to avoid are extreme battery temperatures (above 90 F and below 32 F), leaving the battery at a full charge for more than a day, and running the battery to turtle often. Also, keeping amperage low, especially when cold, will help too. Basically, easy on the accelerator :mrgreen: .
 
imabubble, sorry for my confusion. We normally refer to the unit that interfaces the wall to the car by its generic term - EVSE (Electric Vehicle Supply Equipment), or "Charging Station". In this case, you are referring to the Mitsubishi-supplied unit that came with the car.

In which direction from the San Jose dealership are you located, as I'm up in Los Altos Hills and would be happy to bring my CaniOn over to you and check your battery?

Over the last couple of weeks my new used 2012 i-MiEV exhibited failure symptoms that evolved into symptoms identical to yours and it is presently in the San Jose dealership's shop. I took it there yesterday and I am awaiting their analysis. I have no idea how much care and feeding the battery experienced by the previous owner of my used i-MiEV (from the Bakersfield area).

My first i-MiEV is still going strong at over 36K miles, but I have babied that pack since its birth.

I'm in the process of preparing a separate post with CaniOn screenshots showing what happened inside the pack (love CaniOn!), but for completeness I would first like to have the feedback from the dealership.
 
*Usually* a bad cell will still allow the battery to charge to 3/4ths or 7/8ths full. Your symptom could also be caused (I think) by the delay timer function on the Remote Control. If you have played around at all with the remote just prior to the car exhibiting the current symptoms, that's something I would look into. The remote can cause some pretty unusual problems

Don
 
JoeS said:
I'm in the process of preparing a separate post with CaniOn screenshots showing what happened inside the pack (love CaniOn!), but for completeness I would first like to have the feedback from the dealership.
The dealer confirmed that my pack had a bad cell and asked me to fill in some paperwork which they are submitting to Mitsu HQ.
 
Sorry to hear of the battery failure.
Good news that they will fix things up.

A look at the CanIon read outs will be a super reference for others to evaluate their battery state.
Looking forward to seeing it.
 
imabubble,

How many miles/km on your car at the inception of the charging problem?
Good luck with the battery situation. Symptoms sound similar to my sister-in-laws 2012 battery problem. (Her battery was replaced).

2012, silver ES with QC
16500+miles
 
Rather than start a new thread, I thought I'd just write up my experience here. I had made an appointment to take the i-MiEV in to the dealer on March 9. The car had 24.5K miles on it when I bought it used in December and now has 27.9K miles.

Here's the letter I wrote to the dealer on Sunday, 8 March, but updated the following morning when I saw a further deterioration.

The purpose of this letter is to ensure that I properly communicate to you the problem that I am experiencing with my 2012 i-MiEV.

Over the last several weeks the car's range has been diminishing and battery charging has deteriorated. Here are the symptoms, roughly in chronological order:

1. First of all, the range available seemed to start dropping. I did not worry about this too much as it could have been caused by cooler weather or simply my driving style inconsistency.

2. Then the charging seemed to go much faster than normal. For example, normally, using 240vac, the car recharged at a rate of around 3 bars/hour, but then it seemed to start taking less time to do the same thing.

3. About a week ago I noticed that, when attempting to fully charge, the charger would not turn off for a long time despite the car showing fully charged. Looking at my home power monitor (T.E.D. 5000) I noted that the charge would continue to very slowly taper off all the way down to zero power over a long period of time. Normally, the power is constant, then decays to a certain point, and then the car's charger shuts off well before dropping down to zero power.

4. Then, later last week, something changed and the charger would no longer fully charge the car, and stopped at 13 bars. After this I drove it a bit and when I tried recharging again it has now stopped at 12 bars. As I write this I hope I have enough charge available to drive it the 25 miles to your shop.

UPDATE 9 March, 2015: This morning I tried to top the car up before going to your shop. When I plugged the car in it started to charge and then almost immediately stopped charging. The 'fuel gauge' had dropped to 8 bars!

5. The car's charger properly responds to plugging in and starting the charge. The situation is the same whether I use my 120vac EVSE or my 240vac EVSE.

6. The car has had no failure indications on the dashboard display.

I own two i-MiEVs and really love these cars! I have not had one bit of trouble with my first i-MiEV, which now has over 36,000 miles.


So, on March 6th here's what the CaniOn readout looked like:

MittiBatteryFail030615.jpg


On Monday March 9th, when I tried to top up the car before the 25-mile trip to the dealer and the car basically would not take a charge, here's what the readout looked like (note that it has dropped from 70% to 55% without even having been driven and only having power turned on for the few minutes to take the CaniOn shots):

MittiBatteryFail030915.jpg


I ended up calling Mitsubishi to have the car flatbeded to the dealership, but they told me that since it is two months past the 36-month warranty that I would have to pay $103 for this service. We were on a tight schedule that day so I elected not to try to drive it there as who knows how the battery would respond? The dealer was very cooperative and allowed that if this is a warranty repair then he'd try to get the towing refunded.

In talking with the dealer I elected not to share my CaniOn information with him, as I didn't want to be discourteous and appear to be telling him what to do.

So, the dealer called me on 10 March to let me know that they had found a defective cell in the pack (corroborating what CaniOn had already told me) and wanted me to come in the following day to fill out a questionnaire from Mitsubishi. I did that but for some reason the service manager didn't want to run a copy of it for me. The questions were somewhat generic and presumed an immediate failure rather than a gradual deterioration, so I merely referred to my letter on the questionnaire and asked the service manager to be sure to attach a copy of it to that questionnaire.

Next step was for the dealer to submit the paperwork to Mitsu HQ and he'd let me then know what happens next.

What caused the failure? This car had lived for three years in Bakersfield, California, where it gets awfully hot all summer long. The car was a daily driver and if I recall the owner said he charged it at work and the car was outside (but in the shade), so we can draw our own conclusions as I don't think he had a timer shutoff or used the Remote to stop charging before it went full. But who knows? Even though I got reasonable range from the car when I first bought it, the consumption and RR display seemed a little 'different' than what I was used to with my first i-MiEV. Don't know if these might have been early indications of something wrong. I hadn't noticed any cell problems and only started taking CaniOn screenshots in the last couple of weeks.

I found it very interesting looking at the CaniOn readout. Even though we don't see it in the pictures, the voltage and probably minimal current draw is telling the charger that the battery is fully charged, so it shuts off. Independently, the SoC calculation says that it's only at 55% capacity, which is accurately represented on the fuel gauge as 8 bars and also a corresponding RR decrease. A 200mv drop in one cell is enough to seriously affect this system!

I have every confidence that Mitsubishi will do right by me and repair/replace the pack, and I'll keep y'all informed of my progress.
 
Good news is that I'm not seeing that with my silver car, but my range is shorter than normal (haven't pinpointed it yet). I just checked it the other day, and it seems to be coming into balance.

Joe, it seems that the car will attempt to balance the pack as long as it can, but it has to stop whenever a cell hits 4.105 volts, as that is the cutoff point.

When you think about it, these cells are quite sensitive to even a minor disparity in cell voltages. Over the safe usable range of charge, our cells only lose about 1/2 a volt (500 mV) between full charge of 4.1 volts and the 1 bar ~3.6 volts. Of course, it is possible to run the cells down to the cutoff of 2.75 volts if they have degraded to that point, but the car usually shuts down before reaching that level because the SoC calculation reaches 0%. That fact alone seems to block the possibility of replacing our cells with larger capacity cells on our own, but that's a topic for another thread.

Regarding these failures, it continues to look like one cell in the battery prompts the need to have the pack replaced. For 1,000 i-MiEVs with 88 cells each, given 10 failures for example (extrapolating an estimation) with one cell failure per pack, that is 10 cells out of 88,000 going bad. The failure rate is well below 1%, which is pretty amazing. 1 in 100 cars, not so much.

For a vehicle like the Tesla Model S, which has a bunch of small cells in a parallel-series arrangement (big "cells" made from paralleled batteries, each big "cell" wired in series to each other), a single cell failure wouldn't decrease the range as dramatically as our i-MiEV since one cell losing half of its capacity is a much smaller percentage of total usable capacity. I wonder what Tesla's cell failure rate actually is, and if we'd be better off with twin 25 Ah cells instead of a single 50 Ah cell, but I'm drifting again. Hope they get your i-MiEV fixed up.
 
JoeS said:
Over the last several weeks the car's range has been diminishing and battery charging has deteriorated. Here are the symptoms, roughly in chronological order:
Joe,
Thanks for this very informative post.
Your CaniOn readings confirm that the State-of-Charge percentage (SoC%) is calculated using the weakest cell (Cell 87 in your case).

Here are some assumptions based on your observations:
JoeS said:
1. First of all, the range available seemed to start dropping. I did not worry about this too much as it could have been caused by cooler weather or simply my driving style inconsistency.
As cell #87 started to fail, it discharged faster than the other good cells, and since the SoC% is calculated using the weakest cell your range decreased.

JoeS said:
2. Then the charging seemed to go much faster than normal. For example, normally, using 240vac, the car recharged at a rate of around 3 bars/hour, but then it seemed to start taking less time to do the same thing.
Even though cell #87 had discharged to give a low SoC%, the remaining cells were still near their maximum charge voltage and you were mainly charging bad cell #87. Charging only one failing cell took less time.

JoeS said:
3. About a week ago I noticed that, when attempting to fully charge, the charger would not turn off for a long time despite the car showing fully charged. Looking at my home power monitor (T.E.D. 5000) I noted that the charge would continue to very slowly taper off all the way down to zero power over a long period of time. Normally, the power is constant, then decays to a certain point, and then the car's charger shuts off well before dropping down to zero power.
The battery management system was trying to charge bad cell #87 to 4.105 volts, and as long as current keeps flowing into the battery the charger remains running.

JoeS said:
4. Then, later last week, something changed and the charger would no longer fully charge the car, and stopped at 13 bars. After this I drove it a bit and when I tried recharging again it has now stopped at 12 bars. As I write this I hope I have enough charge available to drive it the 25 miles to your shop.
Bad cell #87 would no longer charge to 4.105 volts and no more charging current would flow into bad cell #87, and all of the remaining cells were fully charged. The car stopped charging at a lower SoC% (or a lower number of bars.)
 
Hey Joe,

Thanks for this post. I believe you are the first to have a cell failure and have canion data to support the failure. I also thought this looks like its in the area of the "hot temp sensor" the one that always seem to be a few deg above all the others when charging. I wonder if there is any correlation. Since this was a hot climate car perhaps that cell just got hotter then the others and that induced the failure. Or it might just be a random bad cell. In any case thanks again for your very informative post.

Don.....
 
RobertC, thank you for your analysis. I had a similar thought process but elected to presently state "just the facts, m'am" rather than try to dissect it.

DonDakin, even though both the failed #87 cell and the typically hot-running sensor are on the right side of the CaniOn screen, I don't know what their relative proximity is within the battery pack. Didn't someone identify that particular temperature sensor as being close to a coolant line?

PV1, with these large-format cells the automakers are really relying heavily on individual cell quality - I'm impressed. Tesla, OTOH, can easily tolerate a fair number of individual cell failures without anyone even noticing; besides, each of their cells is typically lightly stressed. Interesting system design contrast.

A bit off-topic: I've been running one of my Sparrows with two paralleled Enerdel Lithium(NMC) strings of 36 cell pairs each (3x2P12S) without balancing for the last 1-1/2 years (yes, very conservative pack charge and discharge control). Well, I had recently noticed a 0.2v-0.3v lower reading on one of the 12-cell modules so today I took that string apart and found ONE cell pair to read 3.47 volts whereas the other 35 cells in that string read 3.74v :shock: (3.74v represents about 60%SoC). I'm presently simply charging that one cell pair to see if I can bring it back; if not, I'll have to disassemble that module and replace the cell pair. Guess it's not my week for Lithium cell happiness. :cry:
 
JoeS said:
... even though both the failed #87 cell and the typically hot-running sensor are on the right side of the CaniOn screen, I don't know what their relative proximity is within the battery pack. Didn't someone identify that particular temperature sensor as being close to a coolant line?
Temperature sensor #64 runs high when charging:
http://myimiev.com/forum/viewtopic.php?f=23&t=1789&p=16835#p16835

The temperature sensors and voltage sensors are shown in this illustration:
C6DcyF4.png


The battery layout is shown in this illustration:

Image deleted 17 May 2019 due to technical inaccuracy.
For reference, here is what the image had shown:
http://i.imgur.com/7MXf3gr.png

Here is the corrected battery cell layout (courtesy of kiev):

m9i3pVF.png


Here is a picture of an open battery pack:
oHPMWO0l.png


Here is an illustration showing the radiator hose:
nnK9zy9l.png


The battery pack is composed of ten LEV50-8 battery modules with eight 3.7V LEV50 batteries and six temperature sensors each, and two LEV50-4 battery modules with four 3.7V LEV50 batteries and three temperature sensors each, for a total of 88 LEV50 batteries and 66 temperature sensors.
The two LEV50-4 battery modules sit in the battery pack so as to have their battery management circuitry and temperature sensors mounted on top of the batteries, while the ten LEV50-4 battery modules sit in the battery pack so as to have their battery management circuitry and temperature sensors mounted on the side of the batteries.

According to these images, Joe's bad cell #87 lies between temperature sensors #65 and #66.
 
I wonder if this only happens in the 2012's ?
Anyone know if this happened in a 2014 ?
And I wonder if there may have been any changes to the 2014 from the 2012 ?

Is there a way of putting some kind of barrier between the battery and the hoses? Probably not an easy task if someone could do it at all.
 
tigger19687 said:
Is there a way of putting some kind of barrier between the battery and the hoses? Probably not an easy task if someone could do it at all.
I don't think that having temperature sensor #64 read higher during charging has any effect on the battery, and it only an assumption that the radiator hose is causing temperature sensor #64 to read higher.

If someone experiences a battery cell failure once the battery warranty has expired it might be less expensive to replace the faulty cell (and perhaps any other weak cells) than to replace the whole battery pack.
 
I have always thought the 'ideal situation' would be to drive the car for 3 or 4 years and 40 or 50K miles and then experience a cell failure and let Mitsu give you a brand new battery pack . . . .

But then, how do we really know if the replacement is 'brand new' or (more likely) a 'reconditioned' pack that Mitsu puts together with previously used cells from other faulty packs??

Don
 
Back
Top