Battery cell failing at just 2900 km (1800 miles)

[Gyver]

I just bought a Citroen C-Zero (a rebranded imiev) here in Italy. It was used as a demonstrative vehicle in 2012/2013 (exp. date of its last insurance was October 2013) and then it was left alone for 2 years. When I tried it at the dealer it looked absolutely new, even the tyres. It was manufactured on April 2012 so I knew it should have been fitted with a 16 kWh battery pack (unlike new C-Zeros that are sold with a 14.5 kWh pack) It was perfect on the road, I got 1 year full guarantee from the dealer, and the battery pack was guaranteed by Citroen for 5 years starting from registration date (October 2012), so at 12k euros I had no hesitations and I bought it.
As soon as it arrived at home (22nd december, a nice Christmas present!) I plugged in and fully charged it. I jumped in to have my first ride in my new car... but the gauge was showing just 12 bars instead of 16, and the RR displayed just 82 kms left (50 miles). At first I thought it could have been because the battery was standing unused for 2 years and needed some cycling, so I drove till 2 bars were left (the charge only lasted for 45 kms / 28 miles) and charged fully again. Same story: 12 bars and 82 kms left on the RR gauge. :evil: This time I tried do discharge the battery completely, I wanted to see the turtle light on my dashboard. Instead, as soon as I had 1 bar left the EV warning symbol showed up. I slowly drove back home and called Citroen.
Now my car lies at the dealership, they told me they probably have to change the battery pack (they also told me the car needs to go through 3 different recall repairs...)
On one hand it is nice to know I will have a brand new pack (made with the new lev50n cells), on the other hand I will receive a 14.5 kWh pack, so a lower range for my car.
I am sharing my experience because I find interesting that even a battery pack with such a low mileage can fail. I have no information on how the car was stored during these years, but considering the mint condition of the body and tyres I suppose it was kept inside.
Unfortunately I haven't received my OBD interface yet, so I have no CaniOn data to share.

 

[FrostyCanada]

Hello Gyver, sorry for the trouble you're having with your new little EV.

On a positive note you'll most likely be getting a brand new battery pack and they'll also update your car with the latest recalls.

You'll be getting a new car back, but without the new car smell

Please pe keep us posted,
Brian

 

[JoeS]

Gyver, welcome to the forum. Actually, congratulations that you will be receiving a new pack.

I would be surprised if your pack is replaced with one of lower capacity as, if I recall, the lower-capacity pack has a different cell count. A reprogramming would be in order, which may or may not be trivial. Could someone verify this? I think you can make a case that you should receive a replacement of the same capacity, and it would be worth making a fuss over this. If they won't do it I wouldn't worry too much about it as a 10% reduction in capacity will not be noticeable in most peoples' everyday driving.

As to the cause of the original pack's demise: few people know about the proper care and feeding of our batteries - if a car is left unused for an extended period of time, it should be left charged at around 25%-40% SOC (exact number is open to debate) and the 12v battery left on a float charger. Unfortunately, I've seen cars sitting in dealers' lots fully charged in the broiling sun, which is not good for Lithium battery longevity.

 

[Malm]

Another from 2012 with a fail. At least, we know six of them from 2012 with battery failures. From 2011 there is no record of these failures, and there are many in Australia and Europe. Mine is one of them. This can be because a change of the cells, from LEV50 to LEV50N, that took place in 2012.

 

[tigger19687]

I thought the other 2012 failures were from early early 2012 and late 2011?

 

[databeestje]

Well, my 5 year old 2010 definitely had LEV50 Cells, and that had a cell failure too.

You see, it's all about statistics. Put 100 cells in series of 99,993% reliable, and you end up with a 83% reliable string. There is a reason Tesla went with a 96S74P battery, a single cell will reduce the total capacity by 1,3% without failing the pack entirely.

All the other electric cars, i-Miev included, fail entirely when a single cell fails. That's bloody annoying, and I know they did this because it's easier to engineer and build, but it's not better

If you ever replaced the single LEV50 Cells with a similar 18650 construction you would end up with something in the order of a 88S20P which would mean a 5% hit on capacity, but you could still drive and charge. As you can see from these numbers, the difference between 20 or 74 parallel isn't particularly huge, it falls off quite fast, but the benefits are tangible.

 

[PV1]

There was also a process change at the factory as a result of the Boeing Dreamliner incident. There was a faulty screening process that was contaminating the cells, causing internal failures. I'm not sure if it is related or not, but the change happened fairly close to the switch from LEV50 to LEV50N cells.

The LEAF pack has a single string of modules, but each module is made up of 4 parallel cells, so a cell failure in the LEAF would take a 25% hit at the most. I'm not sure of the construction of the others, but Tesla's 18650 design definitely takes the cake for redundancy :lol: , not to mention energy density.

 

[Malm]

Well, my 5 year old 2010 definitely had LEV50 Cells, and that had a cell failure too.

You see, it's all about statistics. Put 100 cells in series of 99,993% reliable, and you end up with a 83% reliable string. There is a reason Tesla went with a 96S74P battery, a single cell will reduce the total capacity by 1,3% without failing the pack entirely.

All the other electric cars, i-Miev included, fail entirely when a single cell fails. That's bloody annoying, and I know they did this because it's easier to engineer and build, but it's not better

If you ever replaced the single LEV50 Cells with a similar 18650 construction you would end up with something in the order of a 88S20P which would mean a 5% hit on capacity, but you could still drive and charge. As you can see from these numbers, the difference between 20 or 74 parallel isn't particularly huge, it falls off quite fast, but the benefits are tangible.

I wouldn´t expect no fails in the LEV50. But the question for me is if the fails are much more frequent in the 2012 i-MiEVs then in the others. For now, just an hypothesis.

 

[Gyver]

Hi guys, I finally received some news from the dealer. They confirmed the battery pack is damaged and has to be replaced.

I went on chatting a bit with the chief mechanic, he told me 2 cells fail to charge: number 6 and 12. He is unsure if there is a pack in stock at PSA in France, if that is not the case they will have to ship it from Japan. They will change both the battery pack and the ECU.

Since they will probably need quite some time to receive the new parts, they will return me the car so that I can drive around with a limited range. Well, I think I'll take some pleasure in trashing that doomed battery in the next days. :twisted: Drag-racing, anyone?

______________
2012 white C-Zero
2008 Fiat Multipla "Natural-Power" (Compressed Natural Gas)

Driving CNG since 1-18-2000.
Driving electric since 12-22-2015.

 

[PV1]

First I've heard of a confirmed pack with two cells going out. Hopefully you get a new one soon.

What you may start to notice is that after every charge, your range will continue to decrease. When mine was going out, a full charge was 4 bars, 20 miles RR by the time I got the car to the dealer. They had my car from 5/26/15 to 6/19/15 (dates dropped off and picked up).

 

[databeestje]

Hi guys, I finally received some news from the dealer. They confirmed the battery pack is damaged and has to be replaced.

I went on chatting a bit with the chief mechanic, he told me 2 cells fail to charge: number 6 and 12. He is unsure if there is a pack in stock at PSA in France, if that is not the case they will have to ship it from Japan. They will change both the battery pack and the ECU.

Since they will probably need quite some time to receive the new parts, they will return me the car so that I can drive around with a limited range. Well, I think I'll take some pleasure in trashing that doomed battery in the next days. :twisted: Drag-racing, anyone?

I'm not sure how many cells failed in my battery, I never got a read out as it was serviced by a dealer quite a bit away. Mitsubishi has just 3 dealers in NL that can service high voltage.

They told me in september that there were no batteries in the EU so it had to be shipped from Japan. It took 2 months before the car was returned, they lent me a Mitsubishi Spacestar during the time, but it depends on the dealer.

 

[Gyver]

I'm not sure how many cells failed in my battery, I never got a read out as it was serviced by a dealer quite a bit away. Mitsubishi has just 3 dealers in NL that can service high voltage.

They told me in september that there were no batteries in the EU so it had to be shipped from Japan. It took 2 months before the car was returned, they lent me a Mitsubishi Spacestar during the time, but it depends on the dealer.

I guess I willl have to be very patient :roll:
when did you had your car serviced?

 

[databeestje]

The car was away from the middle of september to the middle of november, exactly 2 months. The good news is that I now have a new battery, although the old one wasn't that bad after 5 years. It still had 90km even without completing the charge (it stopped before the balance stage where the charge power is reduced ~90%).

I now have 120km range without the heater and 90km with the heater on the 2nd tick.

 

[Gyver]

What you may start to notice is that after every charge, your range will continue to decrease.

Oddly enough, after two weeks of intense use I now have 2 bars more on the battery gauge when fully charged and the car has more range. at first I had 12 bars when fully charged, now I have 14. the RR is gone from 82 km left to 95 (I know this is only indicative, but I also have more true range on the road). In these 2 weeks I experimented a lot and drained the battery completely most of the times (more than 10 kms on turtle), It is like one of the damaged cells had woken up with all these daily profound discharges after 2 years of sitting without charging / discharging.
Meanwhile the new battery pack has arrived. The dealer is waiting for "some pieces of electronic" (ECU?) that should be available next week, so my experimenting with the damaged pack is nearly over.
Unfortunately I discovered my obd reader is based on ELM327, so no CanIon data for now. I placed an order for an OBDLink MX on ebay, but since it is coming from the US (couldn't find it in Europe) I doubt I will have it before the battery swap.

 

[PV1]

That's a shame about not having CaniOn data. This is starting to sound like a lazy cell or two that simply fell out of balance, and your repeated use and charging (driving below two bars then recharging) has managed to bring the cells back into balance. I may try this with Koorz next summer, as I notice SoC sometimes jumps after parking the car for a few minutes, although this is all cells simultaneously and Koorz fully charges.

Keep on driving, as it now sounds like a challenge to get all bars back before getting a new pack.

 

[Barbagris]

Unfortunately I discovered my obd reader is based on ELM327, so no CanIon data for now. I placed an order for an OBDLink MX on ebay, but since it is coming from the US (couldn't find it in Europe) I doubt I will have it before the battery swap.

I think the icarSoft i620 can do the job. Is listed at amazon Italy for (total) 45€, and in a very few days you can run canIon.

If don't work, I suppose you can return it to Amazon...

 

[priusfan]

.......
Unfortunately I discovered my obd reader is based on ELM327, so no CanIon data for now. I placed an order for an OBDLink MX on ebay, but since it is coming from the US (couldn't find it in Europe) I doubt I will have it before the battery swap.

The right device for canion is the OBDLINK LX (not MX): Bluetooth is cheaper and more stable than Wifi with canion....

 

[kiev]

Make sure that it's a real LX from scantool and not a counterfeit scamtool being sold on ebay--there are some out there that 'look' like the LX but are using elm chip internal and won't work.

 

[Gyver]

Make sure that it's a real LX from scantool and not a counterfeit scamtool being sold on ebay--there are some out there that 'look' like the LX but are using elm chip internal and won't work.

This is the one I bought:
http://www.ebay.it/itm/171651829835?_trksid=p2057872.m2749.l2649&ssPageName=STRK%3AMEBIDX%3AIT
100% positive reviews and claims he is an authorized dealer.. I really hope this isn't a scamtool!

Barbagris what a pity I didn't read your post before, anyway tracking says the device is already in Italy, waiting for custom clearance, if the bureaucrats don't mess things up it should be here in some days. Meanwhile the Citroen dealership has called me today postponing the "operation" to the 23rd february, we'll see who comes first!

Now after 1 month, 2000 kms of deep discharging (riding the turtle till limp mode most of the time) and full recharging cycles It now shows 15 bars and 111 km / 105 km on the rr (depending on how I drove the day before) which is quite different than 12 bars and 82 kms I had when I received my c-zero

 

[PV1]

Hopefully your OBDLink gets there soon. This is an interesting development, and appears to be showing the balancing system at work, bringing a cell back into line with the others.