First Post - Looking for experienced guidance on purchase

First, I've put a ton of research into this purchase. We are going to purchase a CPO i3, I would prefer a 2016, actually I would prefer a 2017 but those are even more rare, but we are looking and comparing 2015 and 2016's. We live in south florida, so it's always nice, in the summer I would say in the sun it's brutal and I'm wondering if that is impacting i3's down here that are left out in the sun all day.

Tthis is really a question of how do you gauge the "value" of a used i3? For example, for a typical ICE car, you have the price, and you can look around the engine/trans/suspension and get a general feel for how the car has been treated. Some cars you can instantly tell were not well cared for while others just show zero abuse/use and therefore you use this to make a decision on value.

I'm doing this for the i3, but I have to say the most important factor has to be battery health. I'm trying to leverage the Batt.Max.Kapa value on each as a general indicator on battery health, which appears to be the most important factor to consider when looking at a used i3 (correct me if I'm wrong).

We have narrowed our search to two 2015's, one with a Batt.Max.Kappa value of 18.8 and the other with the value of 16.4, does this "GENERALLY" mean that the lower value should be valued at a lower price?

Will it's range be less?

Based on my calculations of 3.5 miles per kWh (which is what I'm finding), the two cars ranges based on these values are roughly 65 miles vs 57 miles, correct?

I've been educating the salesman, and his boss with this information slowly as we have been negotiating and I'm wondering if I should purchase the one with the value of 16.4 if I can get it for 10-20 percent less than the one at 18.8, OR will I hate myself for limiting my already somewhat limited range?

These are both REX versions and my wife's daily commute is mostly highway with a rough round trip, (with errands) of 47 to 65 miles, based on my calc's it appears that either would work with limited use of the REX.

I'm really looking for others opinions here, hopefully with facts, but I'm more than open to accepting anyones opinion.

Both cars have around 20k miles and the lower kappa value car has ALL the options and is a nice ORANGE color with leather and EVERYTHING ELSE.

Unfortunately, a single Batt. Kapa. max reading is almost meaningless. On our 2014 BEV with just under 7k miles, Batt. Kapa. max values have varied from ~16 kWh to ~20 kWh with no up or down trend obvious.

Fortunately, refrigerant-cooled i3 battery packs have degraded quite slowly in general. BMW guarantees that battery pack capacity won't degrade more than 30% in 8 years/100k miles. So I wouldn't consider battery pack health to be the most important factor were I considering purchasing a used i3. The i3's battery management system makes it difficult for a driver to abuse the battery pack enough to cause rapid degradation. The degradation amount is probably well-correlated to the distance driven, so concentrating on a low-mileage i3 would probably be the best way to find an i3 with a strong battery pack.

I beg to differ yet. I may agree with the statement that Batt.Kapa.max doesn't matter much in the future once I collect more stats, but based on what I've seen so far, it's a good indicator of the current battery condition. So all other things being equal (and even substantially unequal :roll I would go for a higher Kapa.max number.

I'm from Ukraine, and I imported my 2015 BEV from the US in May as well as other people here, over 150 cars during this year already. We don't enjoy BMW warranty, and the battery pack degradation is our risk, so naturally we pay a lot of attention to Batt.Kapa.max and try to closely watch the degradation of the pack with time and mileage. We also actively communicate in our local forums to share data. The degradation is there for most other people cars I've seen, from 1% to 15% for 2014-2015 i3 models based on Batt.Kapa. It usually closely correlates with the available range reported by owners, both estimated and real. That's why I think that this indicator is quite robust.

Mileage is a good predictor of the battery capacity as well, but its strength seems to be lower than Kapa.max.

My BEV's current mileage is 20K, I've covered 4K of these since the end of May, and have everyday stats logged in. I track the battery condition via:
- Batt.Kapa. It's been quite consistent, in the range of 18.3 to 17.7, median and average of 18 and a slight downward trend. That's what I'd expect.
- Batt.Ladung divided by the SOC. This one varies significantly, from 15 to 21, median of 17.5 and average of 17.2. The indicator depends a lot on the SOC, the estimated capacity being closer to Batt.Kapa the higher the SOC. Some extreme high values are possible at very low SOC levels as well (I got 21 kWh at SOC of 10.5% and Batt.Ladung of 2.2 kWh, probably because of rounding).
- Battery capacity estimate calculated via trip computer data. The variance of this output is very significant as well, probably because the trip computer doesn't calculate data accurately and/or consistently. The range here is 12-19.6, median is at 18.7 and average at 18.2. The estimates seem to be significantly lower when I use mid or low SOC levels during my daily commute (e.g. starting at 80% SOC and down to 40% rather than from 100% to 60%), and also at times when my efficiency (in miles/kWh) is lower.

I'm leaning toward believing the value has merit, don't know exactly how much, but I can say after several days of research and reading that the car with the reading of 16.4 has less battery health than the car with the reading of 18.5.

Where I begin to second guess this decision is based on the fact that this car has been sitting in their used car garage, in the south florida heat for sale since early June, so almost three months without regular use which could have impacted this reading and the battery was nearly drained when I tested it.

I like the car and think I can get it for around 19k vs 21, on the other, the 18.5 car has Giga leather/cloth interior and I'm just not sure about the cloth and resale and smells/stains, I'm sure that is much to do about nothing.

If I get the car for a deal and end up hating it due to range issues I will regret it and I'm assuming range is the number one issue here.

Even in it's current state of approx 13% degradation it's not likely to reach the 30% needed for warranty and it's not something I should be planning for anyway.

So thank you for taking the time to respond, I think I've made my decision.

Oleksiy said:

- Batt.Kapa. It's been quite consistent, in the range of 18.3 to 17.7, median and average of 18 and a slight downward trend.

Click to expand...

Are you reading Batt. Kapa. max under identical conditions each time (e.g., same battery pack temperature, same SOC, etc.)? Maybe because I don't try to read Batt. Kapa. max under identical conditions each time, I have seen values as low as 16.1 kWh and as high as 20.1 kWh even though the battery pack temperature doesn't vary much (ambient temperatures don't vary much here in the tropics during the year and our i3 is never driven at high speeds for significant distances on our small island). It's 17.5 kWh today, but it was 16.7 kWh a week ago. During this past week, it hasn't been charged to full, so the battery cells' charge levels haven't been balanced which would have increased capacity if significant imbalance existed. I don't drive on the same routes regularly, so it's difficult to determine whether our range has decreased over almost 4 years of ownership. I can't make any sense of the wide range of Batt. Kapa. max values displayed in our i3.

There must be a way to use Batt. Kapa. max to judge actual battery pack capacity with some precision, but I haven't discovered how to do this.

loneranger said:

I'm leaning toward believing the value has merit, don't know exactly how much, but I can say after several days of research and reading that the car with the reading of 16.4 has less battery health than the car with the reading of 18.5.

Click to expand...

Our i3 had a Batt. Kapa. max value of ~16 kWh months before it displayed a series of values ~20 kWh over several weeks. It had received a system software update in the meantime, so maybe different system software versions calculate Batt. Kapa. max differently. We just don't know.

I think you risk excluding an i3 with just as good a battery as another i3 whose Batt. Kapa. max value happened to be higher at the time that it was read. I would certainly not consider a one-time difference of 2 kWh between two i3's to be significant. If each of these i3's displayed a series of consistent Batt. Kapa. max values over several months that had an average difference of 2 kWh, I would consider that to be significant and meaningful.

Art, all very good points and valid, however, in the used car market we simply don't have access to this information and I have to go with current readings. Eventually these salesmen will figure this out and have the cars in a state that shows the battery health, but I doubt too many people are as research intense when buying....my wife thinks I'm a bit obsessed.

If you know absolutely NOTHING about the type of car your buying, how can you go into a purchase blind is my thinking. I really appreciate the trailblazers on this forum, it's a treasure trove of information which I've utilized as baptism into the i3, so thank you very much.

My wife ended up ending the decision making process last night while I was into hour 3 in front of the computer and we eliminated the orange one, using her quote, "it's too much like a dude car" and it's out so now we just have to close the deal on the other one.

It's the last couple days of the month, they are ready to deal, so I suspect we will be the new owners of a 2015 CPO i3 soon, thanks again.

alohart said:

Oleksiy said:

- Batt.Kapa. It's been quite consistent, in the range of 18.3 to 17.7, median and average of 18 and a slight downward trend.

Click to expand...

Are you reading Batt. Kapa. max under identical conditions each time (e.g., same battery pack temperature, same SOC, etc.)?

Click to expand...

I've been logging Batt.Kapa at different conditions having read your previous posts on the subject and trying to find patterns if any. These are the conditions for the data set I have:

- Ambient temperature at reading. Range - 55-88F, average and median - about 74F
- Battery pack temp at reading. Range - 63-86F, average and median - about 80F
- SOH at reading. Range 100%-11%, average 49%, median 51%

The Batt.Kapa doesn't seem to be influenced by any of these and has varied within the mentioned range (17.7-18.3) randomly. It also doesn't seem to be affected by the balancing or non-balancing of cells. I used to have very low values during weeks of everyday charging up to 100%, and alternatively readings north of 18 during periods of consistent 80-90% SOC after overnight charging. But I haven't done any analysis of the data yet, will get to it one day, probably once I collect more data points.

The other two methods (Batt.Ladung and estimate based on the trip computer data) seem to be very affected by SOC, but not by the ambient or battery pack temperature.

We are approaching cold season shortly, will go into subfreezing territory as early as in October-November, so it would be interesting to check whether more extreme ambient temperature flluctuations change things.

To add to the above, I checked monthly Batt.Kapa averages in my pivot spreadsheet, and it looks lake the capacity went down from 18.1 in May-June to 18.0 in July and to 17.9 in August. Unless it's an aberration or something, at this rate I should expect about 5% degradation every 20K miles. I.e. I still have slightly over 100K miles to drive until I reach 70%. At this time the car will have become unusable for my daily commute, even in summer.

Oleksiy said:

It also doesn't seem to be affected by the balancing or non-balancing of cells. I used to have very low values during weeks of everyday charging up to 100%, and alternatively readings north of 18 during periods of consistent 80-90% SOC after overnight charging.

Click to expand...

To what do you attribute the variation between the very low values and values north of 18 kWh?

Some English i3 owners met with BMW engineers earlier and apparently were told that Batt. Kapa. max is computed from the amount of energy added during charging. However, when you added more energy by charging to 100%, Batt. Kapa. max was lower, and when you added less energy charging to 80-90%, Batt. Kapa. max was higher. Seems like the algorithm used to calculate Batt. Kapa. max must be flawed or maybe is only accurate when charging to full which I rarely do.

For about 2 weeks recently, I charged and left our i3 at full charge with our OUC connected so that it could increase the charge level should it drop due to cell balancing. That should have balanced the cells at least partially, yet that's when our Batt. Kapa. max values dropped to the 16 range whereas when I charge to 80-90% which is typical, our Batt. Kapa. max values are in the high 17's. Seems similar to your experience which is difficult to explain.

I still lack data to answer your question with any confidence. That's in case the variation of Batt.Kapa is not random within some range.

E.g., I have 4 data points with readings of 17.7 and 17.8, two of these occurring after SOC beginning of period of 100%, and two - at about 90%. The ambient temp varies from 65F (one outlier) to 75F (three readings), the pack temp - 75-82F. I have 8 data points of 18.3 and 18.2 readings, these could be connected to high BEP SOC (6 times 100%; 95%, 97%). The ambient temperature is in the range of 71-81F, so not materially higher. Same for the temp of the pack - 79-84F.

I also added linear trendlines to see if the variables I log have any predicting power for Batt.Kapa, and it seems none could be used to predict the variance of Batt.Kapa yet, at least based on the pool of data I have. The highest R-squared of 0.08 is for ambient temperature, slightly lower for the pack temperature, 0.016 for SOH at reading and 0.02 for SOH beginning of period, 0.02 for efficiency in km/kWh, 0.03 for estimated battery SOH based on trip computer based calc, 0.0036 for battery capacity in kWh calculated based on Batt.Ladung from secret menu.

If BMW engineer is correct, Batt.Kapa should be a good indicator of battery health since we are all interested in the amount of energy that was added to the battery during charging sessions, all else being equal. It also means that I'll start to see declining Batt.Kapa readings in winter, and may see some recovery next summer minus normal degradation.