Unfortunately near the bottom of the pack.
What used EVs preserve their batteries the most? What about the ones with the most degradation? Check the best and the worst here.
It would be nice to know what their methodology is. I went to the source of the original article here and then went to the company that gathered the data here and there is no mention of exactly how they determined battery degradation. How many examples of each model are in the data set? What's the average mileage? Are the majority of charges starting from a completely depleted battery? That's likely more common for PHEVs than EVs give the smaller packs on the hybrids. EV drivers likely top up more often and only occasionally plug in with a flat battery.Unfortunately near the bottom of the pack.
What used EVs preserve their batteries the most? What about the ones with the most degradation? Check the best and the worst here.insideevs.com
I thought about the buffer thing, but I'm not sure how you would measure the buffer being consumed without advanced diagnostics. The data presented seems to indicate actual reduction in usable capacity that would be noticed by the driver.While unknown, the Niro PHEV almost certainly has an unused "buffer" of over 10%. Until that is exhausted which will likely take around 5 years (depending on number of charge cycles), owners will not be able to measure any reduction in range. Every owner has had degradation - zero percent quoted for the Bolt is just not possible. Maybe in an alternate universe with different rules of physics!
Assuming the site's numbers are accurate, a big if, I think the most likely source of problems are batteries. They are so scarce that Korean manufacturers may have sourced second quality. The various algorithms supporting lithium battery life are well established so that is an unlikely source of early degradation. Of course, vehicles such as the Leaf with 100% passive cooling will have issues.
The purpose of the buffer is to mask battery degradation, not provide a generally usable reserve. The battery management system uses the buffer to maintain the rated capacity as the cells age. Allowing the reserve to be used during normal driving would make the reserve less effective in preserving performance over time.I would agree that it would make sense to have a buffer at the upper end. However, if they do have one the car isn't permitted to enter it. If I leave my garage and head down the hill, I have roughly 1/2 mile to the bottom and a stop sign. If I leave the garage with a 100% charge (which I almost always do), the ICE will fire just over half way down to add compression braking. I have to coast down the hill in neutral to avoid the ICE starting. If there's a buffer, why not let the regen continue all the way down? What good is a buffer if you can't use it?
That's not the primary reason although I'm sure the marketers love it. Primary purpose of buffer is to reduce degradation by keeping battery neither 100% or 0% charged. Sweet spot is middle 80% for long life. Well, really the middle 60%, but that would only be tolerated by hybrid owners (and we do).The purpose of the buffer is to mask battery degradation, not provide a generally usable reserve.
I agree that the data is confusing. If @yticolev is correct when he says:Jim, that would seem logical, but then why would the car show a 3.5 degradation after one year? If there's a buffer to combat any aging, it shouldn't show any degradation for several years I would think. I mean, what you say makes sense, but it conflicts with what this report appears to show.
We would not expect to see any drop in performance for several years. That's not the case with this dataset as there is a linear downward trend from brand new in the overall data. What does that say about the presence of buffer capacity? We all know that the battery management systems in EVs don't charge the packs to 100% of capacity in order to prevent premature aging. That's not a buffer, that's just the equivalent of the head space left in a gas tank to allow for expansion with temperature. A true buffer would actively mask degradation by adding capacity as the pack ages. For example, if the BMS charged the pack to 80% when new it might charge it to 83% after a year to maintain the effective capacity assuming a 3.5% yearly loss. On further review I don't think that's what's happening.While unknown, the Niro PHEV almost certainly has an unused "buffer" of over 10%. Until that is exhausted which will likely take around 5 years (depending on number of charge cycles), owners will not be able to measure any reduction in range.
That would explain why the individual model trends stay flat at 100% for a while then start dropping.Another issue with this test is that none of the cars were tested new. Thus it is not possible to know actual starting capacity, only calculated capacity. My new laptop and cellphone batteries are plus/minus 5% of rated capacity.
Agree 100%. While the data is interesting I agree that there are clear questions about how and what they actually measured. I plan on keeping my PHEV for the long haul and am not overly concerned about the issue. I ran some numbers and even assuming the 3.5% / year drop claimed it would take 9 years for the range to drop by 25%. I plan to be retired by then so my commuting days should be over. Maybe they'll have a nice ragtop BEV I can slide into by then!Bottom line, don't take this report seriously. I seriously doubt most owner will see any actual range reduction for many years (of course degradation must occur, it just will not affect range due to the software buffered portion of the battery). Many will have traded in before it happens based on normal consumer habits. For the same reason, don't take the displayed range seriously either. Even the actual range will vary due to a number of other variables long before degradation becomes apparent.
I've been able to observe (using EVNotify) that the SoC displayed and the SoC BMS (Real SoC) are never the same. SoC displayed is higher by about 5% at the top end and about 3% lower at the bottom - this is well known for the IONIQ 28 kWh BEV. The cross over is at about 22% SoC. This is the buffer that Hyundai has built into the IONIQ and why our batteries appear to last so long. (They don't ever get fully charged and the BMS also starts using the buffer once it detects any degradation, so your range doesn't drop until you have at least 8% degradation).
AFAIK, the easiest and best way for a layman to determine degradation is to charge to 100% and drive the car as low as you can and record the total energy expended - Torque Pro, EVNotify, etc, you can then proportion this value to calculate the theoretical energy available if you expended 100% of the charge. Then after some time (6 months or a year) perform the test again. If there has been any degradation, then the energy the battery can release will be less.
I think perhaps you are misinterpreting the numbers. It’s also an open question in my mind if it’s safe to extrapolate the numbers for the 2017 HEV to your 2020 PHEV, but let’s assume that it is (aside from the battery size, which we know is larger in the PHEV).I been tracking cell voltage on my 2020 PHEV and what i been noticing is that there isn't a big of buffer at the top end when fully charged.
When the car is fully charged to 100% the cell voltage is at 4.15v. So there is only 50mv of buffer to real 100% at 4.2v.
When the car is fully discharged and where the hybrid engine kick in at 16%, the cell voltage reading is 3.5v. So there is 1v of buffer to the real 0% at 2.5v.
The 4.2V and 2.5V information i got from the tech site below.
So its possible that the top end buffer of 50mv is not enough, and causing more stress when the battery is sitting at 100% constantly. This may be the reason for the degradation.
Personally i been charging my car to 88% and that charge the cell is reading 4.00v.
While Level 2 is often cited as the optimal way to charge an EV, the difference in battery health between cars that routinely charged on Level 2 as compared to those who used Level 1 appeared to be observable but was not beyond the level of statistical significance.