Charge Strategy for Max Battery Life

Discussion in 'Clarity' started by Steven B, Apr 18, 2018.

To remove this ad click here.

  1. Steven B

    Steven B Active Member

    The goal for this thread is to be a listing of links to actual science rather than posts of armchair quarterbacks. Prefer short summaries and then actual links related to science-based battery charging strategies to maximize long-term battery life.

    Background: In the early days of NiCad batteries the guidance was to fully drain before charging, else the battery would have a 'memory' of its new lowpoint and thus shorten its energy providing capacity (duration) per charge.

    I am aware that Lithium-based batteries do not have this 'memory' behavior but I know it is common knowledge that modern Lithium-based batteries have a maximum number of charge cycles as do NiMH-based batteries.

    So, I'd like to see the science that shows that repetitive charges starting from somewhere between 50-80% SOC (State Of Charge) results in lithium-based batteries that have a longer life (total number of charge cycles) versus holding off on charging until SOC has dropped to 20-30%.

    One scientist says modern lithium batteries have a 2500-3000 cycle life. Does the science say a charge starting at 50% is ONE cycle just as a charge starting at 20% is ONE cycle? If that is what the studies show, then draining down to 10 or 20% before charging should be better than daily charging from 60-95%, resulting in a battery that lasts 15 years rather than 8.

    If you have come across science-based studies of this elsewhere, please post links here. Also, not sure how relevant Prii with NiMH batteries would be to Lithium's behavior, but if you find anything on those, it might be all that is available since Lithiums have only been in cars for about 10 yrs.

    Not much interested in anecdotal data such as "I've got a friend of a friend who has X hundred thousand miles on his/her ______ and still has the original battery".

    Thanks.
     
    Daniel M W and Kendalf like this.
  2. To remove this ad click here.

  3. rodeknyt

    rodeknyt Active Member

    Most of the information about lithium batteries centers around cell phones and tablets, but from everything I've read (can't provide links but some were actual battery gurus), a charge cycle is based on percentage. In other words, a battery that is charged full from a starting SoC of 50% represents half a cycle.
     
  4. Steven B

    Steven B Active Member

    This study implies a cycle of 10%-90% SOC is common:
    http://mat4bat.eu/wp-content/uploads/2014/03/Cycling-Aging-of-Lithium-Ion-batteries_Grolleau_EIGSI.pdf

    Check out slide 25 (no number). Lesson from this slide: Charging your new car in in colder temps (41F shown) may permanently stunt its battery capacity during the first several hundred cycles. Warm temp charging eventually drops to similar capacity levels.

    SO, Based on Honda's stated battery capacity (is there any reserve in addition to this?) and accurate charge data that others have posted, we know the usable capacity is somewhere around 82-85% of theorectical capacity. But do we know if there is reserve at the top end as well as bottom end (can the battery only be charged to its theoretical 95%, for instance)? I think we do not know for certain the specific reserve at the low end and whether there is any reserve at the top end. We know what the app says and what the charge bars say, but those are software-based and can be skewed however the programmer wants: display characters "100%" and shut off charger when theoretical SOC reaches 95%; display "11%" and switch to Hybrid mode when theoretical SOC reaches 16%. I believe Honda states that the car will no longer run in EV mode when the capacity reaches 11%. We could take their word that this is 11% of theoretical and then based on the usable capacity data, determine that there must be __% of reserve at the top end. This calculation might indicate something around 5% is in reserve at the top end, so indeed the programmer would code the various displays to report 100% and shut off the charger when theoretical reaches 95%. I feel confident that Honda is keeping the battery away from 0% and 100% for battery health and safety but from a lithium waste disposal perspective, it would be helpful to know what specific charge methodology will keep my car running on the same battery for 15 years (probably long after I sell it to someone else) instead of 8.
     
    AGarg likes this.
  5. AGarg

    AGarg New Member

    I don't have any scientific evidence but this is what the owner's manual states on page 466:

    The High Voltage battery gradually discharges even if the vehicle is not in use. As a result, if your vehicle is parked for an extended period of time, the battery level may get low. Keeping your vehicle’s battery level low can shorten the battery life. To maintain the battery while the vehicle is not in use, recharge the battery at least once every three months.
    The High Voltage battery life can also be affected by ambient temperature. In particular, when it is cold outside, the vehicle’s driving range on electric power can be reduced, and a longer battery charging time is required. In addition, parking in extremely hot or cold environments can accelerate battery drain.

    To help extend the lifespan of the battery, it is recommended that you fully charge the battery each time prior to driving.
     
    Texas22Step likes this.
  6. Viking79

    Viking79 Well-Known Member

    Check at scholar.google.com, tons of studies.
     
  7. To remove this ad click here.

  8. AGarg

    AGarg New Member

  9. Steven B

    Steven B Active Member

    Apparently we should watch the video linked forum to learn how best to treat our battery: https://teslamotorsclub.com/tmc/threads/why-do-li-ion-batteries-die-and-how-to-improve-the-situation.27109/

    Also, this is seven years old but these points may still be valid: http://www.plugincars.com/eight-tips-extend-battery-life-your-electric-car-107938.html

    Sure would be nice to get some straight answers from Honda on what the top end and low end reserves are. Plus, we should ask for this feature improvement to the App: Under the Charge Schedule feature give the user the option to Stop charging when battery is 80% Full.
     
  10. jane smith

    jane smith New Member

    Steven B -

    I have a 2013 Leaf which allows the 80% charge. The EPA wanted to use the 80% figure instead of the 100% to calculate mileage so Nissan removed that feature in later models. While it would be great to have, I doubt Honda would want the EPA to penalize them as well.
     
  11. Steven B

    Steven B Active Member

    Yeah, not asking them to further restrict the available energy, just to confirm what the current restrictions (reserves) of theoretical energy are. The high voltage battery will not drain below x% at low end and will not charge above yz% at high end. We all can agree that the AVAILABLE energy to power the car [(yz-x) x 17hWh= others have said this is approximately 15kWh] is what the EPA based the EV range on, not on the theoretical battery capacity. I have seen the low end percentage as low as 9%.
     
  12. To remove this ad click here.

  13. Viking79

    Viking79 Well-Known Member

    My recommendation is drive the car how you want to drive the car, don't worry about things like battery degradation. The car is going to depreciate rapidly regardless of how well you have cared for the battery.

    From what I have researched, two primary factors effect battery degradation:
    Calendar Aging
    Cycle Aging

    Your battery will go bad sitting stored for a long time, your battery will go bad if you cycle it a lot.

    Some studies:
    https://www.nature.com/articles/srep12967
    http://jes.ecsdl.org/content/163/9/A1872.full
    http://jes.ecsdl.org/cgi/reprint/164/1/A6066

    Cycling cold at high currents is bad for Lithium-ion
    Storing at high temperatures and high SoC leads is bad
    Cycling at low depth of discharge will give you more cycles than at high depth of discharge
     
    insightman likes this.
  14. M.M.

    M.M. Active Member

    Thing about this is, the massive size of the Tesla battery packs make a huge difference. Assuming that these guys have a 100kWh Model S, and they're charging it to 100%, but probably not totally draining it before recharging (say they're getting down to 20%), even without anything special going on you'd expect the battery to only lose a percent or two per 100 cycles.

    And at >300 miles per full cycle, 200,000 miles is only about 650 cycles, plus the climate isn't cold, so 6% degradation sounds about right. It would be a bit worse if the car was chronologically older, as well (there is both time and cycling based degradation).

    PHEVs are more challenging, because a lot of users will mostly drain the battery pack every day during their regular commute, resulting in a lot more cycles per year than a Tesla (or even less-impressive BEV) driver would. Chevrolet's solution to this was to leave a fairly large amount of leeway at both the top and bottom of the battery, which increased the number of cycles and also hid the degradation (since it was all at the top end you weren't using anyway). Mine has 50K miles and is 5 years old and still has the full available capacity showing, though I don't know how much it's lost off the unavailable top.

    I don't know what the specs on the Clarity are, and haven't seen anything clearly published since it doesn't report kWh anywhere in the dash display, only "miles". It definitely reserves a pretty good buffer at the bottom (you actually see that on the display), but not sure about how full "full" is yet.

    Smartphones, incidentally, are treated more like PHEVs--many people fully charge and almost fully drain the thing every day, which can result in a 700 full-depth cycles in a couple of years and a noticeable decrease.

    Phrased more simply, 200K miles on a Tesla with a 100kWh battery pack is 650 cycles. 200K EV miles on a 18kWh Clarity is at least 3500 cycles.
     
    Last edited: Apr 26, 2018
    Steven B and PHEV Newbie like this.
  15. Pushmi-Pullyu

    Pushmi-Pullyu Well-Known Member

    It certainly would, but don't be surprised if Honda refuses to disclose the percentages. GM has obstinately refused to tell us just how much capacity is reserved in the Volt's battery pack, and in fact some GM advocates have even tried to claim that there isn't any reserve there! More plausibly, some claim the Volt reserves ~35% of the new battery pack's capacity, but again GM has never confirmed the number.

    A standard feature to charge to 80% would definitely be appropriate for a BEV, but the Clarity PHEV isn't a BEV, so Honda may be using a different strategy, as GM is with the Volt. It is quite noticeable that apparently nobody has ever reported a range loss in the Volt, so the advantage to GM's advertising in reserving a large portion of the pack's capacity is quite obvious. However, I've read lately that some older Volts have been showing a loss of maximum power, which may well be a sign of battery capacity loss.

    Or to put it another way: If you charge to what the Clarity's instrument panel says is 80%, that may actually be rather far from a true 80% of the pack's capacity... and I'm not talking about the erratic range numbers which several Clarity owners have reported their car displaying, either.
    -
     
    Steven B likes this.
  16. Viking79

    Viking79 Well-Known Member

    The Clarity PHEV uses approximately 11.5 to 12.5 kwh of the 17 kwh pack. This is easy to tell because it takes about 14 kwh from the wall for a complete charge. Figuring 80-90% efficient charging.

    PHEVs have to leave some on the top to prevent damage to the pack when the engine is running. My '15 i3 REx (replaced my Volt) does this by not running engine until it has dropped to 75% SoC (Technically it won't start the engine until 6.5%, but if in service mode or coded to allow higher SoC engine starting, 75% is the cutoff), the Volt and Clarity don't charge it all the way so the engine can run whenever.

    Given Clarity has 20 bars to represent about 12 kwh of power, they are probably around 1/2 to 2/3 of a kwh each. Meaning it keeps maybe a 1 kwh or 1.3 kwh buffer. They probably have a little more beyond that they don't use to protect the pack. This means maybe around 2 kwh on the bottom, 12 kwh usable, and 3 kwh on the top?

    This would be about 85% at full charge, seems reasonable and probably safe point where the engine running wouldn't damage anything and should be good for thousands of cycles. Many Volts have well over 100-150k EV miles now based on Voltstats.net, which is like 3000 or 4000 charge cycles.
     
  17. M.M.

    M.M. Active Member

    Being an engineer, this got way longer than I intended, so here's the executive summary/tl;dr version:

    Based on real-world tests and how a car works, you will probably best extend the lifespan if you:
    1. Charge your car every time you drive it, even if the pack is still half full (most important)
    2. Go easy on the accelerator (probably minor)
    3. Leave your car plugged in all the time, particularly if it's hot outside.
    Also: Your Clarity will probably lose about 1% of its capacity per 10,000-20,000 EV miles you drive it. You probably won't start to see the effects of this until you've driven at least 150K EV miles and/or owned the car for 5-10 years.

    Showing my work, hard numbers, and offering some citations to back it up:

    Finally found the article I was looking for. Battery University has an good write-up of experimental Lithium-ion battery capacity maintenance regarding a number of factors. This page has a bit more on discharge rates.

    Figure 6 on that first article is the most interesting (capacity loss at different max charge and max discharge levels), with figure 7 showing an extrapolation of those numbers. Table 2 provides some less specific information on capacity loss as a factor of just depth of discharge. Table 3 shows the permanent impact of storage temperature on battery capacity, and figure 4 in the second article shows the impact of discharge rate on capacity.

    The data is mostly for either small pouch-style Li-ion batteries or single-battery canister styles (which are the building block of most large battery packs), so the results aren't 100% applicable to high-quality automotive battery packs, but the fundamental chemistry is the same so the general conclusions apply. Likewise, the data is all from lab tests, and real-world batteries will not perform as well (more erratic charge/discharge patterns, less climate control, aging that isn't factored in, and whatever else), but the basic conclusions aren't affected.

    Takeaways:

    Doing a bit of math on the simplified table 3, the deeper you discharge a Li-ion battery, the greater the capacity loss. No big surprise. Discharging to only 80% gets you about a 20% increase in total energy you can extract from a Li-ion battery over its lifespan versus 100% discharge. 60% discharge gets you 50% more use than 100%. 40% depth of discharge (that is, charging when the battery is well past half full) will double the usable energy compared to 100% discharge.

    So, in vehicle terms, charge your car every time you drive it. This would not necessarily be true if the vehicle were being charged to 100% capacity, but indications are that the Clarity reserves 15% at the top, so that's not an issue.

    Figure 4 in the second article shows what happens at C1 discharge rate (that is, drawing enough energy out that the battery will be empty in 1 hour), versus C5 (that is, it'll be empty in 12 minutes), C7.5, and C10 (empty in 6 minutes). After 500 cycles discharged at a C1 rate, the battery has lost about 5% of its capacity. At a C10 rate, it's lost a whopping 30% after the same number of cycles. So, the faster a Li-ion battery is discharged, the faster it loses total capacity.

    Now, a vehicle has active cooling on the battery pack, so this actually won't be as big of an issue as it is with a free-air-cooled cell. While the Clarity doesn't report numbers, the older Volts do, so I can say that at around 60mph on a flat highway those use around 15kW. The Clarity is probably somewhat more, but that says that you're only drawing more than around a C1 or C2 rate when accelerating or driving up a hill.

    So, in in bottom-line vehicle terms, don't lead foot it. It probably won't make a big difference thanks to active cooling and the relatively short periods of time one is accelerating, but it's certainly not helping.

    Table 3 shows what happens at various storage temperatures. In short, the hotter the battery is, the faster it loses capacity. I think (though have not confirmed in the manual) that the Clarity will cool the battery if it's left plugged in and the ambient temperature is high. This would be a very good thing if you live in a hot climate. So with the caveat I haven't confirmed Clarity behavior, leave your car plugged in all the time if you live in a hot climate.

    Figure 6 and Figure 7 in the first article are the most interesting. Those measure and then extrapolate the capacity of a battery at various max charge and max discharge points. Skipping over the initial spike in capacity loss (which I'm guessing might be much less noticeable on a high-quality vehicle battery, if it exists at all), figure 6 shows:

    Used at 100% to 25% SoC, a Li-ion battery will lose about 1% of its capacity every 250 cycles. This is probably the approximate pattern that the aforementioned 200K mile Tesla charged to 100% every time was run at. If it was a 100kWh battery pack model, that would put it at around 900 charge cycles, so a 6% capacity loss is well within the predicted range (worse, actually, but some of that may be due to an initial dip, some is probably environmental factors, and some may be simple age).

    Used at 85% to 25% SoC, a Li-ion battery will lose about 1% of its capacity every 450 cycles. This is probably representative of a Clarity with daily driving of 30-35 miles total. This is probably around what a Clarity is allowing the user to use, based on Viking79's note on the total available capacity.

    (Showing my work, if 14kWh goes in at the wall, and we assume battery has a 88% charge efficiency and the AC/DC converter is about 97% efficient, that means there's about 12kWh usable, which is 70% of the total capacity, meaning around 15% at the top and 15% at the bottom are reserved. That very roughly agrees with the two bars left at which it kicks in the ICE. That would put full cycle capacity in the 85% to 15% band, if you fully emptied it every time you drove it.)

    Used at 75% to 45% SoC, a Li-ion battery will lose about 1% of its capacity every 830 cycles. This isn't realistic for a vehicle, but isn't all that far off from what you'd see if you didn't bother to plug your Clarity in if you only drove a couple of miles, and the rest of your driving was in the 20-30 mile-per-day range.

    So, let's that 85% to 25% SoC as a baseline, keeping in mind that the pack has to lose 15% of its capacity before it will actually impact usable range (I'm assuming here the Clarity does the same thing as the Volt and "hides" the capacity loss in the non-user-accessible top portion of available charge). The reports of Volts that haven't lost any capacity after 100K EV miles are no surprise. That's about 3000 charge cycles, which would only be 7% capacity loss under ideal conditions, and even assuming it's twice as bad in reality you still would only be down 14%, which doesn't yet hit the "visible" portion of the capacity.

    In the case of a Clarity, assuming real-world you get around 40 miles to a charge, most of the time the ICE hasn't kicked in when you plug in to charge, you're theoretically looking at 270K EV miles before you start to see decrease in battery capacity. If real-world is twice as harsh as a lab, that's still 130K EV miles. Not bad.

    Not shown in any of those charts is age-related battery decay. I've seen industry numbers in the range of 1% decay in total capacity per year based on aging. This would start to become significant after 5-10 years of ownership, depending on how much you drive. Nothing you can do to help with that, though.
     
  18. M.M.

    M.M. Active Member

    I just realized I forgot one other thing and it's too late to edit: If you want to be really fanatical about maximizing battery life, you should not plug the car in to charge if you're at around 90% charge or higher and know that your next trip won't be long enough to drain the battery down to below, say, 20%. Not topping up the charge at that point will very slightly reduce the reduction in capacity. Honestly, though, it's so minor that unless that was a very regular part of your commute it's not worth worrying about. (For example, if your daily routine were to drive 5 miles, park for two hours at a place you could charge, then drive 25 miles before returning home, it would be somewhat better to not plug in during that 2-hour break.)

    The evidence for this can be inferred in the charge ranges noted above, although I suspect the difference is extremely small.
    Your calculation about the actual used capacity of a Clarity battery are quite useful, but I'm fairly certain that the reason most PHEVs leave extra at the top has little to do with protecting the pack from the engine and mostly to do with extending the life of the battery pack, as noted in my lengthy post above. The i3 you mentioned is kind of a special case since it the engine test mode is essentially forcing engine generation and therefore needs somewhere to send that power; in a Clarity (or an i3 under normal circumstances) if the ICE were running "naturally" and the battery were full, it could (and presumably would) either drop the engine to idle or just shut it off if it needed to.

    Saving that 15% at the top significantly increases the number of cycles you get out of the battery, which is much more important with the comparatively smaller battery pack in a PHEV compared to a BEV with a large battery pack. To a lesser degree (but psychologically significant) it also insulates the user from capacity decrease; if you lose 10% of the capacity when it's all visible, you know you lost 10%, but if you didn't have access to the top 15% anyway nothing seems different.

    If you drive an older Leaf with only 100 mile range and nothing to fall back on, you definitely want to be able to use all 100 miles if you need it, but you'll also definitely notice when it turns into 90 miles (my boss had an older one and has commented on how much capacity he's lost over the years). You might only have 35 miles of range in an older Volt, but since you've got the ICE to fall back on you don't have any range anxiety, and after putting 100K miles on it you still see 35 miles and can drive it the same you always did, even though technically the battery has decreased capacity.
     
    Johnhaydev, atr, Steven B and 2 others like this.
  19. Viking79

    Viking79 Well-Known Member

    Great replies by the way.

    I think it is safe to say it could be for both, protection from damage of over voltage/over charging while the engine is running (it is bad to charge a battery at 75 to 85% SoC at 20-40 kW rate), and as you say the SoC will improve storage life.

    Numerous studies show impact of SoC on storage, and all agree that high SoCs will cause increased capacity fade. This one is fairly comprehensive and shows that there are two distinct plateaus depending on the state of the graphite anode. http://jes.ecsdl.org/content/163/9/A1872.full
    They point out some issues with other studies that don't really keep the SoC steady through the aging period, but when you do you end up with these plateaus. If the car uses NMC cells (unknown I think), keeping at 85% would be beneficial. Volt uses NMC.
    [​IMG]

    Also worth noting is the Leaf. The Leaf uses NMC. If you look at what happens with high SoC and high temperatures with NMC and I think it might become obvious there is an issue using that chemistry in that way. At 25 C it is fine, but 40 C (104 F) it is not and would see rapid capacity loss.

    This also points out that if you try to store your battery at 70% SoC to save it, you probably won't actually help much, charging up every night is just fine. If you wanted to extend the life you would probably have to keep it around 40 to 50% to be safe, which would make the range not useful.
     
    M.M. likes this.
  20. Viking79

    Viking79 Well-Known Member

    PS, I am not disagreeing with your assessment, the Clarity's operating window of 15-85% SoC is definitely chosen for longevity, should make battery life more than twice as long as a car with SoC window of 0-100% (source: http://mat4bat.eu/wp-content/uploads/2014/03/Cycling-Aging-of-Lithium-Ion-batteries_Grolleau_EIGSI.pdf ). Should easily last 4000 cycles

    I am stating that the engine shouldn't run above a certain SoC threshold to avoid damage to the battery as well. If the engine is running in the Clarity or most PHEVs, the generator is always generating power. If it is, this has to be dissipated as heat, driving the wheels, or charging the battery. So yes, you can stop the engine (what the i3 does), but also limit capacity so the engine doesn't overcharge the battery.

    It may very well be the primary reason is battery life, but it definitely protects the battery from the engine charging it at what is a DC fast charging rate.
     
    Johnhaydev likes this.
  21. KentuckyKen

    KentuckyKen Well-Known Member

    M.M and Viking79, thank you both so much for giving us good old hard core facts and not mere opinions so we can use your info to make informed decisions.
    God bless you both.
     
  22. M.M.

    M.M. Active Member

    I was not disagreeing with you, either, and you're certainly correct about overcharging a Li-ion battery. My point was simply that since a PHEV can turn off the ICE at will at any moment without penalty, there's no particular reason for it to worry about the ICE causing overcharge.

    They kind of have to leave some extra overhead for a different overcharge reason, though, since someone who lives at the top of a hill could easily be doing regenerative braking for a while on a "fully" charged battery. I'm actually curious what would happen if/when you did a lot of regen on a full battery to the point that it was approaching overcharge. I assume it would force use of the mechanical brakes at some point. Come to think of it, I know someone with a Leaf who lives several miles up a steep hill, I should ask her.
     

Share This Page