Honda Clarity Top End Battery Reserve

Discussion in 'Clarity' started by Lowell_Greenberg, Sep 21, 2019.

  1. Lowell_Greenberg

    Lowell_Greenberg Active Member

    The attached article is a technical summary of the Clarity PHEV powertrain and a performance comparison with Honda's older generation PHEV powertrain. Among other things it describes the rationale for liquid cooling, how effective cabin space utilization was achieved and why engine size could be reduced from 2 to 1.5 liters.

    Also, figure 12, suggests that the upper end charge reserve is considerably less than the lower end. Assuming the line chart is drawn to scale, based on this, the question arises as to why Honda engineers designed it this way and whether it might be prudent for the user to "virtually" charge to 85%-90% to help preserve Lithium ion battery life.

    Two things are frustrating: (1) Why there is no car indicator of real charge percentage and (2) Why the car/Hondalink app doesn't support charging to X percent. Of course assuming an 8.3% charge rate/hr and knowing the starting percent would allow you to approximate a desired virtual charge limit. But this is inconvienant and prone to error.

    Nor do I think the question is academic. Charge levels above 85% to 90% may not ideal for a lithium ion battery, particularly under high ambiant temperatures and/or extended periods of sitting.

    Finally, whatever the actual upper and lower reserve limits, continually draining the virtual battery percentage to the lower limit and charging to the maxium limit is probably not ideal either. Something like 85-90/25-30 seems best based on my prior reading and limited technical knowledge. Being more conservative than this may extend battery life, but compromise PHEV range so that the car can't be driven primarily on EV. View attachment Development of Electric Powertrain for CLARITY.pdf

    Sent from my SM-G975U using Tapatalk
     
  2. Phil_Meyers

    Phil_Meyers Member

    Thanks for sharing that article. I haven't read it before. Honda only charges to 4.07 volts. To me, that represents a pretty large top buffer, so you're really not charging to 100% in reference to the battery SOC.

    I also understand it as this is a Honda, and not a Tesla. I don't think they intended their perspective buyers to have much technical knowledge on battery longevity and how to maximize it. They don't even give you the charge rate in Kw.

    Frankly, it's something I don't worry about. They did a good job with the top and bottom buffer and it's liquid cooled. I intend to keep driving it as intended.
     
  3. Landshark

    Landshark Active Member

    Great article. The layouts really clarify what is going on under the hood and floor.

    Can you expand in your interpretation of Fig 12? To me it looks like a comparison to their previous PHEV. I don’t see how you’ve been able to determine “charge reserve” from that graphic, nor am I certain what you mean by charge reserve.

    Many of the graphs show no numbers on either the horizontal or vertical, or both, so in those cases all we can determine is that the new PHEV is “Good” compared to the old one as indicated by the stars and arrows.

    I did not see in the document where Honda indicated that the cells are charged to 4.07v. These are 3.7v cells, so charging to 4.07v would be a pretty full charge in my book. The onboard charger has a rated output of 6.6kW at 240v (Level 2). The included “charger” has a rated output of 1.4kW at 120v. It’s really just a power cord.

    Li-ion batteries can deliver 3000-5000 life cycles while still retaining 70% of their original capacity. A life cycle being defined as going from 100% SOC to 20% and back to 100%. Unlike LA batteries, Lithium can be discharged to 20% or even 10% without risking damage. They can also be charged at very high rates as evidenced by Level 3/Chademo chargers. They will take whatever can be thrown at them up to about 90%, at which point the BMS will gradually finish the job.
     
  4. Phil_Meyers

    Phil_Meyers Member

    Hi, 3.7v is the nominal voltage. I will find my reference to 4.07 and provide the citation.
     
  5. Lowell_Greenberg

    Lowell_Greenberg Active Member

    I am referring to the red dotted lines in fig 12 that appear to show available battery energy.

    Sent from my SM-G975U using Tapatalk
     
  6. KentuckyKen

    KentuckyKen Well-Known Member

    Thanks for reposting that article. It was posted a long time ago in the forum but with so many threads and posts, it’s good to churn it back up to the top.

    It’s difficult to get hard numbers from that graph, but using the old Mark 1 eyeball, it’s appears to be ~1 kWh at the top and ~3 kWh at the bottom. And we must remember that this is not the metric that graph was created to show. Nevertheless, that’s not entirely unreasonable since with a 17 kWh battery that would be ~ 5.9% buffer at the top and ~17.6% at the bottom for a total buffer of ~23 to 24%.

    We can also calculate this by using the most widely reported amount of power drawn from an EVSE to charge from 0 EV as 14.1 to 14.2 kWh. Assuming an onboard charger efficiency of ~92% that works out to ~13.1 kWh usable allowed power or about a 23% buffer (13.1/17) split in an unknown ratio between top and bottom of the SOC.
    So both ways are in approximate agreement.

    The 92% onboard charger efficiency estimate is from several sources I found (not Clarity or even Honda specific) whose estimates ranged from 85 to 95% with most in the low 90s. That also is in the ballpark with @Landshark’s post above that the onboard charger is rated at 6.6 kW. My ChargePoint reports 7.1 to 7.2 kW going into the car. And 6.6/7.2 is ~92%. So again I think we have confirmation that we’re at least in the ballpark with all this.

    And that graph does shed a little more light (if not quantitatively) on how the buffer is split between top or bottom. Bottom line (bad pun intended) is that it appears we have approximately a 23% or so buffer split with more at the bottom than top. This would seem to indicate that the Honda engineers are more concerned with depth of discharge and are comfortable with coming closer to a full charge as long as they avoid overcharging. And we see the prevention of over charging by how regen is limited for the first few miles after a full charge.

    So the real question is how much benefit in battery longevity will be gained by charging to a little less than the assumed ~94% that the BMS is allowing. No one knows and to be sure we’d have to wait and compare results and by then it will be too late. My personal opinion and that’s somewhere between a complete WAG and an educated guess, is that it won’t make but a few % difference at best and so isn’t worth the trouble to me. However, if Honda had provided the option, I would probably choose a 90% of allowable charge on days I knew I wouldn’t need that 10% additional range. YMMV as always.
     
  7. Kerbe

    Kerbe Active Member

    I just had a full diagnostic run on my 2018 which I purchased new after it had been sitting - uncharged - on the dealer's lot for nine months. I was concerned about the traction battery's capacity. The test results - a six-page printout - showed that the car has a capacity of 55 Ah and that the "Battery Usable Capacity" is 86%. I would venture to guess, then, that 14% of capacity is split between the upper and lower end for battery protection. The "common knowledge" that one should not charge a lithium traction battery pack to more than 90% of capacity seems to be more about DC charging - which stresses the battery - than it does Level 1 or 2 "trickle" charging...
     
  8. Ray B

    Ray B Active Member

    I think it is important to think ahead. The published buffer in the BMS is for a brand new battery, but obviously there will still be degradation in the battery capacity over time, so the buffer will be consumed to maintain the same usable capacity. This is shown in this chart from the battery university:

    [​IMG]

    Honda has published a similar chart for the Fit EV:

    upload_2019-9-22_10-42-32.png

    So I had adopted the charging strategy that @KentuckyKen mentions, but I tend to target ~93% as the maximum charge, and I try to avoid going all the way down to 0EV when possible. I don't generally need the full range on a daily basis and I think it is a little easier on the battery. I will attach a couple of pdfs from Honda which gives some background.

    The files are sourced from here: https://www.hondarandd.jp/index.php
     

    Attached Files:

    Last edited: Sep 22, 2019
    KentuckyKen likes this.
  9. Landshark

    Landshark Active Member

    After reading just the first attached document, I’m convinced that Honda has established a charging protocol, designed a BMS and set parameters that will prevent any of us from being able to cause harm to the batteries. I’m also convinced that I can’t outsmart this thing and that attempting to do so may, in the long, run have detrimental effects.

    It would be an improvement to have access to additional information, such as battery voltage or SOC as a percentage. This is my first Honda automobile and their policy seems to be to provide the customer with only the information they deem necessary. The battery gauge has 20 bars, so we could determine that each one represents 5% of capacity. Likewise, if the EV range displays 50 miles after a charge we could estimate that each mile uses 2% of capacity.

    I read in one review that the car will switch from EV to HV mode when the battery reaches 11% capacity. That seems low.
    What is 11% to Honda? Is it 11% of useable capacity or 11% SOC?
    What is 86% useable battery capacity to Honda?

    A 100% charge on a 3.7v cell would be 4.2v. That would be a 19kWh battery.
    A 90% charge would be 4.1v. That’s 18.6kWh. It was mentioned previously that Honda charges the cells to 4.07v. That is just touching 90%.
    For those attempting to charge to ~93%, are you charging to ~93% of 90% or are you attempting to charge the battery to a higher voltage than Honda has established with their charging protocol?
     
    Peter CC likes this.
  10. I bought a 2 year old Nissan LEAF a few weeks ago and it's been fun so far.

    Nissan did a lot of things wrong with the LEAF. Their primary sin was in opting out of active battery temperature control. LEAFs in severely hot climates experience disappointingly quick battery degradation, particularly if routinely charged to 100%. I bought a Michigan car, with only one Level 3 charging session to his history so the battery is still relatively healthy.

    The good news for the LEAF owner community is that there is a third party app available in iOS and Android called Leaf Spy which can extract detailed vehicle data including battery condition information through the OBD port.

    It would be wonderful if a similar app could be developed for the Clarity, as I suspect there's much that the car could report through such software.
     
    Peter CC likes this.
  11. Phil_Meyers

    Phil_Meyers Member

    According to this printout, Honda charges to 4.14v.
    https://www.insideevsforum.com/comm...e5ff1f-391f-42ad-b1df-0679d72aff37-jpeg.4924/
     
    Kailani likes this.
  12. Landshark

    Landshark Active Member

    Thank you.

    It shows 4.14v as a limit and 4.09v as a target. Also shown is a charger inlet temperature of 82F. It appears that Honda is targeting a ~90% SOC, at least for a new battery.

    This practice could delay the onset of battery degradation. It also allows for a higher percent SOC in the future in an effort to maintain consistent useable capacity over time.
     
    KentuckyKen likes this.
  13. Lowell_Greenberg

    Lowell_Greenberg Active Member

    I think the attached from the NREL is one of the best papers I have seen on optimizing PHEV battery usage and management.

    Key factors relate to BMS design/effectiveness (something the consumer has no control over), Lithium ion battery type, average outside temperatures, storage times at maximum SOC, daily number of charging cycles/DOD (depth of discharge), etc.

    Some of it's more interesting conclusions:

    (1) Issue: xEV battery packs are oversized & controls are conservatively tuned to achieve typical life of 10 years.

    (2) Oversizing is expensive. Solution: 35% smaller HEV battery by providing vehicle controller with real-time knowledge of battery degradation

    Item (2) is interesting and would reduce costs while maintaining battery life. Item (1) is depressing for those that want to hold onto a car beyond the 10 year point. However, fewer charge cycles, milder climates, fewer high DOD events, and fewer hours having a car/battery sit at maximum SOC may improve upon average battery life, assuming the BMS system offers effective, balanced cell cooling.

    As for the upper reserve, 90%-95% maximum SOC, should be fine, particularly on a PHEV, where there is a significant range tradeoff by significantly lowering this max number. And finally, this points out the principal drawback of a PHEV- smaller batteries mean higher daily DOD and more charging cycles- which equates to shorter battery life. View attachment Battery Life Cycle.pdf

    Sent from my SM-G975U using Tapatalk
     
    KentuckyKen likes this.
  14. KentuckyKen

    KentuckyKen Well-Known Member

    Thanks to all and especially @Ray B for posting all that glorious data from Honda engineers. I’ve saved all of it in my Acrobat and it will take some time to digest it all.
    This is exactly what we need in our attempts to figure out the Clarity (without any help from Honda, I might add!).
    And thanks to all the others who posted very informative data and sources too.
    Mo’ data!!!!!!
     
    Ray B likes this.
  15. AnthonyW

    AnthonyW Well-Known Member

    The battery charges to 342 volts on the top end. Nominal is 311. Since the battery is configured as 84p2s that equates to 4.07 on the top and 3.70 as nominal per cell.


    Sent from my iPhone using Inside EVs
     
  16. Landshark

    Landshark Active Member

    Thanks. Do you know the voltage when the display shows EV Range 0.0?
     
  17. Hoon

    Hoon Member

    The lower end having much larger buffer/reserve on a PHEV probably have to do with HV operation mode. They need that reserve to operate the vehicle with sufficient power as the gas engine alone is not powerful enough for normal acceleration.

    Being a PHEV, there's going to be much more stress on the battery through depth of discharge and more cycles compared to a BEV. Still, I would think Honda designed it to hold at least 80% covering the warranty period (10 years in CA), which is pretty good if you think about the li-ion battery.

    If there were a user-selectable charge upper limit setting, I'd set it to 90%. I usually time it when I'm able to, not being paranoid about it though. I'm using the 120V OEM charger on a 240V outlet, which gets me about 20% per hour. I stop charging from my phone through HondaLink app if I remember to do so. I don't like how the regen brake feels different when you have a full charge, and it's theoretically better for the battery to not be in fully charged state.
     
  18. Mark W

    Mark W Active Member

    Great information. Thanks for posting it here.
     
  19. Ray B

    Ray B Active Member

    The EV 0 can be a moving target, as it can normally be ~12% SoC when it just ticks down to 0.0, but I've seen lower, and I've seen it as high as 15%. It may depend on a lot of factors such as ambient temp, battery temp, recent driving patterns/aggressiveness. Of course, it is possible to pull the SoC down a lot lower than 12% - I have gotten into the single digits at times, maybe as low as 5-6%, and I believe some have reported 0% (and no bars on the gauge).

    So we'd probably need someone with enhanced ODB scanners to pull info on EV voltages at 15%, 12%, and 10%, which should be easy to arrange, but it may be tough to configure a test at lower SoC% as the car tries to prohibit it as much as possible. Of course, 0% SoC voltage may be published elsewhere, but I haven't dug into it. If I find it somewhere I will post it.
     
  20. 2002

    2002 Well-Known Member

    I use scheduled charging, what I do is always keep the end time at 7:00 am then I adjust the start time so that by 7:00 am it will have reached whatever SOC I want. Cheaper electric rates during the night for me, also the garage is cooler.
     
    Walt R likes this.

Share This Page