TMC is an independent, primarily volunteer organization that relies on ad revenue to cover its operating costs. Please consider whitelisting TMC on your ad blocker and becoming a Supporting Member. For more info: Support TMC
  1. TMC is currently READ ONLY.
    Click here for more info.

How I Recovered Half of my Battery's Lost Capacity

Discussion in 'Model 3: Battery & Charging' started by SomeJoe7777, Aug 27, 2020.

  1. SomeJoe7777

    SomeJoe7777 Marginally-Known Member

    Joined:
    Mar 28, 2015
    Messages:
    2,164
    Location:
    Houston, TX
    I do not know the values of the resistors, nor do I know exactly how they're wired into each brick (parallel group of cells) or if they can be removed from the circuit if the pack isn't balancing. I wish I knew these items because that would tell us a lot.

    One further note is that I'm being somewhat deliberately vague in my description of the battery construction. I do not want to reveal the exact items that I was informed about by the Tesla technician, as I believe some of those items may have been proprietary information that he wasn't supposed to share. Thus, the post is a mashup of what I was told and what I can infer on an engineering basis.

    I intend the focus of the post to be more on the procedure I used and the results. Changing my charge pattern to allow the car to sleep and sample the battery at different states of charge has contributed to an increase in indicated miles. I am assuming this is predominantly a change in indication and/or a change in how much of the battery's capacity is accessible for use, not an actual change in real battery capacity.

    The 270 miles was indicated after a 100% charge where the car did not sit at 100% for more than about an hour.

    I agree with you that real capacity loss is inevitable and unrecoverable. So we must assume that the increase in indicated capacity is the result of either (1) better estimation of battery actual capacity by the BMS, (2) accessibility to more energy of what the battery pack is actually holding (i.e. the BMS believes we can discharge deeper than before, or the BMS recognizes we can charge more than we were able to before), or (3) a combination of these two.

    I don't know of a good way to compute or observe the actual capacity of the battery without relying on the BMS, other than perhaps discharging the battery to the point where protective circuits will no longer let the vehicle drive, and then charging to the point where the BMS stops charging at 100%, and independently measuring the total energy that flowed into the vehicle. But this is what it might take to get a better idea of what is actually going on here.
     
    • Like x 1
  2. Wennfred

    Wennfred Supporting Member

    Joined:
    Apr 4, 2019
    Messages:
    2,907
    Location:
    San Diego
    Thanks Alan, will give that a go.

    Fred
     
    • Like x 1
  3. AlanSubie4Life

    AlanSubie4Life Efficiency Obsessed Member

    Joined:
    Oct 22, 2018
    Messages:
    8,904
    Location:
    San Diego
    #103 AlanSubie4Life, Aug 29, 2020
    Last edited: Aug 29, 2020
    Yeah, I agree, I don't think that is inconsistent with what I've said. It's really delving into the technical details of BMS, which of course will be quite complicated and intricate since there is likely tons of optimization and cases to deal with. I certainly don't know all of those details at all, and a lot of what I am saying is somewhat speculative, just based on basic electrical engineering knowledge. The devil is in the details!

    I would be careful not to confuse self-discharge of cells with internal resistance. They're kind of different equivalent circuit elements (a leak is a parallel shunt across the internal ideal voltage source, while the internal resistance is a series element modeled in series with the ideal voltage source of a Thevenin equivalent circuit). The shunt resistance of a leaky cell will very very slightly change the short-circuit current, and thus effect the equivalent series resistance, but presumably not by much! (We should certainly hope not!!!)

    Yes, it would increase voltage faster because it has lower Ah capacity. To be clear, in this case internal resistance is not really that relevant (the internal resistance only affects the voltage when charging is taking place, when current is flowing).

    But certainly internal resistance and changes in that value DO affect energy available (and the efficiency of charging!!!). Hopefully internal resistance stays VERY low!

    It's not clear to me whether or not the balancing system typically operates during a charge. I got the impression it really only became active above 90%, when not charging, but there's not a requirement that it behave that way. I really have no idea. Obviously the balancing is really only necessary at high SoC (as long as no cell has hit maximum or minimum voltage it isn't really needed, at least temporarily - of course you do have to rebalance, in general).
     
  4. Battpower

    Battpower Supporting Member

    Joined:
    Oct 10, 2019
    Messages:
    1,950
    Location:
    Uk
    Model S originally balanced at higher SOC. Since total energy disipated depends on time, balancing disipation can be increased by allowing it to happen for longer periods. Now I believe S balances most of the time.

    Yes, I thought that when I grouped those!

    Norton and Thevenin! They seem a long time back in my life, but they still apply!
     
    • Like x 1
  5. AlanSubie4Life

    AlanSubie4Life Efficiency Obsessed Member

    Joined:
    Oct 22, 2018
    Messages:
    8,904
    Location:
    San Diego
    #105 AlanSubie4Life, Aug 29, 2020
    Last edited: Aug 29, 2020
    Yeah, my guess would be that 270 rated miles result was transiently low (and a real limitation of capacity of your battery during that particular charge), caused by a voltage imbalance. Wouldn't take too much imbalance to lose a couple kWh. But I guess we'll never know. Whenever I ask people for imbalance info from SMT, the data never confirms any imbalance...so...shrug. I could be totally wrong about the reason for the 270 rated miles result.

    I'm just saying you can measure how much energy you add to a battery (with reasonable precision) by measuring the length of a charging event, and knowing the charging overhead - which is quite well understood and as long as you turn off any parasitic drains and you're careful to measure the available voltage and current being fed into the car. To get a very close estimate of capacity, you don't have to drain ALL the way to zero.

    We've attempted measurements like this before on TMC for SR vehicles and compared ones showing 200 miles @100% with ones showing 220 miles @100%, and sure enough, the one with 200 miles finishes charging 10% faster, with all charging conditions exactly the same. Fairly definitive. Add 10% more rated miles...takes 10% longer. Obviously you have to be careful during the experiment to not hit the taper, or if you do, account for the increased overhead during that time period (above 97% or whatever) through the appropriate formula.
     
  6. TEG

    TEG Teslafanatic

    Joined:
    Aug 20, 2006
    Messages:
    21,703
    #106 TEG, Aug 29, 2020
    Last edited: Aug 29, 2020
    So there are so many factors to contemplate on charging regimen above and beyond just trying to let the computer learn the actual battery capacity.
    Along with the "be nice to your battery" discussions, there is also a consideration of "how much performance will I get?" and "do I have enough stored range for any trip I might take?"

    Just doing a little personal seat-of-the-pants guesswork, my personal habits have changed a lot during COVID.

    Pre-Covid I would nightly charge to 80%, commute to work (which brought it down to ~60%) and then leave it that way during the day, and have a charge timer to go back to 80% first thing in the morning before I left again.
    If I was trying to stretch out battery lifespan a little, I could get by doing a daily 60% charge, and drive it down to 40% keeping the car near a battery happier 50%.... *but* I notice less acceleration at lower SoC, so I want to keep it higher for driving fun factor and quick passing safety.
    Also, I was tempted to use 90% (for even more performance), but I felt that 80% was nicer to the battery, so went that way instead. 100% is "off the table" for me unless I was doing a once in a long time attempt to retrain the SoC computer.

    Now that COVID hit, and my car sits a lot more, and I am not commuting to work, I changed my charge timer to keep it at 50%. This is more sort of a "storage mode" for me, but does give me enough range to be able to make a sudden decision to drive somewhere and not worry about having to charge first.

    And yes, some of us spend too much time contemplating little details that probably don't matter to most people. Is it worth planing your whole day around making your battery last 15.2 years instead of 15.1 years? Or so you get 0.1 second quicker 0-60 time?
     
    • Like x 2
  7. Battpower

    Battpower Supporting Member

    Joined:
    Oct 10, 2019
    Messages:
    1,950
    Location:
    Uk
    Absolutely.

    My own interest is to be able to differentiate between 'normal', 'abnormal' and 'failure'. Tesla aren't always straightforward in acknowledging when something 'fails' or what the cause might have been. I just want to be on top of that as much as I can so I can relax and enjoy the car.
     
    • Like x 2
  8. AlanSubie4Life

    AlanSubie4Life Efficiency Obsessed Member

    Joined:
    Oct 22, 2018
    Messages:
    8,904
    Location:
    San Diego
    Yeah. The only reason I keep coming back and discussing these things here is that there seems to be a persistent idea out there amongst some Tesla owners that somehow the rated miles reduction observed on every Tesla vehicle (and every EV in general!) is somehow "not real" or "temporary." But I think we've proved elsewhere here that this is definitely not true. Measured carefully via means other than the car itself, any owner can demonstrate that cars showing lower rated miles take less energy to fully charge. That indicates definitively that vehicles with fewer rated miles at 100% contain less available energy after being fully charged. QED. (I guess you could argue that cars with lower rated miles have lower charging overhead but that would make no sense.) But somehow the idea that this is just an estimate (to be clear: it is an estimate, but accurate within 1-2%, probably) persists!!! It's weird!
     
    • Love x 2
  9. drtimhill

    drtimhill Active Member

    Joined:
    Apr 25, 2019
    Messages:
    1,533
    Location:
    Seattle
    I wasn't denying that some/many people see real degradation. Just that there are an awful lot of posts on this forum where people scream about "degradation" based on a quick glance at the range shown on the screen, without stopping to think about all possible causes. One of the interesting points the OP made was that Tesla tune the range to be pessimistic (low) to avoid people running out of battery on a journey (which is sensible).
     
    • Like x 3
  10. AlanSubie4Life

    AlanSubie4Life Efficiency Obsessed Member

    Joined:
    Oct 22, 2018
    Messages:
    8,904
    Location:
    San Diego
    Agreed. It appears to be rather too variable for people to fully comprehend. People aren't used to the size of their gas tank changing with temperature, for example.

    But I'd argue that when the range shows a particular number, it really is an extremely accurate estimate (within 1-2% I would think, not including the buffer) of actual energy available at that point in time at that pack temperature.

    I think we generally understand from SMT that (for Model 3) there is a 4.5% buffer below 0 rated miles, which is part of the available pack energy, which is not *exactly* shown on the display. But this manifests as each rated mile containing roughly 234Wh/rmi (pre-2020 AWD M3) rather than 245Wh/rmi (which is the actual total available energy content of the battery, including the buffer, when you take the number of rated miles at 100% and multiply by that scalar). These values differ by 4.5% of course. This results in about 4.5% of pack energy being available still when you hit 0 rated miles. Hard to say how much of that is truly usable though. Don't want to press that accelerator too hard at that point or you'll get a brownout (hopefully not in your shorts though)!

    Tesla definitely doesn't want to have the car shutting down when you hit 0 rated miles (or slightly before). That seems to me to be the reason for the buffer.
     
    • Like x 1
  11. TEG

    TEG Teslafanatic

    Joined:
    Aug 20, 2006
    Messages:
    21,703
    I changed my Model 3 battery fullness to report in % instead of miles, so I don't have to sweat the inaccuracy too much.
    Heck, the actual miles of range is so variable depending on other factors, like terrain, traffic, wind, lead-food, AC/heat, windows open/closed, etc.

    At the end of the day, I won't be planning any trips where I am trying to depend on the range predictions to be precise to get me there.
    Tesla has enough superchargers, enough range in the big pack 3s, and a smart enough nav system that I can head out and "play it by ear" and figure it out as a I go.
    Watching amount of free slots at upcoming charging locations is another factor as well. An even bigger factor is "when do I want a food and potty break, and what are the nearby facilities like?"
    Some, to some degree, all this sweating over range accuracy is just an academic exercise.

    I lived through earlier generation EVs with < 100 mile range and limited charging options, so I know how frustrating inaccurate range predictions can be when planning a trip, but to me the 3 just avoids all that concern.
     
    • Like x 1
    • Disagree x 1
  12. aTri

    aTri Member

    Joined:
    May 27, 2019
    Messages:
    16
    Location:
    Storrs CT
    I am no rocket scientist. Just a 70 yo grandmother of 11. I gave up 50 years of driving manual transmission ICE cars for the simplicity of a TESLA. I hope you learn to enjoy driving it. Not only is it Simpler, and Safer, but it definitely keeps me as the most popular grandparent around!
     
    • Like x 4
    • Love x 2
  13. AlanSubie4Life

    AlanSubie4Life Efficiency Obsessed Member

    Joined:
    Oct 22, 2018
    Messages:
    8,904
    Location:
    San Diego
    Yeah, a 1-2% error in the estimate is pretty inconsequential, but the 10% drop in rated miles (a separate and distinct factor) is important for me to track since it tells me what I can do relative to what I have done in the past (assuming similar conditions of course). That’s personally why I like to track the energy I have. Just a personal choice - I understand the other perspective - and in the end of course I rely on a % number when using the Nav so at that point the miles number “doesn’t matter.”

    I do this all the time on trips; I try to target arrival at 5-10%. So accuracy of the estimate is very important (and I’ve never had a problem with the accuracy!). Another reason the occasional conversations going roughly: “it is just an estimate and could be far off” are strange to me. The estimate must be very, very accurate! I would not want to have arrival at 2% rather than 5%, by surprise, for example - it would be stressful, or I might have to slow down (or search desperately for a good rock-spewing draft), which would be super bad - and that would be potentially just a 3% error if you start close to 100%. A headwind could cause that, of course, but that is a known factor early in the drive, so that sort of deterministic error is not my point here.
     
  14. TEG

    TEG Teslafanatic

    Joined:
    Aug 20, 2006
    Messages:
    21,703
    From the LEAF days, I would never try to do anything with a plan to arrive at 5%... When you started getting to the lower end of charge, the GOM ("Guess O'Meter") would just change to dashes and pictures of turtles since the electronics really didn't know how much energy was left.
    I guess those past experience have me always trying to keep a 10% bottom buffer (on top of whatever "reserve tank" Tesla already has built-in.)
    Along with "range anxiety" finally kicking in down at the bottom of SoC, I also start to feel like I am hurting the battery, so I try to avoid doing that.
     
    • Like x 2
  15. AlanSubie4Life

    AlanSubie4Life Efficiency Obsessed Member

    Joined:
    Oct 22, 2018
    Messages:
    8,904
    Location:
    San Diego
    #115 AlanSubie4Life, Aug 29, 2020
    Last edited: Aug 29, 2020
    Certainly understand that!

    But that’s the beauty of Tesla’s estimates (which are very very good in my experience notwithstanding known environmental factors like temp or headwind). Target arrival above 5% (not 5%!) and it really can work very very well, without pestering you about having to slow down to “make it”.

    Having remaining energy go to “unknown” would certainly be very disturbing.

    The battery is what it is for me. It’s not like I draw it down to 5% on a daily basis, and if I’m on a road trip, I have to drive, and I want to get there nearly as fast as a gasoline car (which is quite possible with V3s). For that, 5-10% arrival is optimal.

    I don’t find any range anxiety when I’m actually at 5-10%. It’s much more anxiety when I’m at 50-100%, and the prediction is arrival at 5%, or it tells me I have to slow down to make it. The probability of this happening on any given Supercharger hop goes up as my rated miles at 100% becomes lower, of course.
     
  16. Daniellane

    Daniellane Active Member

    Joined:
    Aug 13, 2017
    Messages:
    2,367
    Location:
    Camas, Washington
    That’s a crazy amount of technical detail and execution! I was not even aware of OCV before reading your excellent post. I can say from my own personal experience that your OCD readings are off the chart! ;)
    Take this as a compliment. Thanks for all the hard work.
     
  17. SigNC

    SigNC Active Member

    Joined:
    Aug 23, 2017
    Messages:
    1,461
    Location:
    NC
    I still don't understand the letting batteries at low SoC "rest." I get that they will be warm due to high amps but the car should be able to just cool them off right? Seems like it would be best to get it back up to 25%+ as soon as possible. What am i missing?
     
    • Like x 1
  18. AZAV8R

    AZAV8R Member

    Joined:
    Aug 29, 2020
    Messages:
    88
    Location:
    Arizona
    (Thank you so much for the informative post! I just joined this group to escape the troll-battles going on at the Telsa Forum.)

    This is the kind of info that I've been hoping to find on my M3. Aside from the crap on the Tesla forum driving me away, it surprised me that Telsa is not posting these vs. us having to kind of drag it out of them. There is so much to learn about our amazing machines. It's nice to see fact vs. the rampant speculation we have to engage in to better understand them better.
     
    • Like x 2
  19. immolated

    immolated Member

    Joined:
    Oct 3, 2016
    Messages:
    223
    Location:
    Las Vegas
    After your experiment was complete, continuing the calibration routine seems like a lot of work to change what is essentially a display inaccuracy. As long as you understand the "why", I am not concerned about small changes in the displayed range estimate until I'm at <100 mi and I need to worry about making it to the next charging stop. My old gas car had a "distance to empty" estimate that I treated the same way. Its full tank value changed by +/- 10% easily
     
  20. TEG

    TEG Teslafanatic

    Joined:
    Aug 20, 2006
    Messages:
    21,703
    Welcome to TMC. I abandoned Tesla's official forums years ago for many reasons. Glad you found us.
     
    • Like x 1

Share This Page

  • About Us

    Formed in 2006, Tesla Motors Club (TMC) was the first independent online Tesla community. Today it remains the largest and most dynamic community of Tesla enthusiasts. Learn more.
  • Do you value your experience at TMC? Consider becoming a Supporting Member of Tesla Motors Club. As a thank you for your contribution, you'll get nearly no ads in the Community and Groups sections. Additional perks are available depending on the level of contribution. Please visit the Account Upgrades page for more details.


    SUPPORT TMC