How I Recovered Half of my Battery's Lost Capacity

[AlanSubie4Life]

My EPA range is 316 but reported max range has mostly been between 305 and 309 since I've owned the car. My gut says not likely I've lost that much range in 2 months. Hopefully this exercise will get me back to a reported 316 mile range. Fingers crossed.

I would ignore data points that are gathered with your SoC below 90%, as they are prone to excessive rounding/extrapolation error. Charge to 90% and then extrapolate using the slider in the Tesla app to see what it says for your 100% range (will be within ~+/-1.5 miles of correct). Also ignore data points gathered when the car battery is cold soaked below about 65 degrees F.

This range loss is real, and normal. The Model Y has very little capacity loss that is required to occur before you see a loss in rated miles (unlike the 2018/2019 Model 3s - the 2020 Model 3 is similar to Model Y in that capacity loss can be minimal and a decrement in the 322 mile range of the 2020 Model 3 will show quite quickly).

I don't track the constant carefully in the Model Y, but you're likely at about 3% capacity loss, which is quite normal after a couple months with some use.

Capacity loss is quite fast at first. Early Model 3 owners didn't see this because they could go from 78-79kWh to 76kWh without seeing any loss of rated miles (all evidence seems to suggest that Tesla appears to enlarge the energy content of each rated mile, until going below a certain energy threshold, whereupon energy content per rated mile hits a minimum, and remains constant from there), but looking at Stats and other TeslaFi charts, it looked pretty typical to lose 3% capacity in the first 3-6 months (most people started seeing a small decrement (say 309 rather than 310) in their rated range when the vehicle was 3-6 months old - which corresponds to about 3% capacity loss - 309 rated miles is 75.7kWh, and most people started with 78kWh+).

Part of this change (capacity loss showing sooner on 2020 vehicles) likely has to do with tighter control of initial capacity at Tesla with newer packs (might be easier to guarantee that every pack starts above 77.6kWh (which is the new requirement (instead of 76kWh) to show full rated miles on 2020 Model 3) than it used to be). That means that vehicles can all show the same initial range as long as they are all over 77.6kWh (this may not have been the case before and obviously Tesla would want to avoid that customer upset - hence the modification of rated mile energy content initially to make sure everyone shows the same rated miles at first, even if their energy content starts at slightly different levels - everyone is within a couple % so it's not a big deal). But the capacity loss characteristics are likely still similar on 2020 vehicles (even though probably there have been some small tweaks to the chemistry). The tighter initial starting point and the slightly lower energy content per rated mile (there are more rated miles in a pack that is the same size) means that that rated mile number is likely going to start ticking down sooner on newer vehicles than in prior years, even though loss of capacity is not actually occurring any faster.

To be clear, this is somewhat speculative on my part, but the data all appear to point in this direction and I haven't seen any SMT data or other CAN bus data which indicate it works in any other way. And it's the only way to make sense of the EPA data, SMT data, etc. And it does fit precisely with the SMT data once the degradation threshold has been met. (77.6kWh for 2020 Model 3, 76kWh for 2018/2019 Model 3) Unfortunately it's rare for brand new owners to immediately connect SMT (in the first week!) so we never catch those initial data points, which is what we'd need to confirm the rampant speculation above.

 

[ewoodrick]

As in those who say "you really haven't lost range, just switch the display to percentage"?

Exploring that idea: When new, my midrange M3 had a range of 260 miles. I charge it to 100% and then look at the predicted range I see it is 260 miles. I then make a trip of exactly 260 miles under ideal conditions. Sure enough, when I look at the display I see 0 miles and 0%.

Now, two years later, my display tells me that when charged to 100% I can go just 230 miles (pretty close to my current car). So, I charge to 100% and take off and drive. When the display says 0%, I stop or(or car quits). At that point my M3 has been ringing alarm bells for some time. When I switch the display to miles, am I not going to see I've only gone 230 miles?

So, I guess my basic question is, what is battery voltage at 100% and at 0% for the second case? I assume for the "new car" case the voltages would be ( from graph provided later by dhrivnak )58.2 and 44 volts. My question is: Wouldn't those two voltages be different for my two year old car?

The number varies, and that's the whole basis of this discussion.

When charging batteries, the most common way is to apply higher than the cell voltage until the charging current comes to near 0A (or below a specified threshold)> You then allow the batteries to settle down (lots of chemical reactions need to complete) and measure the voltage and that becomes 100%.
If you don't charge to 100%, by watching the current go to 0, then you are only guessing.

The low voltage is measured differently, it's a combination of engineering and analysis. Basically, it is determined by the lowest voltage that nominally that doesn't cause enhanced cell damage. So while part of it comes from chemistry, real life plays in and you basically need to run a bunch of cells through testing to see what their MTBF is.

And whatever that number is, Tesla has increased it. That's how cars that set in auction houses or used car dealer lots can stay at 0 for long periods of time, because 0 really isn't 0. At a certain percent, the car starts to turn things off. And before it hits real 0%, it has turned EVERYTHING off. And, in case you didn't realize it, to measure a battery, requires the battery meter to be on, so at 0, it has to be charged, just to get the car to turn back on.

Taking the battery to 100% is a recommended practice, the manual and other places recommend it when needed.

Taking a battery to 0 isn't a great idea. I don't recommend it.

 

[Torquer]

Were there other changes to your charging routine? eg: use of superchargers.

Looking at your post makes me wonder if the title makes sense. If this is showing 'battery degradation' then you just reversed degradation. If we are looking at 'calibration error' then you change in charging allowed the BMS to revise its calculations. If it is showing 'Inaccessible capacity due to imbalance-related factors' then your new routine has addressed that.


View attachment 582024

Thanks Battpower, good point on the chart title. I doubt too that degradation has been reversed! I agree it is 'calibration error', though I had tried a couple of times to see if it could be reset with the fabled 'bounce' but with no result. l believe it was my 'old' routine that caused the accumulated 'calibration error' and agree its the new routine that caused the BMS to revise its calculations.

The most noticeable change in the routine was charging to 80% when battery is below 80 miles. I used to plug in every night and had charging set to 90% or 95% completely unnecessarily.

I have used superchargers rarely except for two 1000 mile trips.

Thanks for the comment - helps to confirm my suspicions.

 

[Battpower]

the chart title. I doubt too that degradation has been reversed

I think we all tend to accept charts and data when it is clearly and professionally presented. Especially with the kind of discussion here, it's easy to format that a lot of our source information is the product of very dilligent but non the less non-guaranteed reverse engineering of some sort. I have seen quite a lot of rambling debate about cell balance based around 3rd party tools that give a reading called 'imbalance'. Data that only loosely relates to claims being supported by it really can get everyone very confused.

 

[ewoodrick]

As an Electrical Engineer, I would think you would better appreciate the value of calibrated instruments.

Sure, the readings are not perfect, but the Tesla Guess-O-Meter is WAY better than my old Chevy Volt's meter, mainly because Tesla knows the route you're taking and elevations along that route.

Another thing I would like you to clarify for me ...

You are the 1st person I've seen to state that leaving your battery charged to 100% for days or weeks up to a month is not harmful. (!)

On what do you base that claim?

Because the evidence that charging a battery to 100% is damaging has timeframes for many months and years. Leaving it at 100% for a year will cause a problem. And honestly, if you look at what the car tells you, it's the same thing.
Do you get a message if the car is left at 100% for a long time? No
You will only get a message if you start to show a habit of charging to 100%, i.e. multiple charging to 100% sessions.

Your suggestion that Tesla is better than the Volt, while it is probably true, your reasoning is flawed. The Tesla GOM doesn't look at the destination. How could it? I hazard to say that a lot, maybe even most driving is done without a destination defined in the car.

It is probably true because a LOT more engineering has been accomplished by all car and battery manufacturers and then I dare say that Tesla has gone well beyond Chevy in the amount of engineering put into the car. Chevy seems to be more of a design it, get it out the door and then forget about it organization. (and then a few years latter create a new team for a new car)

But even Tesla still gets it WAY wrong. Let's wait just a few months and see the yearly resurgence of "I'm not getting my full range" posts that occur when the temperatures get colder. While I'm sure that Tesla could probably get this number a lot better, they don't because if they did, they'd get. "I'm getting what the GOM say, but that's only 60% of what I got a few months ago)

And probably the biggest difference between the Tesla and Volt/Leaf. If your guess is off by 30-50 miles in the Tesla, you still get home with plenty of range. In the Volt/Leaf, you can get stranded on the way home.

 

[ewoodrick]

I think we all tend to accept charts and data when it is clearly and professionally presented. Especially with the kind of discussion here, it's easy to format that a lot of our source information is the product of very dilligent but non the less non-guaranteed reverse engineering of some sort. I have seen quite a lot of rambling debate about cell balance based around 3rd party tools that give a reading called 'imbalance'. Data that only loosely relates to claims being supported by it really can get everyone very confused.

And every one of these charts starts with an estimation/guess.

Rechargeable battery charge isn't something that can really be measured without destroying the cell. Everything else is truly an engineering estimate and, as indicated from this thread, the estimate can (and does) easily drift over time.

But yet you can create some highly detailed charts that are quite worthless.

I'm now on my 4th EV, and have learned that the battery charge is an ESTIMATE and that the battery range is an ESTIMATE.
I do know that my Model 3 with over 40,000 miles is still within 1-2 miles of initial estimates and that in real life, I'm getting about the same as initial and that I can get more than that if I want to.
And, knowing how/where I'm about to drive, be a lot more accurate than the GOM in the Model 3 or my Model Y.

 

[SageBrush]

Tesla is primarily interested in getting past their warranty obligations.

Like the battery degradation warranty for the Model S ?

 

[Battpower]

Like the battery degradation warranty for the Model S ?

FWIW!

 

[AlanSubie4Life]

I do know that my Model 3 with over 40,000 miles is still within 1-2 miles of initial estimates and that in real life, I'm getting about the same as initial and that I can get more than that if I want to.

I'm not sure what you mean by that. I guess it depends on what your rated miles at 100% is. Maybe it hasn't changed for you.

In any case, if you compared the trip meter kWh reading for a full 100% to 10% discharge (without stopping or spending time in park during the discharge) when the car was new, to what you see now, for a 100% to 10% discharge (without stopping or spending time in park), I'd expect that you would see a substantial difference (and it's unlikely due to software changes).

I could predict what the difference would be but it depends on what your 100% rated miles value is. And your Model 3 type (I can't remember if you have a RWD - obviously for RWD owners the situation is more complex since some additional energy was unlocked (I don't know whether it was at the top or bottom) in early 2019 - that's kind of a special case).

Anyway, all that really matters is how much energy you can get out of the battery, not how far you can go (which is obviously going to depend on many factors outside of your control). You want to have as much energy as possible that you can extract from the battery, because that means you will be able to go further, for a given set of conditions.

 

[ewoodrick]

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!

No, the rate reduction is not REAL on EVERY vehicle, but especially in the Tesla, (who is one of the few that recommends charging to less than 100% daily) it is common for a LARGE number of vehicles to show a range not consistent with reality. The BMS wants a higher charge to calibrate the range and if it doesn't get it, the range decreases over time.

This thread is about the apparent FALSE decrease because of the BMS.
Yes, a few percent is expected on a new battery pack. But it wouldn't surprise me if this may actually be built into to Tesla's range estimates.

 

[camalaio]

These are the bleed resistors from a Model S. 4 x 158 ohm (the ones marked 1580 in 6 rows). 4 resistors effectively balance 1 brick of parallel cells. In the S these are then in modules of 6 bricks, so this picture is of the balancing resistors for one module of bricks. They are tiny resistors and on the face of it totally not up to the task of balancing 70 - 100kwh of cells (although of course there are around 15 such boards for a whole battery.)

View attachment 582335

Ok, neat. Really hard to tell, but it looks like those 4 are in parallel (the vias seem to also bridge over the pads). If so, 39.5 Ohm. The larger picture shows it's one of these banks per group of parallel cells I believe. They also seem to have transistors (Q6, Q5, etc.) that are probably switching the balancing resistors on/off, basically. Of course, that's a wild guess, but likely not too far from the truth.

I wish we had pictures of the backside of the Model 3 BMS. I'm pretty certain the balancing isn't happening on the front side of the board that we have pictures of (I can't see anything that would serve this role).

The BMS needs to see everything. Charge, discharge, heavy / light loads, settling times.... the lot. Not driving means it has loads of time watching 'settling' but no corresponding energy use the effects of which are worth gathering data about. I'm sure Tesla gather as much useful data as possible to guide the battery management process. EM has said 'good for the car to just leave it plugged in on AC'. This is likely correct as it leaves the car to decide exactly what it needs to do, including a top-up charge to whatever SOC limit is set.

I should've seen this response coming and given more detail haha. It's "seeing" just about everything it can still, just driven less on a daily basis (it still goes on day trips too).

I only disagree that the car gets to decide what to do, and to be fair I disagree with both Elon's wording and the wording of the owner's manual on the subject. The car doesn't get to decide anything about the SoC. I, the owner, set that. I even set what time it charges! It can technically modify things within those constraints (e.g. cap my max voltage permanently), but those are outliers and haven't been observed in the wild so far for Model 3.

The main benefit to keeping it plugged in is simply to avoid a case where the battery may accidentally discharge more than expected.

I think there is an important distinction needs making between batteries that are basically in' good shape' and capable of doing the required job, and batteries containing bricks / cells that for whatever reason have a significantly degraded and become sufficiently 'different' in their characteristics that the balancing systems can't keep the pack usefully uniform / coherent in its energy storage and release function. If you reach that situation then whatever you do, it's just trying to polish the proverbial turd. Hopefully M3 / MY cars won't see this for some time but there is plenty of evidence amongst older / heavily used MS.

Absolutely! There are cases like this for the Model 3 on these forums, where the battery is clearly physically faulty in some way and that is the limiter on capacity. Unfortunately there's just no easy well to tell this via the car itself, as it appears the same: reduced range.

I tried to find a picture of the disipating components for the M3 and found this:

model-3-battery-pack-bms - TESLARATI

but am not sure what's what.

Unless the dissipating components have some provision made for their dissipation needs, (in Model S they are internal to each module so presumably experience whatever the internal module temp is) I can see that their ability to function as required could be compromised.

I wonder if there isn't a better approach to balancing using ultra capacitors? This would seem especially useful in an application with high transients especially while driving and would allow high level regen even at low temperatures. It could also allow charge to be shuttled from one brick to another in an active manner rather than passively bleeding off energy - even if when everything is in good shape you don't actually have much energy balancing to do.

I linked to a similar photo in the part you quoted (links aren't super clear on these forums, are they?)

The Model 3 BMS boards sort of just sit on top of the modules.

But anyways, congrats, you've discovered "active balancing"! This actually exists, with different ways to accomplish it. Basically, tying a higher group of cells to a lower one temporarily: using the energy of the "too high" cells to raise the energy of the "too low" cells.

It's a less wasteful approach, but on average balancing energy is not a significant portion of standby energy usage (at least, not in a Tesla where there are so many other ways it wastes energy).

I agree that the resistor explanation is a lot of voodoo.

So let me clarify that when I'm talking about vampire drain in this context, I'm speaking of the inherent drain inside the main pack itself. The 60W median you're speaking of is the drain for the entire vehicle. That includes energy use that's far larger than the internal balancing resistors, i.e. charging the 12V via the DC-DC converter. So your calculations that show a very low and impractical resistor values I don't think are accurate.

I do wish I knew more of the engineering design choices behind the bleed resistors and we could estimate better what is going on.

Ah, ok. Since it was coming from Tesla service folks, I had assumed "vampire drain" would include all user-observable drain since they field those concerns.

I was definitely presenting the impossibility of the claims as-is. I know the ~60W number is due to various other reasons (the car being routinely "awake" is the actual primary factor, in which many other things use far more power). So yeah, my calculations shouldn't reflect reality - they were to demonstrate that the appearance of the claim cannot be true.

We can see from the picture posted by @Battpower that there are 158 ohm resistors in the Model S/X pack. Without looking at the back of the board, I can't tell whether these are in parallel, series, or some combination so I don't know the exact resistance presented across each module. But if the Tesla tech was accurate on the one point that the Model 3 battery aims to reduce the drain through the balancing resistors, then the resistors in the Model 3/Y packs should be higher in value.

It looks to me like they're in parallel, so 39.5 Ohm, implying 0.4W dissipation (~100mA). This appears to be much more reasonable. There's also nearby transistors (labelled Q) that are probably (dis)engaging those balance resistors. That is, they are not engaged full time, but engaged as necessary and individually (each group of components seems to go to a different group of cells in parallel).

I can't find the post anymore (sorry!), but someone rightly pointed out that when being switched on/off anyways, the resistor value only serves to change the amount of time needed to balance, but the same amount of energy will be used since dealing with a certain amount of imbalance via discharge requires removing a certain amount of energy. One reason to increase the resistance could be to give a more accurate reading while balancing though, as less current would lead to less voltage sag (we're talking tiny amounts though, not sure this is a problem in reality at these values).

While my post may indeed raise more questions than answers about the battery construction (and I wouldn't have posted if all I had was just the Tesla tech's information) the results speak for themselves. How bad does the BMS have to be at range estimation if I can gain a full 20 miles in indicated range just by changing my charging habits? I think this is a very significant finding.

I strongly suspect this is confirmation bias, but read on.

I fully believe there was something else goofed up with your case. In fact, though I've been avoiding the topic, I think your pack is (only) the second case I've read on that it appears balancing may have been the result of your range recovery.

Our current understanding of balancing outside of this thread is that the battery needs to be at rest and above approximately 73% (I'll have a thread on this later today). You appear to have been consistently using Sentry mode, which would mostly not allow balancing (or at least decrease the total time it can spend balancing).

So in my mind, the BMS was actually being very good at energy estimation. It was probably accounting for your (lightly) imbalanced pack. When given the opportunity to balance (you stopped using Sentry, etc.), it then correctly reported you can extract more energy before it needs to cut off the supply.

The finding is significant for that reason. It shows a case where doing "the bad thing" (e.g. Sentry 24/7 in an extreme case) could lead to reduced range, but also have that "lost range" be recoverable. Again, it's something we knew about, but it just doesn't matter to many people. Most people's charging habits and battery health will result in well-balanced packs, by design.

This is just one of many oversights and side effects, IMO, regarding flashy features like Sentry Mode.

I agree with you on most of these, but I'll point out a few things:

1. If you're going to charge to 100% for range, like for a road trip, then let the car charge fully. This may take a while. Incomplete charges to 100% where the balancing doesn't finish can make the calibration worse.
2. Obviously, don't strand yourself by leaving the car with a low charge. For my family, we have two Model 3's, and my wife's is nearly always at 80%, so we can always use hers if we need to go somewhere and my Model 3 doesn't have sufficient charge. Furthermore, we have a supercharger about 4 miles away from us, so we can use that if required. Indeed make sure you have a plan if you want to leave your car in a low state of charge.
3. I didn't make my battery meter slightly more accurate, I made it a LOT more accurate. Like 8% of total capacity more accurate. That's very significant.
4. While the 3rd party tools may have some idiosyncracies, all of their data is from the car. It may not be presented in the best way, but it's not false. And, comparisons that you make within the same tool can indeed tell you things. If TeslaFi's 270 mile estimate when my battery was at it's low point wasn't precisely accurate, that's OK, because we can assume that the same inaccuracies are present in the current reading of 290 miles. I gained 20 miles in this tool, calculate from the car's data.

I do hope we can eventually answer a lot of these questions, but without more teardown and analysis of the Model 3 pack, I don't see it happening. We can banter back and forth here until the next blue moon, but still may not get anywhere without more facts.

I guess it's just easier to keep the numbers going eh?

1. I will still state that balancing and calibration are separate concerns. But it can only balance after charging is done, and balancing itself is a bit of a slow process.
2. I pointed out "stranding" because you have the burden of a stickied post in which you implied a recommendation by stating what worked for you. The internet sucks, doesn't it? Suddenly you're on the hook for everyone's personal cases without even having given context to your own.
3. See previous chunk of my post. You probably ended up balancing it (and thus increased extractable energy), not necessarily made it more accurate. Significant yes, but potentially not the cause you are thinking.
4. I will very strongly disagree here, elaborating below. Even the official app (but not car) gets this wrong!

The third-party tools use the Tesla API data which, yes, is the very same data you see in the app (and thus in the car as well to some extent). There are higher inaccuracies at lower states of charge - you cannot assume the "absolute error" for range at a lower SoC is the same at higher SoC.

Let's take my car's data, right now, from the API. In-app it's displaying 262km of range, which is about 162.8mi, thus the 162.85 below.

'battery_level': 55,
'battery_range': 162.85,
'est_battery_range': 236.2,
'ideal_battery_range': 162.85,
'usable_battery_level': 55,​

There are a few very important things to note here:

• Any representation of percentage is rounded. This is the primary source of error. At 20%, this could be either 19.5% or 20.49%. Since you divide by this to estimate range at max, this introduces huge variance at lower percentages (note how at 1%, it ranges from 0.5% to 1.49% - that's a 50% swing, showing where this increased variance comes from).
• I have no idea what "est_battery_range" is. Assuming it's miles that's ~380km, far too low for representing the 100% state and far too high for its current state. Assuming it's somehow in km, it doesn't jive with anything else.
• The numbers the apps use are the temperature-varying ones. For tracking actual battery health, this is unfair. Lower capacity is reported in cold, though this is not an actual indicator of degraded battery health.

So let's do the math these services would do. Take the current range (162.85) and divide by the percentage (0.55). That comes out to 296.1mi, or 476.5km. Also, in the app, if I drag the slider to 100%, reports that my 100% range is also about 476km, because it's doing the exact same calculation on the exact same data. So far so good, right?

Now recall the issue I mentioned with rounded numbers. I happen to know via the CAN bus that the actual SoC right now is 54.5%. If I use that instead, I get about 481km (and this is pretty much the actual 100% range of my car right now). But had it actually been 55.49% (and still reported "55"), that would falsely indicate 472km. That's already a 9km (5.6mi) spread.

At lower SoC, it gets even worse. Say at 10%, it was reporting 48.10km. But that "10" could be 9.5 or 10.499, which would indicate 100% ranges of 506km and 458km respectively. That is now a ridiculous 48km (~30mi) spread. In other words, that's 10% of the max range.

100% range estimates are estimates, whether it's a third-party service or the Tesla app, and their accuracy is worse at lower SoC.

Now, that is not to say the current range number in the car is inaccurate. That one is a pretty dang accurate representation of the current energy in the pack. The error is introduced when using a rounded, low significant digit percentage to estimate the 100% capacity.

These services have their place and can provide value, but the recommendation to throw away data for <90% SoC (if the service allows this) is absolutely warranted. Anything much below that has too much error for discussion on single-digit percentage degradations.

After reading this thread, I agree with SomeJoe777's conclusions but I think there is a much easier way to get there. I am a retired engineer, who has build several EV's, owned a Roadster for 10 years, a Model 3 for 2 1/2 years (47,000 miles) and in all have over 135,000 gas free miles under my belt. I charge each day to 90% and about 4 times a year take a trip where I charge to 100% and then the car normally sets for a few hours as I set the charge at night and the car fully charges while I sleep and then I head off in the morning. As Joe777 alluded to this ensures my battery stays balanced AND calibrated. I learned with the Roadster (and parameters may be different) that there is no balancing below 82% charge and only partial balancing at the default 83% charge level. We were schooled to occasionally range charge for full balancing and calibration and that served me well for 10 years. The reason is there is very little voltage difference in lithium batteries between 20 and 80% charge level. So if you routinely charge to say 70% there is not enough voltage difference for balancing to occur. You only see differences at the extreme levels, below 10% and above 90%. So my default charge level is 90% and that is why I do not stress about an occasional 100% charge. The BMS needs to see both the top and the bottom on occasion and if it does not the estimates, being conservative will drop your range. Part of this is real as a pack is only as good as the weakest brick. And if your pack is out of balance that low brick will not be brought in line and it will limit your range. It is also true that if you always or eve frequently charge to 100% that too will damage your battery.

Now look at the second graph of my Model 3 over the past 47,000 miles. My range is actually up 1 miles if you compare my first 15 readings (278.2) versus the last 15 readings (279.4). One could argue I am actually down 4.5% as I did get a 5% range boost from Tesla during this time. But I have also noticed the average can jump up or down several miles after a firmware update. But If you look at that second graph I think you will find loss in capacity is negligible.

So my advice to those who stress about range, or loss of range. Charge daily to 90%. On the few times you could use 100% use it. And if you occasionally go below 5% that too is fine as the BMS needs to see both the top and the bottom for it to be accurate.

An analogy if you like at the bottom.

View attachment 582406


Lets say you start off on a hike to an overlook 3 miles away. Knowing your pace of 2000 steps/mile that is 6000 steps. So you start off and after 4000 steps, you realize you need your hiking stick that you set down on a break so you hike back 3000 steps. Then you start back and after 3500 steps you realize you dropped your jacket out of your pack and turn around again and take 1500 steps back. Then you resume your hike. Using math, you should still be 3000 steps from the overlook. But are you what if your pace today is 2100 steps per mile, or the trail was more rocky than normal so you had to take a lot of little steps. The only way to really know is to get to the overlook or back to the start. And once there you can now calibrate your position and for the first time in many hours you for sure know where you are at. The rest of the day was just educated guesses. If the BMS does not know the start or end, you are left with estimates and guesses. If those estimates are estimates of estimates for over a year the accuracy is fair at best.

I'm getting winded with this at this point (I have a post coming I can finally just link to), but balancing is probably occurring at 73% and above. There is absolutely no need to charge to 90% on Model 3. I'd also point out that the only thing the Roadster and the Model 3 really have in common is that they are electric vehicles - everything is different otherwise.

@camalaio You handled being "that guy" very, very well. Seems like every time you post, I learn something.

Thanks, I really try. It's exhausting if I'm honest. For every factual post, there are 10 other more popular posts with varying amounts of fact (not saying this thread specifically!). I guess I've just learned to try to explain, even if it's an information overload.

Except Tesla says we should always keep the car plugged in.

from the Model 3 Owners manual. “The most important way to preserve the Battery is to LEAVE YOUR VEHICLE PLUGGED IN when you are not using it. “

I wish we had more details but until then I plan to keep my car plugged in.

The main (only?) benefit to keeping your car plugged in is to not be surprised with a lower-than-expected state of charge (of which there could be numerous causes, including not fully closing your door and thus keeping the climate control on for a while).

The other dubious but related benefit is blissful ignorance of how much energy the car uses in standby!

Understood but was curious the reasoning for the faulty valves themselves and if it affected the battery's ability to properly control temp in anyway. Then if true, cause battery issues down the road. Far fetched I know however, My pack was "popping" so hard during charging that I could feel it in the floor. Quite alarming. Like someone taking a sledge hammer to my battery pack. I'm probably going off topic. Sorry about that.

It's not related to temperature management at all, just pressure changes (which come about via elevation and temperature changes, yes). Basically it seems like their thresholds were too high and would allow a sheet to flex and pop out/in, but this isn't actually hurting anything. It was just too good at keeping the same air in the battery

Question regarding how the car reports degradation. If the car is displaying range in Percent, does that value change if there is real degradation. Will it charge to 100% no matter the battery capacity? or does it max out at 95% or whatever if the battery capacity has diminished? Is having the display set to "miles" the only way to see variance in range..real or not?

Others answered the 100% thing, but yes, distance display is the only way the car itself can show you actual remaining capacity. Especially on newer vehicles, this is actually a pretty accurate lens into the actual capacity of the battery.

What an incredibly helpful and amazing post.

THIS post should be sticky.

Thank you so much for posting this.

Heh, thanks. Unfortunately my information isn't delivered with any sort of virality or fantastic claims. Too often my posts are depressing for fans, if more realistic.

Everyone wants to know how they can "gain range", which is why this thread got stickied. If I made such a post, it would be about how you would "maybe 'gain' some range, maybe 'lose' some range, but actually nothing changes". It just doesn't got the same oomph, y'know? Most of what I said only needs to be brought up in the context of incorrect information, which is a hard thing to create from scratch.

 

[AlanSubie4Life]

No, the rate reduction is not REAL on EVERY vehicle

Sorry. Was applying the constant Tesla energy metrics per rated mile to other EVs (which eliminates any GOM effects). I meant to say available energy reduction is real on every EV, Tesla and otherwise.

This thread is about the apparent FALSE decrease because of the BMS.

No, it's not a false reduction. It's likely that the 270 estimate was real (since it was on a 100% charge), and correct. It represented that that much energy was available. Full stop. This thread is about how to charge the battery to get it to contain more energy after hitting local minima (for whatever reason).

Of course it is also possible that sometimes the BMS can get confused, and the estimate may actually truly have higher error. But there are many ways for the battery to end up in a state where less energy can be drawn from it - and for sure the BMS needs to present that info to the user! Otherwise it leads to a very very bad time.

For this thread, I'm making the assumption that the BMS was correct about 270 rated miles - it might not be the case in this specific situation - but it's not that unusual for available energy to be recovered on subsequent recharge and rebalancing, as occurred here. The state of that worst case brick, and the effect it can have on the state of other bricks if rebalancing has not occurred, is super important!

 

[Zoomit]

Because they are designed to allow 100%. And no where in the instruction does it say take to 100% 2 minutes before your trip.
If 100% was going to damage your battery, don't you think that Tesla would have mentioned that or set real 100% to more than apparent 100%? The damage of 100% comes from leaving it there for long periods, months, years, not minutes, hours or days. Do you see anywhere in the warranty that talks about it? I'm pretty sure that the ONLY place that talks about it is the nag that when you leave it at 100% and charge MULTIPLE times, it suggests that you decrease it.

I dare say that there is a high probability that the battery may last longer than most of the cars. Higher probability that the car will get totaled than the battery needing to be replaced. Actually that's pretty much a duh statement. Many more cars have been salvaged at this point.

If you stop trying to treat the battery so gingerly, it will work a LOT better. Isn't that what this thread is about?

Taking the battery to 100% is a recommended practice, the manual and other places recommend it when needed.

Tesla Range Tips: "Whenever possible, don’t let the battery go above 90% or below 20%."

This thread is not about actual battery degradation. It's about predicting battery capacity and false indications due to charging habits.

IF you have any third party apps, disable them, so they can't talk to the car. Besides that, close the door and never check the app, that allows the car to go to sleep in a few hours. It will wake up to check a few things periodically, and if you check it with the phone, that wakes it up.

Just like every technologist tell people these days. For better sleep remove all technology and just go to sleep.

This is outdated advice. I use TeslaFi and my car sleeps just fine. In fact, I use TeslaFi as a reference, on charge state for example, so I don't have to check the app and wake the car.

 

[SageBrush]

The main (only?) benefit to keeping your car plugged in is to not be surprised with a lower-than-expected state of charge

I'm also in the club that does not buy the notion that always plugging in retards battery degradation although I can see how the manual might be interpreted that way.

 

[camalaio]

The number varies, and that's the whole basis of this discussion.

When charging batteries, the most common way is to apply higher than the cell voltage until the charging current comes to near 0A (or below a specified threshold)> You then allow the batteries to settle down (lots of chemical reactions need to complete) and measure the voltage and that becomes 100%.
If you don't charge to 100%, by watching the current go to 0, then you are only guessing.

The low voltage is measured differently, it's a combination of engineering and analysis. Basically, it is determined by the lowest voltage that nominally that doesn't cause enhanced cell damage. So while part of it comes from chemistry, real life plays in and you basically need to run a bunch of cells through testing to see what their MTBF is.

And whatever that number is, Tesla has increased it. That's how cars that set in auction houses or used car dealer lots can stay at 0 for long periods of time, because 0 really isn't 0. At a certain percent, the car starts to turn things off. And before it hits real 0%, it has turned EVERYTHING off. And, in case you didn't realize it, to measure a battery, requires the battery meter to be on, so at 0, it has to be charged, just to get the car to turn back on.

Taking the battery to 100% is a recommended practice, the manual and other places recommend it when needed.

Taking a battery to 0 isn't a great idea. I don't recommend it.

This is not how charging Li-ion batteries works.

Some pre-determined max threshold voltage is used for 100%. For most Li-ion cells, this is 4.2V/cell for example.

Li-ion chargers generally use constant-current charging up until the voltage necessary to deliver that current meets/exceeds the max threshold (4.2V). Then it switches to constant-voltage charging, holding that 4.2V until the current going into the battery drops below some determined threshold (not actually 0A, they all stop before this, sometimes by a surprising amount). This usually results in the cells being very close to 4.2V after charging, within a few mV.

You can absolutely charge a "4.2V" cell to 4.3V or even 4.4V. The cells don't inherently limit this (not at those voltages anyways), it's an engineered determination. They just get a bit explode-y beyond the common 4.2V/cell, so we call that 100%.

I think your assertion was that it sort of "figures out" what voltage it settles to and calls this 100%, but that's simply not the case.

 

[AlanSubie4Life]

Ok, neat. Really hard to tell, but it looks like those 4 are in parallel (the vias seem to also bridge over the pads). If so, 39.5 Ohm. The larger picture shows it's one of these banks per group of parallel cells I believe. They also seem to have transistors (Q6, Q5, etc.) that are probably switching the balancing resistors on/off, basically. Of course, that's a wild guess, but likely not too far from the truth.

I wish we had pictures of the backside of the Model 3 BMS. I'm pretty certain the balancing isn't happening on the front side of the board that we have pictures of (I can't see anything that would serve this role).



I should've seen this response coming and given more detail haha. It's "seeing" just about everything it can still, just driven less on a daily basis (it still goes on day trips too).

I only disagree that the car gets to decide what to do, and to be fair I disagree with both Elon's wording and the wording of the owner's manual on the subject. The car doesn't get to decide anything about the SoC. I, the owner, set that. I even set what time it charges! It can technically modify things within those constraints (e.g. cap my max voltage permanently), but those are outliers and haven't been observed in the wild so far for Model 3.

The main benefit to keeping it plugged in is simply to avoid a case where the battery may accidentally discharge more than expected.



Absolutely! There are cases like this for the Model 3 on these forums, where the battery is clearly physically faulty in some way and that is the limiter on capacity. Unfortunately there's just no easy well to tell this via the car itself, as it appears the same: reduced range.



I linked to a similar photo in the part you quoted (links aren't super clear on these forums, are they?)

The Model 3 BMS boards sort of just sit on top of the modules.

But anyways, congrats, you've discovered "active balancing"! This actually exists, with different ways to accomplish it. Basically, tying a higher group of cells to a lower one temporarily: using the energy of the "too high" cells to raise the energy of the "too low" cells.

It's a less wasteful approach, but on average balancing energy is not a significant portion of standby energy usage (at least, not in a Tesla where there are so many other ways it wastes energy).



Ah, ok. Since it was coming from Tesla service folks, I had assumed "vampire drain" would include all user-observable drain since they field those concerns.

I was definitely presenting the impossibility of the claims as-is. I know the ~60W number is due to various other reasons (the car being routinely "awake" is the actual primary factor, in which many other things use far more power). So yeah, my calculations shouldn't reflect reality - they were to demonstrate that the appearance of the claim cannot be true.



It looks to me like they're in parallel, so 39.5 Ohm, implying 0.4W dissipation (~100mA). This appears to be much more reasonable. There's also nearby transistors (labelled Q) that are probably (dis)engaging those balance resistors. That is, they are not engaged full time, but engaged as necessary and individually (each group of components seems to go to a different group of cells in parallel).

I can't find the post anymore (sorry!), but someone rightly pointed out that when being switched on/off anyways, the resistor value only serves to change the amount of time needed to balance, but the same amount of energy will be used since dealing with a certain amount of imbalance via discharge requires removing a certain amount of energy. One reason to increase the resistance could be to give a more accurate reading while balancing though, as less current would lead to less voltage sag (we're talking tiny amounts though, not sure this is a problem in reality at these values).



I strongly suspect this is confirmation bias, but read on.

I fully believe there was something else goofed up with your case. In fact, though I've been avoiding the topic, I think your pack is (only) the second case I've read on that it appears balancing may have been the result of your range recovery.

Our current understanding of balancing outside of this thread is that the battery needs to be at rest and above approximately 73% (I'll have a thread on this later today). You appear to have been consistently using Sentry mode, which would mostly not allow balancing (or at least decrease the total time it can spend balancing).

So in my mind, the BMS was actually being very good at energy estimation. It was probably accounting for your (lightly) imbalanced pack. When given the opportunity to balance (you stopped using Sentry, etc.), it then correctly reported you can extract more energy before it needs to cut off the supply.

The finding is significant for that reason. It shows a case where doing "the bad thing" (e.g. Sentry 24/7 in an extreme case) could lead to reduced range, but also have that "lost range" be recoverable. Again, it's something we knew about, but it just doesn't matter to many people. Most people's charging habits and battery health will result in well-balanced packs, by design.

This is just one of many oversights and side effects, IMO, regarding flashy features like Sentry Mode.



I guess it's just easier to keep the numbers going eh?

1. I will still state that balancing and calibration are separate concerns. But it can only balance after charging is done, and balancing itself is a bit of a slow process.
2. I pointed out "stranding" because you have the burden of a stickied post in which you implied a recommendation by stating what worked for you. The internet sucks, doesn't it? Suddenly you're on the hook for everyone's personal cases without even having given context to your own.
3. See previous chunk of my post. You probably ended up balancing it (and thus increased extractable energy), not necessarily made it more accurate. Significant yes, but potentially not the cause you are thinking.
4. I will very strongly disagree here, elaborating below. Even the official app (but not car) gets this wrong!

The third-party tools use the Tesla API data which, yes, is the very same data you see in the app (and thus in the car as well to some extent). There are higher inaccuracies at lower states of charge - you cannot assume the "absolute error" for range at a lower SoC is the same at higher SoC.

Let's take my car's data, right now, from the API. In-app it's displaying 262km of range, which is about 162.8mi, thus the 162.85 below.

'battery_level': 55,
'battery_range': 162.85,
'est_battery_range': 236.2,
'ideal_battery_range': 162.85,
'usable_battery_level': 55,​

There are a few very important things to note here:

• Any representation of percentage is rounded. This is the primary source of error. At 20%, this could be either 19.5% or 20.49%. Since you divide by this to estimate range at max, this introduces huge variance at lower percentages (note how at 1%, it ranges from 0.5% to 1.49% - that's a 50% swing, showing where this increased variance comes from).
• I have no idea what "est_battery_range" is. Assuming it's miles that's ~380km, far too low for representing the 100% state and far too high for its current state. Assuming it's somehow in km, it doesn't jive with anything else.
• The numbers the apps use are the temperature-varying ones. For tracking actual battery health, this is unfair. Lower capacity is reported in cold, though this is not an actual indicator of degraded battery health.

So let's do the math these services would do. Take the current range (162.85) and divide by the percentage (0.55). That comes out to 296.1mi, or 476.5km. Also, in the app, if I drag the slider to 100%, reports that my 100% range is also about 476km, because it's doing the exact same calculation on the exact same data. So far so good, right?

Now recall the issue I mentioned with rounded numbers. I happen to know via the CAN bus that the actual SoC right now is 54.5%. If I use that instead, I get about 481km (and this is pretty much the actual 100% range of my car right now). But had it actually been 55.49% (and still reported "55"), that would falsely indicate 472km. That's already a 9km (5.6mi) spread.

At lower SoC, it gets even worse. Say at 10%, it was reporting 48.10km. But that "10" could be 9.5 or 10.499, which would indicate 100% ranges of 506km and 458km respectively. That is now a ridiculous 48km (~30mi) spread. In other words, that's 10% of the max range.

100% range estimates are estimates, whether it's a third-party service or the Tesla app, and their accuracy is worse at lower SoC.

Now, that is not to say the current range number in the car is inaccurate. That one is a pretty dang accurate representation of the current energy in the pack. The error is introduced when using a rounded, low significant digit percentage to estimate the 100% capacity.

These services have their place and can provide value, but the recommendation to throw away data for <90% SoC (if the service allows this) is absolutely warranted. Anything much below that has too much error for discussion on single-digit percentage degradations.



I'm getting winded with this at this point (I have a post coming I can finally just link to), but balancing is probably occurring at 73% and above. There is absolutely no need to charge to 90% on Model 3. I'd also point out that the only thing the Roadster and the Model 3 really have in common is that they are electric vehicles - everything is different otherwise.



Thanks, I really try. It's exhausting if I'm honest. For every factual post, there are 10 other more popular posts with varying amounts of fact (not saying this thread specifically!). I guess I've just learned to try to explain, even if it's an information overload.



The main (only?) benefit to keeping your car plugged in is to not be surprised with a lower-than-expected state of charge (of which there could be numerous causes, including not fully closing your door and thus keeping the climate control on for a while).

The other dubious but related benefit is blissful ignorance of how much energy the car uses in standby!



It's not related to temperature management at all, just pressure changes (which come about via elevation and temperature changes, yes). Basically it seems like their thresholds were too high and would allow a sheet to flex and pop out/in, but this isn't actually hurting anything. It was just too good at keeping the same air in the battery



Others answered the 100% thing, but yes, distance display is the only way the car itself can show you actual remaining capacity. Especially on newer vehicles, this is actually a pretty accurate lens into the actual capacity of the battery.



Heh, thanks. Unfortunately my information isn't delivered with any sort of virality or fantastic claims. Too often my posts are depressing for fans, if more realistic.

Everyone wants to know how they can "gain range", which is why this thread got stickied. If I made such a post, it would be about how you would "maybe 'gain' some range, maybe 'lose' some range, but actually nothing changes". It just doesn't got the same oomph, y'know? Most of what I said only needs to be brought up in the context of incorrect information, which is a hard thing to create from scratch.

I hate to spam the thread, but just wanted to say what a great post this is, so I'm going to do it anyway. Thank you for clarifying these points for users.

 

[Battpower]

the vias seem to also bridge over the pads

That's how it looks to me, and the 100mA figure I have seen mentionned elsewhere too.

The main benefit to keeping it plugged in is simply to avoid a case where the battery may accidentally discharge more than expected.

Yes.

It's a less wasteful approach, but on average balancing energy is not a significant portion of standby energy usage (at least, not in a Tesla where there are so many other ways it wastes energy).

If degraded / old batteries do not become dangerous, then presumeably more agressive, active balancing could keep them going a bit longer and without keeping the bleeding circuits on for ages.

balancing and calibration are separate concerns. But it can only balance after charging is done

Can you clarify? Are you saying it must charge, then balance? I believe I have seen my MS balancing while charging from 50%. But, since the significance of several values changes depending on what else is going on, I am ready to revise that statement!

blissful ignorance of how much energy the car uses in standby!

I'm quite surprised more interest isn't taken in this point. Compared with my other small EV, the MS does need continous topping up so I'm quite conscious of quiescent energy use. When bricks go bad, this can shoot up quite dramatically too.

 

[camalaio]

If degraded / old batteries do not become dangerous, then presumeably more agressive, active balancing could keep them going a bit longer and without keeping the bleeding circuits on for ages.

Solid point. I wonder how much of a factor this is towards end of life?

Can you clarify? Are you saying it must charge, then balance? I believe I have seen my MS balancing while charging from 50%. But, since the significance of several values changes depending on what else is going on, I am ready to revise that statement!

For sure.

My understanding, which could be wrong, is that balancing has to be done once open-circuit voltages are well-known. You can't do this while charging, because differences between the cells could augment these numbers (not to mention you're applying voltage to the cells to charge it, which changes their apparent voltage anyways). For the same reason, you can't start balancing while driving - discharge sags the voltage the parallel bricks of cells slightly differently from each other (capacity, internal resistance, etc.). Also we have regenerative braking while driving, so the charging problem exists too!

That why you can't have a load or charge occurring. One is augmenting the voltage directly (charging), and one gives a different view of imbalance (discharge), making determining the imbalance at rest impossible. That discharge view could have some merit and tell you something valuable, but that's getting really complicated really quickly and would require a ton of data.

I think the most improvement you could make is to estimate and make a note of how much energy needs to be bled off of which cells to attain balance (when the car is at rest), then keep balancing active regardless of whatever the car is doing (sleep, driving, etc.) until that amount of energy has been bled off. It still requires a stable period to determine the imbalance and estimate that energy though!

The Model S could (and does) behave very differently from the Model 3, especially regarding the battery. If they didn't have the same logo, a strong case could be made that they were built by different companies entirely.

Balancing is also ridiculously hard to observe compared to basically every other thing we can measure.

I'm quite surprised more interest isn't taken in this point. Compared with my other small EV, the MS does need continous topping up so I'm quite conscious of quiescent energy use. When bricks go bad, this can shoot up quite dramatically too.

Ugh. I've wrote a dang book about this on this forum previously. It got a total of zero replies haha. Though I'd have some corrections and additions to that post today.

I care so much about this point because of the "green" aspect, and "the future". Thought someone like Tesla would be responsible about this more than others. Nope, they care about their 0-60 times and flashy techy features. People further justify the wasted energy by it being so low in monetary cost, which isn't the concern I have at all (though it does change cost of ownership calculations significantly for the low-mileage ones).

It's really a shame. The battery, viewed as a component, actually has very little quiescent current. Many parts of the car are additionally designed to take very little power in standby. I'd be pissed if I was responsible for those designs and then looked at the massive average "standby" power of the fleet. Designing the system to run on 2W standby means squat when Sentry draws about 250W. (I'm guessing at the 2W). Shall we also talk about Summon Standby? Gah, I'm just getting mad now.

I hate to spam the thread, but just wanted to say what a great post this is, so I'm going to do it anyway. Thank you for clarifying these points for users.

Means a lot coming from you, thanks

 

[Battpower]

balancing has to be done once open-circuit voltages are well-known. You can't do this while charging, because differences between the cells could augment these numbers (not to mention you're applying voltage to the cells to charge it, which changes their apparent voltage anyways). For the same reason, you can't start balancing while driving - discharge sags the voltage the parallel bricks of cells slightly differently from each other (capacity, internal resistance, etc.). Also we have regenerative braking while driving, so the charging problem exists too!

I have a different impression.

Given that over time the BMS can easily learn which bricks are lazy, at any point in time it knows how much energy needs to be bled off and where from to keep things balanced. All those things you mention are going on some of the time, but bleeding off energy is what it is any time of day. Even if you are drawing energy from all cells / bricks, you can always draw a little extra from some to keep them in line. Likewise, when charging, you can give the current a parallel path through a resistor so that not quite as much flows through the brick. As long as you have good data about the battery overall, you can apply balancing for as long as the balancing circuits can be powered on. I think this requires the battery to be active as I think the balancing circuits are powered by the modules themselves. This is a guess - assuming need to keep HV connected circuits as isolated / self contained as possible.

The open circuit state where the only currents in the battery are between the cells themselves is when the voltages settle and become a meaningful indicator of each brick's soc.

I think the most improvement you could make is to estimate and make a note of how much energy needs to be bled off of which cells to attain balance (when the car is at rest), then keep balancing active regardless of whatever the car is doing (sleep, driving, etc.) until that amount of energy has been bled off. It still requires a stable period to determine the imbalance and estimate that energy though!

Like this ^^^^

because of the "green" aspect, and "the future". Thought someone like Tesla would be responsible about this more than others.

As some older S's do more pack heating and cooling - almost simultaneously in some cases! - and older cells have higher levels of internal heat generation - the parasitic dissipation can evidently become quite huge. Having to heat the whole pack before charging, then cool it down again when done must waste huge amounts of energy.

 

[rrmagnuson]

Like many others, I have been concerned with loss of 100% indicated battery range on one of my Model 3s. My P3D (build date 9/13/2018, delivery date 10/8/2018) had gotten down to 270.3 miles at 100% charge on January 20, 2020, at about 30,700 miles, which is a loss of 40.8 miles since the car was new.

I posted about going to the service center to talk with them about battery degradation, which I did on March 9, 2020. It was a great service appointment and the techs at the Houston Westchase service center paid attention to my concerns and promised to follow up with a call from the lead virtual tech team technician. I detailed this service visit in the following post:

Reduced Range - Tesla Issued a Service Bulletin for possible fix

While that service visit was great, the real meat of addressing the problem came when I spoke to the virtual tech team lead. He told me some great things about the Model 3 battery and BMS. With the knowledge of what he told me, I formulated a plan to address it myself.

So here is the deal on the Model 3 battery and why many of us might be seeing this capacity degradation.

The BMS system is not only responsible for charging and monitoring of the battery, but computing the estimated range. The way it does this is to correlate the battery's terminal voltage (and the terminal voltage of each group of parallel cells) to the capacity. The BMS tries to constantly refine and calibrate that relationship between terminal voltage and capacity to display the remaining miles.

For the BMS to execute a calibration computation, it needs data. The primary data it needs to to this is what is called the Open Circuit Voltage (OCV) of the battery and each parallel group of cells. The BMS takes these OCV readings whenever it can, and when it has enough of them, it runs a calibration computation. This lets the BMS now estimate capacity vs the battery voltage. If the BMS goes for a long time without running calibration computations, then the BMS's estimate of the battery's capacity can drift away from the battery's actual capacity. The BMS is conservative in its estimates so that people will not run out of battery before the indicator reads 0 miles, so the drift is almost always in the direction of estimated capacity < actual capacity.

So, when does the BMS take OCV readings? To take a set of OCV readings, the main HV contactor must be open, and the voltages inside the pack for every group of parallel cells must stabilize. How long does that take? Well, interestingly enough, the Model 3 takes a lot longer for the voltages to stabilize than the Model S or X. The reason is because of the battery construction. All Tesla batteries have a resistor in parallel with every parallel group of cells. The purpose of these resistors is for pack balancing. When charging to 100%, these resistors allow the low cells in the parallel group to charge more than the high cells in the group, bringing all the cells closer together in terms of their state of charge. However, the drawback to these resistors is that they are the primary cause of vampire drain.

Because Tesla wanted the Model 3 battery to be the most efficient it could be, Tesla decided to decrease the vampire drain as much as possible. One step they took to accomplish this was to increase the value of all of these resistors so that the vampire drain is minimized. The resistors in the Model 3 packs are apparently around 10x the value of the ones in the Model S/X packs. So what does this do to the BMS? Well, it makes the BMS wait a lot longer to take OCV readings, because the voltages take 10x longer to stabilize. Apparently, the voltages can stabilize enough to take OCV readings in the S/X packs within 15-20 minutes, but the Model 3 can take 3+ hours.

This means that the S/X BMS can run the calibration computations a lot easier and lot more often than the Model 3. 15-20 minutes with the contactor open is enough to get a set of OCV readings. This can happen while you're out shopping or at work, allowing the BMS to get OCV readings while the battery is at various states of charge, both high and low. This is great data for the BMS, and lets it run a good calibration fairly often.

On the Model 3, this doesn't happen. With frequent small trips, no OCV readings ever get taken because the voltage doesn't stabilize before you drive the car again. Also, many of us continuously run Sentry mode whenever we're not at home, and Sentry mode keeps the contactor engaged, thus no OCV readings can be taken no matter how long you wait. For many Model 3's, the only time OCV readings get taken is at home after a battery charge is completed, as that is the only time the car gets to open the contactor and sleep. Finally, 3 hours later, OCV readings get taken.

But that means that the OCV readings are ALWAYS at your battery charge level. If you always charge to 80%, then the only data the BMS is repeatedly collecting is 80% OCV readings. This isn't enough data to make the calibration computation accurate. So even though the readings are getting taken, and the calibration computation is being periodically run, the accuracy of the BMS never improves, and the estimated capacity vs. actual capacity continues to drift apart.

So, knowing all of this, here's what I did:

1. I made it a habit to make sure that the BMS got to take OCV readings whenever possible. I turned off Sentry mode at work so that OCV readings could be taken there. I made sure that TeslaFi was set to allow the car to sleep, because if it isn't asleep, OCV readings can't get taken.

2. I quit charging every day. Round-trip to work and back for me is about 20% of the battery's capacity, and I used to normally charge to 90%. I changed my standard charge to 80%, and then I began charging the car at night only every 3 days. So day 1 gets OCV readings at 80% (after the charge is complete), day 2 at about 60% (after 1 work trip), and day 3 at about 40% (2 work trips). I arrive back home from work with about 20% charge on that last day, and if the next day isn't Saturday, then I charge. If the next day is Saturday (I normally don't go anywhere far on Saturday), then I delay the charge for a 4th day, allowing the BMS to get OCV readings at 20%. So now my BMS is getting data from various states of charge throughout the range of the battery.

3. I periodically (once a month or so) charge to 95%, then let the car sleep for 6 hours, getting OCV readings at 95%. Don't do this at 100%, as it's not good for the battery to sit with 100% charge.

4. If I'm going to take a long drive i.e. road trip, then I charge to 100% to balance the battery, then drive. I also try to time it so that I get back home with around 10% charge, and if I can do that, then I don't charge at that time. Instead, let the car sleep 6 hours so it gets OCV readings at 10%.

These steps allowed the BMS to get many OCV readings that span the entire state of charge of the battery. This gets it good data to run an accurate calibration computation.

So what's the results?

On 1/20/2020 at 30,700 miles, I was down to 270 miles full range, which is 40.8 miles lost (15.1 %). The first good, accurate recalibration occurred 4/16/2020 at 35,600 miles and brought the full range up to 286 miles. Then another one occurred on 8/23/2020 at 41,400 miles and brought the range up to 290 miles, now only a 20 mile loss (6.9 %).

Note that to get just two accurate calibration computations by the BMS took 7 months and 11,000 miles.

So, to summarize:

1. This issue is primarily an indication/estimation problem, not real battery capacity loss.
2. Constant Sentry mode use contributes to this problem, because the car never sleeps, so no OCV readings get taken.
3. Long voltage stabilization times in the Model 3 prevent OCV readings from getting taken frequently, contributing to BMS estimation drift.
4. Constantly charging every day means that those OCV readings that do get taken are always at the same charge level, which makes the BMS calibration inaccurate.
5. Multiple accurate calibration cycles may need to happen before the BMS accuracy improves.
6. It takes a long time (a lot of OCV readings) to cause the BMS to run a calibration computation, and therefore the procedure can take months.

I would love if someone else can perform this procedure and confirm that it works for you, especially if your Model 3 is one that has a lot of apparent degradation. It will take months, but I think we can prove that this procedure will work.

I'm trying to relate this procedure to my personal situation. I am retired so don't have the daily commute pattern, but I do usually go out some every day, whether for shopping, doctor appointments, day trips, or whatever. I'm rarely home no later than 6-7PM and often much earlier. I plug in every day, and charge to 80% at midnight. So it sleeps at least 5 hours at some reduced SOC, charges to 80%, then sleeps typically another 6-8 hours. And since it sleeps a lot at home, Sentry mode is not enabled. Also I've changed how I interpret 80%. Since I have a SR+ with 250 mile initial range, I used to charge to 200 miles. But after 8 months, the range is reduced to about 225 so now I charge to 180ish. Anyway, I don't see a lot of difference between your logic, which is good, and my usage pattern. Do you?