Almost 15% range loss Model 3 Awd

[ajdelange]

So far so good, I am not arguing to that. But so far the heat loss is not in this calculation. And the regen as well.

The current that causes the heat loss is included in the current integral and the heat lost is included in the energy integral. You cannot separate them out just by measuring current. You would have to do a separate estimation of what portion of the energy measured by this method goes to heat by estimating the amount that goes to traction and subtracting that from the integral. Regen is also captured when total current is measured. Current draw becomes negative during regen and the integrals get smaller at such time as regeneration is taking effect.

I hate to argue with the scientific explanation you just did, but this doesn't seem to be the way the BMS calculates consumption shown under "used" for the trip.

Seem? Is there any evidence that it is done otherwise? As I said I don't have any details but this is certainly the common sense approach and any data I am able to see in the car or get through the API supports it.

And from the observations, this is not what the BMS is doing.

Again, it is consistent with any observations I can get from the car or the API.

I think they don't want to do the hard work of calculation and messing it up, especially when 99% of their customer base will never drive above 75-80mph and experience massive heat loss.

No, I don't think they do and I thus think they don't and thus I think the displayed energy use must include the I^2R losses just as it includes all the other losses.

Losses are appreciable at lower speed too. If you drive from point A to point B and back to point A the state of the car (once it cools down) is the same as when you departed. No energy was used in an ideal world. But in the real world you used quite a bit of energy and all of the energy you used is, thus, loss and not only loss but heat loss. You warmed the inverter transistors, you warmed the stator, you warmed the tires, you warmed the bearings, you warmed the gearbox, you warmed the air, you warmed the road bed, you warmed the water on road if it were present. How can the car separate out the I^2R losses? Why would anyone want it to? If it did somehow exclude I^2R losses I would lobby hard to get them included.

This is evident by the SOC readings from the CAN bus on a high-speed stretch, which doesn't match the kWh used in the trip meter, based on % of available capacity.

I think you must be misinterpreting what you are reading from the CAN bus because the notion that one type of loss is somehow separated out and deducted from all the others just doesn't make sense. It would be hard to do and we, as users, certainly wouldn't want it done.

What, after all is SoC? It is the amount of charge in the battery relative to the battery's usable capacity. And what is the battery's usable capacity? The amount of charge it take to move it from one arbitrarily chosen state (empty - the state variable here is OC voltage) to another arbitrarily chosen state (full). This changes with time (the battery ages), temperature and load. It's a pretty simple matter to determine SoC if you know how many coulombs of charge are in the battery relative to the empty state, what the capacity is, and how many you have taken out or put in over a time period. As we have noted above it is pretty easy to accurately measure the total charge added or deducted but it is not at all simple to determine the capacity because our only observable is OC voltage and that doesn't change much with SoC. The Tesla's thus have very sophisticated algorithms that attempt to do this and they are probably the best in the world but even they cannot beat what we call "dilution of precision". Thus capacity estimates are all over the map for the same car over, say a month. Look at some of the plots people have posted to this thread.

 

[diamond.g]

The current that causes the heat loss is included in the current integral and the heat lost is included in the energy integral. You cannot separate them out just by measuring current. You would have to do a separate estimation of what portion of the energy measured by this method goes to heat by estimating the amount that goes to traction and subtracting that from the integral. Regen is also captured when total current is measured. Current draw becomes negative during regen and the integrals get smaller at such time as regeneration is taking effect.

Seem? Is there any evidence that it is done otherwise? As I said I don't have any details but this is certainly the common sense approach and any data I am able to see in the car or get through the API supports it.



Again, it is consistent with any observations I can get from the car or the API.




No, I don't think they do and I thus think they don't and thus I think the displayed energy use must include the I^2R losses just as it includes all the other losses.

Losses are appreciable at lower speed too. If you drive from point A to point B and back to point A the state of the car (once it cools down) is the same as when you departed. No energy was used in an ideal world. But in the real world you used quite a bit of energy and all of the energy you used is, thus, loss and not only loss but heat loss. You warmed the inverter transistors, you warmed the stator, you warmed the tires, you warmed the bearings, you warmed the gearbox, you warmed the air, you warmed the road bed, you warmed the water on road if it were present. How can the car separate out the I^2R losses? Why would anyone want it to? If it did somehow exclude I^2R losses I would lobby hard to get them included.

I think you must be misinterpreting what you are reading from the CAN bus because the notion that one type of loss is somehow separated out and deducted from all the others just doesn't make sense. It would be hard to do and we, as users, certainly wouldn't want it done.

What, after all is SoC? It is the amount of charge in the battery relative to the battery's usable capacity. And what is the battery's usable capacity? The amount of charge it take to move it from one arbitrarily chose state (empty) to another arbitrarily chosen state (full). This changes with time (the battery ages), temperature and load. It's a pretty simple matter to determine SoC if you know how many coulombs of charge are in the battery relative to the empty state, what the capacity is, and how many you have taken out or put in over a time period. As we have noted above it is pretty easy to accurately measure the total charge added or deducted but it is not at all simple to determine the capacity because our only observable is OC voltage and that doesn't change much with SoC. The Tesla's thus have very sophisticated algorithms that attempt to do this and they are probably the best in the world but even they cannot beat what we call "dilution of precision". Thus capacity estimates are all over the map for the same car over, say a month. Look at some of the plots people have posted to this thread.

Just for reference, CAN IDs 132, 292, 2D2 and 352 all reference battery "health"/state. 352 flat out tells Full kWh, Remaining kWh and Expected Remaining kWh. While 292 shows UI SOC, Min SOC, Max SOC.
Jack R has said before in one of his (long) videos that the UI kinda of lies to the end user about some things (like charge state and probably kWh used), which when looking at the CAN data kind of supports that.

For the folks that are having capacity issues, it would be really interesting to see CAN ID 352 Remaining kWh and Expected Remaining kWh to see how close they are, and to see which one is used in calculating how many miles shows up in the UI (unless they are flat pulling 292).

 

[turtlesz]

Curious if you know the charging regimen for each?

On weekdays she supercharges 1x per week as she lives away for work. Weekends she is mostly home where the car is charged. Still seems excessive amount of loss to me. We have have multiple Xs and 3s between family members but have never seen this kind of loss. My LR 3 is a range champ. OC to Vegas with 0 stops and LA to Bay with 1 20 min charge stop. Her 3 cannot do the same.

 

[ajdelange]

Just for reference, CAN IDs 132, 292, 2D2 and 352 all reference battery "health"/state.

If the guy that did the CAN bus reader for Android ever does one for iOS this will be useful info.

352 flat out tells Full kWh, Remaining kWh and Expected Remaining kWh.

So what do "Remaining kWh" and "Expected Remaining kWh" mean? The first I can guess at. The car is saying "I'll let you take another X kWh". But what's the second? The word "Expected" has special meaning to a statistician. It's a statistic used in place of a solid number when a solid number isn't available. And I expect that's the case here but without the manual I can't be sure.

While 292 shows UI SOC, Min SOC, Max SOC.
Jack R has said before in one of his (long) videos that the UI kinda of lies to the end user about some things (like charge state and probably kWh used),

Is it really lying? Were I designing this I would set the UI Empty SoC state at a level above the level to which the battery can be discharged before damage starts to occur. This gives the driver a "reserve" just as an ICE car often has a few gallons left when its gas gauge reads E. I wouldn't consider that lying. In fact, that's what I would want to see. This is what I consider "usable capacity" - the difference between what the UI shows me as full and what it shows me as empty.

 

[TimothyHW3]

So what do "Remaining kWh" and "Expected Remaining kWh" mean? The first I can guess at. The car is saying "I'll let you take another X kWh". But what's the second? The word "Expected" has special meaning to a statistician. It's a statistic used in place of a solid number when a solid number isn't available. And I expect that's the case here but without the manual I can't be sure.

There is nominal full which is a value X, which at the start is anywhere between 77-78 depending on how lucky you are.
This value includes the buffer of roughly 3.5kWh.
The available value is then nominal minus buffer and this is what you get from0-100%
I am not quite sure why there are expected and nominal, but on my recent trip I did see 0.2kWh difference, where expected is more than nominal remaining.

Is it really lying? Were I designing this I would set the UI Empty SoC state at a level above the level to which the battery can be discharged before damage starts to occur. This gives the driver a "reserve" just as an ICE car often has a few gallons left when its gas gauge reads E. I wouldn't consider that lying. In fact, that's what I would want to see. This is what I consider "usable capacity" - the difference between what the UI shows me as full and what it shows me as empty.

They are, sort of, kind of, lying. I explained this at least twice here. At least in terms of rated EPA.

The 100% SOC is based on available capacity (nominal- buffer)
310 miles on the other hand is based on full nominal including buffer. So 100% and 310 miles use two different capacities for their calculation. This is the misinformation part.

When you go to 0%/0km you can use the buffer for a bit and go in negative %, but technically at this point you drove 280-290 miles instead of the 310miles. So this is the lying part.

 

[gecko10x]

The 100% SOC is based on available capacity (nominal- buffer)
310 miles on the other hand is based on full nominal including buffer. So 100% and 310 miles use two different capacities for their calculation. This is the misinformation part.

When you go to 0%/0km you can use the buffer for a bit and go in negative %, but technically at this point you drove 280-290 miles instead of the 310miles. So this is the lying part.

I find this piece very interesting, and quite misleading on Tesla's part (assuming true).

I wonder (long shot) if the recent full SOC mileage drops everyone is seeing could be Tesla removing the buffer from the UI readout...

 

[gecko10x]

On weekdays she supercharges 1x per week as she lives away for work. Weekends she is mostly home where the car is charged. Still seems excessive amount of loss to me. We have have multiple Xs and 3s between family members but have never seen this kind of loss. My LR 3 is a range champ. OC to Vegas with 0 stops and LA to Bay with 1 20 min charge stop. Her 3 cannot do the same.

I should have been more specific. I meant what is typical usage and charge settings? 90-80 daily? 100-10? 60-40?

 

[ajdelange]

I find this piece very interesting, and quite misleading on Tesla's part (assuming true).

I wonder (long shot) if the recent full SOC mileage drops everyone is seeing could be Tesla removing the buffer from the UI readout...

Remember Tesla has the flexibility to chage the full (100%) and empty (0%) voltages around as it sees fit. If they move them closer together they gain battery protection and thus, presumably, battery life but they do so at the expense of "full" capacity range. If they keep the range the same and slide the window towards lower voltage they protect more at the high end and sacrifice either reserve or battery safety at the low end. I think they do constantly adjust these as they gain more and more data from the cars in the fleet and as they tweak battery chemistry.

I don't find it at all misleading. The problem is that it can certainly be confusing to people who don't know anything at all about batteries and I think that would include 98% of new BEV owners.

 

[TimothyHW3]

I find this piece very interesting, and quite misleading on Tesla's part (assuming true).

I wonder (long shot) if the recent full SOC mileage drops everyone is seeing could be Tesla removing the buffer from the UI readout...

No, there is always a buffer. And no, so far we haven't observed any firmware changes, but let's see what happens with V10.

If anything, they will change the voltage on the main capacity, but the buffer is pretty much a standard and it hovers around 3.3-3.5kWh.

The only thing that Tesla does with the buffer, besides protecting the battery while charging, is to release it below 0% for emergencies. Which I still find good. Other cars like the Etron carry around 12kWh buffer that is never released and is used to hide battery degradation and allow for insane charging speeds, because the car is so inneficient. Imagine carrying 150kg of waste, just because your engineers couldn't do a proper testing or develop a better battery pack.

As ajdelange mentioned, the people posting here are new BEV drivers, some even new drivers at all and have fairly limited understanding of rated miles, capacity and buffer so it is hard to make anything out of it really. If we can fit them all with CAN readers that would at least help us out.

 

[AlanSubie4Life]

Cool, and I appreciate the thoughtful conversation. No reason why reasonable people can't philosophize about these seemingly disparate data points.

A datapoint from my drive today. Everything shows that the discharge constant is ~230Wh/rmi

Start: 307rmi

Drove 170.8mi @ 309Wh/mi

Remaining rated miles: 76rmi

Wh/rmi = 170.8mi*309Wh/mi / 231rmi = 228Wh/rmi

Don’t have time to summarize now and can’t upload my pictures due to insufficent BW on North Rim, but every other segment also worked out to ~230Wh/rmi - I went as low as 14% on the SoC.

It did not seem to depend on efficiency (but for the record I averaged something like 330Wh/mi over the entire drive, traveling 75-80+ basically the whole time on the interstate. And it was nearly 9000 feet uphill.

On the return it will be interesting to compare since I should be closer to 280Wh/mi average with all the downhill over 600 miles.

Anyway, this is a key constant you need to know, if you want to predict AWD rated mile use. As far as I can tell, for continuous drives, it’s completely deterministic.

It is not the same for other types of Model 3. Only applies to AWD.

It is easy to find out for yourself - just do a road trip and record the details from one large segment.

This is totally off topic from the original post at this point, but just wanted to close this out as I said I would post a result if I could.

 

[TimothyHW3]

If you still have the pictures. At this consumption, what was the kWh used shown in the trip meter and what was the SOC at that point and with how much did you start the trip?

 

[ajdelange]

Everything shows that the discharge constant is ~230Wh/rmi

Start: 307rmi

Drove 170.8mi @ 309Wh/mi

Remaining rated miles: 76rmi

Wh/rmi = 170.8mi*309Wh/mi / 231rmi = 228Wh/rmi

Your car, as do all the other Teslas, has a rated consumption. It is, approximately, the usable capacity of the battery divided by the EPA range. For example in my X (don't know how I wound up over here but it is an interesting thread) my battery's usable capacity is 98287 Wh and the EPA range is 295 miles (pre Raven) so my rated consumption is 98287/295 = 333.21 Wh/mi. That number should not change. It is a rating for the car. During a trip the navigation system measures the net energy drawn from the battery (everything - traction, windshield wiper motor, your cellphone charging, drag, rolling resistance, stiction - all of it) and divides it by the rated consumption to calculate the rated miles used.

To find out what your rated consumption is you can do what you did based on drives or better yet calculate it from charge data.

In this example 43330 Wh were added to increase the rated range by 129.87 so the rated consumption is 43300/129.87 = 333.41. To bring this back around to the OP these data also give us an estimate of the battery's usable capacity. Adding 43.33 kWh to this battery increased its SoC by 76 - 32 = 44% so the capacity must be 43.33/0.44 = 98.47 ± 0.91 kWh. Clearly there is uncertainty is the estimate because we really don't know whether 76 was 75.54 rounded up to 76 or 76.4 rounded down. Based on this we can estimate the uncertainty and it is appreciable - about 1%. The message is don't rely on a single charge. Process as many as you practically can.

It is not the same for other types of Model 3. Only applies to AWD.

As noted above this concept applies equally well to any of the Teslas. The example here is from an X.

 

[TimothyHW3]

Your car, as do all the other Teslas, has a rated consumption. It is, approximately, the usable capacity of the battery divided by the EPA range. For example in my X (don't know how I wound up over here but it is an interesting thread) my battery's usable capacity is 98287 Wh and the EPA range is 295 miles (pre Raven) so my rated consumption is 98287/295 = 333.21 Wh/mi. That number should not change. It is a rating for the car.



As noted above this concept applies equally well to any of the Teslas. The example here is from an X.

We actually, him and me, discussed this topic already So without going too deep into the rabbit hole you can Maybe read our discussion before.

His "rated miles", as he calls it, is the result of that "misinformation" we talked about previously - 100% SOC rating using the full capacity including buffer, but once you start driving the buffer "magically" disappears below 0%/0km.

So basically what he is doing he is calculating the "disappearance" of the buffer below the line. On a 3 the buffer is roughly 4.5-5% so I told him to just use the rated EPA of about 245 and deduct the buffer.

And what he means by only for AWD is that there is a slight difference between cars, even though the EPA rated, wrongfully, AWD and P 20" the same.

@Alan, would be great to get the used kWh, SOC and the other data

 

[ajdelange]

He's doing it right i.e. dividing the reported kWh used (calculated by multiplying actual miles driven by actual Wh/mi) by the reported rated miles consumed. The result is the car's rated Wh/mi. There is no need to know anything about battery capacity or reserve to do this.

In this example the car used 3290 Wh and 10.48 rated miles so the rate Wh/mi are clearly 3290/10.48 = 313.931. Except that for my vehicle the rated Wh/mi are 333.4 the point being that one cannot accurately determine this from a single drive. In fact it is better to deduce this number from charge data where everything is static but even so you must average over several charges to get a good number.

There is nothing rated wrongfully. The rated consumption reflects the rated range and battery capacity but it is just a number. It is not the consumption rate one should expect to realize any more than he should expect that he can drive the EPA range each time he gets in his vehicle. Each car has a rated Wh/mi number. All you have to do is figure out what it is and that is, as I have shown, very simple to do. I don't know why people have so much trouble understanding this but you are not alone by any means.

 

[diamond.g]

We actually, him and me, discussed this topic already So without going too deep into the rabbit hole you can Maybe read our discussion before.

His "rated miles", as he calls it, is the result of that "misinformation" we talked about previously - 100% SOC rating using the full capacity including buffer, but once you start driving the buffer "magically" disappears below 0%/0km.

So basically what he is doing he is calculating the "disappearance" of the buffer below the line. On a 3 the buffer is roughly 4.5-5% so I told him to just use the rated EPA of about 245 and deduct the buffer.

And what he means by only for AWD is that there is a slight difference between cars, even though the EPA rated, wrongfully, AWD and P 20" the same.

@Alan, would be great to get the used kWh, SOC and the other data

I bold the EPA range bit. The P was never tested EPA with the 20's. It wasn't required because at the time the 20's were an optional package. So yes with 18's the P and the AWD "should" see the same EPA rating based on the EPA cycle (aka they don't go deep in the accelerator).

 

[TimothyHW3]

He's doing it right i.e. dividing the reported kWh used (calculated by multiplying actual miles driven by actual Wh/mi) by the reported rated miles consumed. The result is the car's rated Wh/mi. There is no need to know anything about battery capacity or reserve to do this.

View attachment 462180
In this example the car used 3290 Wh and 10.48 rated miles so the rate Wh/mi are clearly 3290/10.48 = 313.931. Except that for my vehicle the rated Wh/mi are 333.4 the point being that one cannot accurately determine this from a single drive. In fact it is better to deduce this number from charge data where everything is static but even so you must average over several charges to get a good number.

There is nothing rated wrongfully.

Not sure why I have to explain this again, but you will do me and us a favor if you read the previous posts and comments...

There actually is a lot to do with BMS and capacity and degradation, because the rated miles will not be the same, once your car gets older. It is 3.1m per% at the beginning, but it gradually drops with degradation and imbalance.

Your calculation is ok when the car is fairly new, but once the rated miles drop(the ones the BMS shows at 100% and calculates based on the EPA constant*kWh nominal) then your calculation will not be any more meaningful then what TeslaFi is trying to calculate. Not much meaningful at all.

Also, there might be heat loss and imbalance in the BMS which has to be accounted for and they will differ from drive to drive hence the one time measurement doesn't work(like you rightfully mentioned). Reading the BMS does work all the time

But all you are really doing or trying to do is putting a Wh/m number to something we already know - that the buffer, which is being used at 100% disappears once you drive the car, hence the discrepancy between EPA constant and the actual rated miles. This is why the number you are measuring will go up with degradation

And yes, trust me, I really DO understand this as I have access to the CAN, like diamond.g explained. It will be fairly better and easier for you if you get an OBD II and observe the kWh directly. TeslaFi doesn't know a lot of things.

 

[AlanSubie4Life]

As noted above this concept applies equally well to any of the Teslas. The example here is from an X.

I just meant the constant is different. Obviously the concept applies to all Teslas.

There is a lot of complication being discussed here. But let’s keep it simple:

For AWD Model 3:

These numbers do not include the reserve.

Maximum available kWh for driving as displayed on trip meter:

230Wh/mi * (rated miles at 100%)

Maximum battery capacity:

245Wh/mi * (rated miles at 100%)

For my vehicle, I have 307 miles (Sometimes 306 miles) at 100%.

So my maximum trip meter usable energy is:

70.6kWh

My maximum battery charge capacity is:

75.2kWh

Again, not including reserve.

It’s very clear. No need for CAN bus access, though obviously that allows more visibility into the size of the reserve.

I’ll see whether I can post pictures. I still have them. It is all supportive.

I have not seen any evidence of any dependence of the discharge constant on discharge rate.

 

[AlanSubie4Life]

Segment 1 (did not capture the charging data after this segment, but I will do that for next segment - but I know I added 177 miles to get to 253rmi, or 43.4kWh):

Easy to verify discharge constant from this.

Start:

End:

 

[ajdelange]

Not sure why I have to explain this again, but you will do me and us a favor if you read the previous posts and comments...

You have to keep explaining it because your explanations don't make sense no matter how many times one reads them.

There actually is a lot to do with BMS and capacity and degradation, because the rated miles will not be the same, once your car gets older.

Perfect example. The rate at which the car uses energy has no dependence on the state of the battery. If it takes 300 Wh to go a mile today it will take 300 Wh to go a mile 5 years from now regardless of how much the battery has degraded.

It is 3.1m per% at the beginning, but it gradually drops with degradation and imbalance.

Yes, but we aren't talking miles per percent here are we? We are talking miles per kWh (or, actually the reciprocal).
Beyond that, if the rated miles did change over time the calculation would pick it up. Rated Wh/miles is found by dividing the Wh used or added by the rated miles used or added. There is just no way to make it any more complicated than that.

So the obvious thing to do is plot estimated rated Wh/mi against the odometer and look for a trend. I've only got about 6500 mi on the car but if I do that I do see a trend but Pearson's r is 0.06. When a statistician (I am not a statistician) sees r that low he doubts there is any trend at all and calculate the probability that he might see r that big or bigger were there no correlation at all. That probability is 47% thus we have insufficient support for rejecting the null hypothesis (that there is no trend) and conclude that there is none. The estimates are not changing over time even though the battery capacity is (increasing by 169 watts per thousand miles driven).

Your calculation is ok when the car is fairly new, but once the rated miles drop(the ones the BMS shows at 100% and calculates based on the EPA constant*kWh nominal) then your calculation will not be any more meaningful then what TeslaFi is trying to calculate. Not much meaningful at all.

Of course it will be meaningful. When my battery capacity has dropped to 90 kWh then the displays will show that I have 90000/333 = 2070.27 rated miles range at 100% rather than 98000/333 = 294.3 at 100% rather than it does now when the capacity is 98000 kWh as it is now. This is exactly what it is supposed to do. This is what common sense says it should do and what I want it to do.

Also, there might be heat loss and imbalance in the BMS which has to be accounted for and they will differ from drive to drive hence the one time measurement doesn't work(like you rightfully mentioned). Reading the BMS does work all the time

As I have carefully previously explained heat losses in the system have nothing to do with any of this beyond the fact that those losses are counted with all the other energy drains on the battery when battery current is measured. Battery imbalance is another question and also has nothing to do with this approach. Keep in mind that the readings sent to the API and to the displays come from the BMS.

But all you are really doing or trying to do is putting a Wh/m number to something we already know - that the buffer, which is being used at 100% disappears once you drive the car, hence the discrepancy between EPA constant and the actual rated miles. This is why the number you are measuring will go up with degradation

The buffer is always there. It does not dissappear when you drive the car. It is not visible to the user because it represents the range of SoC below the "empty' SoC but above the serious battery damage SoC.

I don't think you understand what "rated" means. The Wh consumed on a given trip depends on a whole lot of things external to the car such as speed driven, weather, road surface conditions... But it also depends on things internal to the car such as how efficient the motors and inverters are. So how do we compare a Tesla to a Kona? We try to standardize the externals to the extent we can and make them represent something meaningful. This is done by prescribing a driving speed profile that represents some mix of highway, town, freeway... etc. The manufacturer has to come up with a set of convincing (to the EPA) loads to put on the brake of a dynamometer which represent drag etc at the various speeds and the tests are run. One of the things that is, of course, standardized in these tests is that there be a fresh battery charged to some level that the manufacturer convinces the EPA is reasonable. Over the course of the test protocol the amount of energy withdrawn is measured and the number of miles "driven" is also and the two divided to give the average watt hours per mile under the EPA driving conditions. As the goal is to come up with some apples to apples comparison of the range of the vehicles this Wh/mi number is divided into a number which the manufacturer has convinced the EPA represents the capacity of the battery and the result is the EPA range which goes in the advertisements for the car. The rated Wh per mi number is the result used to determine the EPA mileage rating. It never changes. The range, expressed in rated miles does as illustrated above. The rated Wh per mile is a rating. To speak of an EPA rating and an actual rating does not make sense. To what authority does one go or to what set of standards does one refer to find the "actual" rated miles.

And yes, trust me, I really DO understand this as I have access to the CAN, like diamond.g explained.

You may have access to CAN bus data but it is pretty clear from the above that you do not know how to interpret it. I suppose that's why Tesla goes to some trouble to conceal it. My suggestion would be to step back a bit and try to understand what the forest is telling you before attempting to interpret the leaves on the trees.

It will be fairly better and easier for you if you get an OBD II and observe the kWh directly. TeslaFi doesn't know a lot of things.

I would love to have CAN bus data but I am not paying Agilent or Fluke for a packet sniffer nor will I patronize Google so until someone comes up with something for Apple I will rely on what TeslaFi and common sense tell me.

 

[TimothyHW3]

So my maximum trip meter usable energy is:

70.6kWh

My maximum battery charge capacity is:

75.2kWh

Again, not including reserve.

It’s very clear. No need for CAN bus access, though obviously that allows more visibility into the size of the reserve.

No, actually it is not very clear, because the nominal full you are quoting at 75.2(might very well be that to be honest) includes the buffer and the usable you quote is far from the real value since the most buffer on Model 3 is 3.5kWh. So your calculation is off by 2kWh.

And yes, if you had the CAN, you can actually see this for yourself

Was the 43.4kWh you posted the used quoted by the car as used under Trip or something else? A calculation, the charge going inside the car?