Welcome to Tesla Motors Club
Discuss Tesla's Model S, Model 3, Model X, Model Y, Cybertruck, Roadster and More.
Register

What's your 90%?

This site may earn commission on affiliate links.
No it doesn't! Do the math. They are using nominal capacity (or the adjusted number they start using under 80%) and the trip meter is buying it hook, line, and sinker.

Here, I'll do the math. Usable is going to be 190 at that level, nominal is 205.

The. Car. Is. Lying.
This is the math: Usable % remaining=.747 x 274 rated miles = 205 miles
Your numbers are there for everyone to see.
 
  • Disagree
Reactions: supratachophobia
No I don't. BMS reading from New Tesla divided by window sticker range. Plus decompiling the firmware. Burden of proof isn't on us.
The only thing Tesla specs is the range in miles. Not Wh/mi. My understanding is that they get that range by driving the car until it stops.
I do agree that something seems curious about the BMS numbers, and that the energy graph suggests rated miles at 290 Wh/mi, which is definitely wrong. But Tesla would have to explain that before I would accuse them of any intentional deception.
 
Last edited:
  • Helpful
Reactions: supratachophobia
The only thing Tesla specs is the range in miles. Not Wh/mi. My understanding is that they get that range by driving the car until it stops.
I do agree that something seems curious about the BMS numbers, and that the energy graph suggests rated miles at 290 Wh/mi, which is definitely wrong. But Tesla would have to explain that before I would accuse them of any intentional deception.

They can't just make up the EPA number. It has some basis in data beyond "driving until it dies".

0% usable in the BMS is zero percent on the dash. You can't drive past zero, according to Elon.

All the cars have different EPA consumption numbers, and the 90kwh batteries are 290. But beyond that, think what the car is doing intentionally: It's calculating range based on 1 number, nominal remaining, from 80-100%, another floating number from 20-80%, and then the usable remaining number below 20%. I would actually agree with you if the car just picked a single number to calculate and stuck with it.
 
They can't just make up the EPA number. It has some basis in data beyond "driving until it dies".

0% usable in the BMS is zero percent on the dash. You can't drive past zero, according to Elon.

All the cars have different EPA consumption numbers, and the 90kwh batteries are 290. But beyond that, think what the car is doing intentionally: It's calculating range based on 1 number, nominal remaining, from 80-100%, another floating number from 20-80%, and then the usable remaining number below 20%. I would actually agree with you if the car just picked a single number to calculate and stuck with it.
I don't know if they use a different number. From what I have seen they just use usable all the way. Remember, there are rounding errors involved so it's hard to know exactly what numbers they are using. If what you say is happening, show us the readout for 80% and 60%, or something similar to demonstrate they use different numbers.
 
It's violating the laws of physics for the REAL capacity to not have changed in 2 years. I'm afraid to get my car under 10% now. It's never a good time to find out what the real capacity is.

I have an 85D with 100,000 miles. The 90 % has decreased to 230 from 244 when new. The capacity prediction winding down to 0 seems to be spot on. I have made it to a supercharger with less than 5 miles twice in the past week. Several months ago I pulled in with 1 mile left. I drive with roof top racks usually with a surfboard on top and just don't seem to have any range issues. The software combined with my many miles of experience has really made range a non-issue

Based on my over 3 years and 100K experience, I trust this car more than any car I have ever owned...no issues whatsoever. Never took it in for suggested maintenance. Drives like brand new. tires are lasting 50K and I probably could have gotten more miles out of them.
 
I have an 85D with 100,000 miles. The 90 % has decreased to 230 from 244 when new. The capacity prediction winding down to 0 seems to be spot on. I have made it to a supercharger with less than 5 miles twice in the past week. Several months ago I pulled in with 1 mile left. I drive with roof top racks usually with a surfboard on top and just don't seem to have any range issues. The software combined with my many miles of experience has really made range a non-issue

Based on my over 3 years and 100K experience, I trust this car more than any car I have ever owned...no issues whatsoever. Never took it in for suggested maintenance. Drives like brand new. tires are lasting 50K and I probably could have gotten more miles out of them.

Must be the Michelins... ?
 
  • Like
Reactions: Kacey Green
yes. It came with the 19 inch Michelin msm4s and I replaced with the same at around 50k. I just replaced these with the Goodyear eagle touring that are touted to have low road noise and I do like them. We'll see how they last.
Service center will confirm, when pressed, that Michelin's last longer than the Goodyear's by at least 20%.
 
  • Like
Reactions: Kacey Green
It's violating the laws of physics for the REAL capacity to not have changed in 2 years. I'm afraid to get my car under 10% now. It's never a good time to find out what the real capacity is.

I'm out at the end of the bell curve, but I'm not alone. There are some other people reporting very low degradation.

So, if you have a 90kwh battery and access to the CANbus, you'll see that the value Range Remaining is actually calculated based on the BMS reported "Nominal Capacity" (includes unusable brick protection). However, once you start driving and you get below 80%, this Range Remaining begins a correction using some number between nominal and usable, down to 20% where it ends up using what it should; just Usable Capacity.

Nominal includes the 4kwh brick protection you can't use. And 0% = 0% Usable (Nominal will read 4kwh remaining when battery reports 0% to the driver).

I've been meaning to get a CANbus reader and see what was going on in there, but I haven't gotten around to it. I have too many projects sitting around waiting for attention as it is.
 
Pretty Graph Time!

This should be fairly self-explanatory, but I'll try to explain anyway. For the 90kwh batteries, this is what is going on behind the scenes: The car starts out reporting to the driver that it can travel XXX miles on it's current charge of 100%. It gets this number by assuming that it can use the Nominal Pack capacity to do XXX range. But we all know that the number used for Nominal includes the 4kwh brick protection and can't be used for driving. So the car has another number, Usable Capacity, which has that 4kwh subtracted. This is supported by the fact that 0% on your dash corresponds with Usable Capacity in the BMS also being 0%.

At some point below 90% and ~80% (closer to 80), something magical happens. The car stops using Nominal Capacity to calculate range remaining, and instead uses a number between Nominal and Usable. That is, until it gets down to 20%, by which time, it's only using Usable. Notice how the the distance between the two lines gradually, and linearly, gets closer together as SOC drops.

upload_2018-7-18_16-32-45.png



It might be tough to see since I shrunk the data-set down for ease of use, but refer to the rate in change on the small graph; the lines are flat between 100 and ~80, and also between 20 and 0.
upload_2018-7-18_16-41-48.png


I'm going to continue to make bold statements and say that Tesla knows about this huge "What's your 90%?" thread and that is why their magic only starts to happen below 90%. They want their best foot forward so-to-speak, so as not to raise concerns from 90kwh battery owners that their range has degraded so severely.
 
Pretty Graph Time!

This should be fairly self-explanatory, but I'll try to explain anyway. For the 90kwh batteries, this is what is going on behind the scenes: The car starts out reporting to the driver that it can travel XXX miles on it's current charge of 100%. It gets this number by assuming that it can use the Nominal Pack capacity to do XXX range. But we all know that the number used for Nominal includes the 4kwh brick protection and can't be used for driving. So the car has another number, Usable Capacity, which has that 4kwh subtracted. This is supported by the fact that 0% on your dash corresponds with Usable Capacity in the BMS also being 0%.

At some point below 90% and ~80% (closer to 80), something magical happens. The car stops using Nominal Capacity to calculate range remaining, and instead uses a number between Nominal and Usable. That is, until it gets down to 20%, by which time, it's only using Usable. Notice how the the distance between the two lines gradually, and linearly, gets closer together as SOC drops.

View attachment 317809


It might be tough to see since I shrunk the data-set down for ease of use, but refer to the rate in change on the small graph; the lines are flat between 100 and ~80, and also between 20 and 0.
View attachment 317813

I'm going to continue to make bold statements and say that Tesla knows about this huge "What's your 90%?" thread and that is why their magic only starts to happen below 90%. They want their best foot forward so-to-speak, so as not to raise concerns from 90kwh battery owners that their range has degraded so severely.
I don't know why, but your graph is completely distorted because the x-axis is not linear, not even close. The first two points are both at 100%, then it goes down to about 98, 95, and then a big jump to 80! Of course it will look strange when plotted like that. You should plot it on a linear scale, and you should include the actual numbers for the data points in order to make sense of it.
 
  • Helpful
Reactions: supratachophobia
I don't know why, but your graph is completely distorted because the x-axis is not linear, not even close. The first two points are both at 100%, then it goes down to about 98, 95, and then a big jump to 80! Of course it will look strange when plotted like that. You should plot it on a linear scale, and you should include the actual numbers for the data points in order to make sense of it.

I'll smooth it out with more data points. But the general shape will not change. Notice how nicely the two lines come together. This graph was made using snapshots for two different legs that had different consumption. I did it that way to show that regardless of consumption, the algorithm doesn't change. Some were trying to argue the consumption was what was throwing the discrepancy. For the next certain, I'll use this last 100% to 1% complete dataset for version 2.
 
It's not the quantity of data points, it's that the horizontal is not to any scale. It starts with 2 points at 100, creating a horizontal line segment. Those 2 points should be on top of each other. Then it drops by different percentages while moving horizontally an arbitrary amount. The horizontal distances should scale relative to the %SOC change.
Something like this: (I tried to guess your mileage numbers from your graph)

upload_2018-7-19_14-58-0.png
 

Attachments

  • upload_2018-7-19_14-50-1.png
    upload_2018-7-19_14-50-1.png
    6.7 KB · Views: 16
It's not the quantity of data points, it's that the horizontal is not to any scale. It starts with 2 points at 100, creating a horizontal line segment. Those 2 points should be on top of each other. Then it drops by different percentages while moving horizontally an arbitrary amount. The horizontal distances should scale relative to the %SOC change.
Something like this: (I tried to guess your mileage numbers from your graph)

View attachment 318083
Leave it to Excel to screw things up. Here are the datapoints I used:
upload_2018-7-19_15-47-44.png
 
  • Informative
Reactions: ran349