What Tesla should have done with range.

[brur]

It has become a recurring theme on the Tesla forums, increased range, losing range, range worry, and it all is a waste of time and energy.
The car screen should have an icon showing the amount of energy in the "tank". As in half a tank or a third of a tank, just like an ICE car would have had twenty years ago. The ICE driver would have to think about how far they could go based on geography and speed and history of previous drives.

The problem with the current screen seems to suggest that there is a real science-based method of determining mileage your car will get. It isn't science, it's a guess based on assumptions. There is no 100% way to know the range the batteries can produce other than by running them down on charge.
As for software updates giving you more range... I don't know, maybe it's a recalibration of the range indicator, maybe it's a tiny bit more energy taken from the battery by way of discharging or charging. There is no magic button or mushroom, they might find a bit here or there in the HVAC, throttle application or whatnot but in the overall scale of day to day life, it's nothing.
If you need to go further on a charge drive slower. Think of an hour clock (sand running through an orifice) every time you punch the accelerator you make that orifice larger, every time you coast or accept the notion of being passed you make the hole smaller.

btw watch youtube sensation Jack Rickard, if you want real entertaining lessons on battery systems.

 

[Big Earl]

It has become a recurring theme on the Tesla forums, increased range, losing range, range worry, and it all is a waste of time and energy.
The car screen should have an icon showing the amount of energy in the "tank". As in half a tank or a third of a tank, just like an ICE car would have had twenty years ago. The ICE driver would have to think about how far they could go based on geography and speed and history of previous drives.

The problem with the current screen seems to suggest that there is a real science-based method of determining mileage your car will get. It isn't science, it's a guess based on assumptions. There is no 100% way to know the range the batteries can produce other than by running them down on charge.
As for software updates giving you more range... I don't know, maybe it's a recalibration of the range indicator, maybe it's a tiny bit more energy taken from the battery by way of discharging or charging. There is no magic button or mushroom, they might find a bit here or there in the HVAC, throttle application or whatnot but in the overall scale of day to day life, it's nothing.
If you need to go further on a charge drive slower. Think of an hour clock (sand running through an orifice) every time you punch the accelerator you make that orifice larger, every time you coast or accept the notion of being passed you make the hole smaller.

btw watch youtube sensation Jack Rickard, if you want real entertaining lessons on battery systems.

You can change the battery display from rated miles to % remaining in the center screen. I prefer % for exactly the reasons you describe.

 

[Chisale]

I've tried it both ways and I prefer the mileage display. You just have to think of those numbers as nominal mileage and you have to add 20% or 30% to whatever actual miles you are driving depending on the circumstance. The energy graph shows in percent for exactly the same reason you are talking about. Still, even if every Tesla driver switched to the "percent left" choice, I really doubt that would stop all the bitching about range.

 

[dhrivnak]

On trips I use the range estimates a LOT. They are very helpful in guiding me to see if I need to change my driving in any way. And fortunately I have found them to be quite accurate.

 

[ajdelange]

The car screen should have an icon showing the amount of energy in the "tank". As in half a tank or a third of a tank, just like an ICE car would have had twenty years ago.

It does.

As for software updates giving you more range... I don't know, maybe it's a recalibration of the range indicator, maybe it's a tiny bit more energy taken from the battery by way of discharging or charging. There is no magic button or mushroom, they might find a bit here or there in the HVAC, throttle application or whatnot but in the overall scale of day to day life, it's nothing.

There are a couple of ways to increase range. One, and this is, I think, what you have in mind, is to simply move the E and F marks on the gas gauge farther apart i.e. to make some of the low end and high end buffers available to the driver. This, of course, reduces the safety factor that those buffers provide.

The other is to change the motor control algorithms to enhance the efficiency of the inverters and motor(s). There isn't much to be squeezed out here very probably as these systems are incredibly efficient already,

If you need to go further on a charge drive slower. Think of an hour clock (sand running through an orifice) every time you punch the accelerator you make that orifice larger, every time you coast or accept the notion of being passed you make the hole smaller.

The sand runs out faster but the car goes further so this is not the way to explain why going slower increases range. Drag from air resistance is proportional to the square of the vehicle velocity as does the energy per unit distance required to overcome drag. Thus reducing speed from 60 to 40 mph decreases the component of the Wh/mi ascribable to drag to 44% of what it is at the higher speed. While drag isn't the only Wh/mi requirement the 56% reduction in it is enough to make an appreciable difference in the overall.

The motors are designed to be most efficient at moderate speed. Running them at higher speed implies more Wh/mi lost to heat.

 

[brur]

On trips I use the range estimates a LOT. They are very helpful in guiding me to see if I need to change my driving in any way. And fortunately I have found them to be quite accurate.

Yes, I use the navigation miles too. Pretty helpful unless you are going up a mountain.

My real point, if I could communicate with any skill, was the continual reference to software updates that claimed to add X% more range. There is always the, I didn't get it, it's not for your model, what software is it on....
As I see it Musk is working a marketing game with these claims. Sooner or later the mileage indicator is going to disappoint the owner.

 

[MichaelP90DL]

Yes, I use the navigation miles too. Pretty helpful unless you are going up a mountain.

My real point, if I could communicate with any skill, was the continual reference to software updates that claimed to add X% more range. There is always the, I didn't get it, it's not for your model, what software is it on....
As I see it Musk is working a marketing game with these claims. Sooner or later the mileage indicator is going to disappoint the owner.

Trust me, you communicate very well.

 

[dhrivnak]

Yes, I use the navigation miles too. Pretty helpful unless you are going up a mountain.

My real point, if I could communicate with any skill, was the continual reference to software updates that claimed to add X% more range. There is always the, I didn't get it, it's not for your model, what software is it on....
As I see it Musk is working a marketing game with these claims. Sooner or later the mileage indicator is going to disappoint the owner.

That I agree with. I have seen my 90% range change from 280, to 285, to 287 to 284 as various software updates hit and I am still not convinced I can go farther, or less based on the latest update.

 

[HankLloydRight]

I've tried it both ways and I prefer the mileage display. You just have to think of those numbers as nominal mileage and you have to add 20% or 30% to whatever actual miles you are driving depending on the circumstance.

This is exactly what I do.

 

[ajdelange]

Let's suppose your 90% range is 285. That implies 100% range of 317 miles which is three miles per percent. Thus just because of rounding you can have 3 miles displayed variation.

 

[ajdelange]

Miles and percent convey the same information. The car is continually estimating the battery's SoC. The displayed range is simply the SoC (in percent) multiplied by the miles per percent. The problem with the miles display is that there are several mile per percent value and it is easy to be confused about which one is being displayed. To add to the confusion Tela changes the those ranges from time to time. I was purring down the freeway a month or two ago and the range was jumping from 100 to 500 miles in front of my eyes. WTF?? Then I notice that there is a new option "Instantaneous" which was never there before. So that is (was - don't know if it is still available) one type. Then there are values based on the last 5, 10 or 30 miles driven and finally the cars "rated" number which they can change in firmware if they want to.

Thus I want to see percent. It tells me how much battery I have. The distance I can go on one percent of battery can go from infinity (the maximum the car displays is 999) to a fraction of a mile. It is incumbent on me to know enough about my route and driving conditions to have a pretty good idea as to how many miles I will actually get per percent and act accordingly. Of course the energy display graph is there to help me do that and, of course, experience driving the car and familiarity with the route are a big help too.

It goes without saying that the individual driver should use the system he prefers.

 

[ralph142]

It is science. But the governing equations for the estimates have a huge pile of variables, as they must. Surprise!

 

[ajdelange]

Range = SoC*miles_per_unit_SoC.

I wouldn't glorify something that simple with the label "science". It's grammar school math. Or at least the second factor is. You divide the miles you went by the amount of SoC you used. Now estimating SoC is based on one important scientific fact - that the batteries are extremely coulomb efficient. Thus it is pretty easy to get a good estimate of SoC by counting the coulombs that went into and came out of the battery. The problem is, as in any case where the observable is an integral, that errors integrate up too if they are not dealt with properly. In a battery we can observe two state variables: temperature and voltage. Using voltage the SoC can be recalibrated from time to time. Thus it is apparent that the challenge here really in coming up with the art that allows good SoC estimates or ones that are adequate for the purpose at hand. Note that SoC estimates are only given to 1% precision which says that in absolute terms the SoC estimates are not that good. But they are much, much better than the estimates of miles_per_unit_SoC to be encountered in the next x miles to be driven. This is another argument for working in SoC rather than range as range contains the uncertainties in both factors (and the uncertainty in miles_per_unit_SoC contains the uncertainty in SoC used to determine it) . But, if range goes to 0 or SoC goes to 0 you are in trouble and so you can really use either.

 

[ralph142]

Range = SoC*miles_per_unit_SoC.

I wouldn't glorify something that simple with the label "science". It's grammar school math. Or at least the second factor is. You divide the miles you went by the amount of SoC you used. Now estimating SoC is based on one important scientific fact - that the batteries are extremely coulomb efficient. Thus it is pretty easy to get a good estimate of SoC by counting the coulombs that went into and came out of the battery. The problem is, as in any case where the observable is an integral, that errors integrate up too if they are not dealt with properly. In a battery we can observe two state variables: temperature and voltage. Using voltage the SoC can be recalibrated from time to time. Thus it is apparent that the challenge here really in coming up with the art that allows good SoC estimates or ones that are adequate for the purpose at hand. Note that SoC estimates are only given to 1% precision which says that in absolute terms the SoC estimates are not that good. But they are much, much better than the estimates of miles_per_unit_SoC to be encountered in the next x miles to be driven. This is another argument for working in SoC rather than range as range contains the uncertainties in both factors (and the uncertainty in miles_per_unit_SoC contains the uncertainty in SoC used to determine it) . But, if range goes to 0 or SoC goes to 0 you are in trouble and so you can really use either.

^^ lol, its not science in the active research sense surely, but its based on a lot of science, unless of course you're just making that all up.

 

[DirtyT3sla]

I've tried it both ways and I prefer the mileage display. You just have to think of those numbers as nominal mileage and you have to add 20% or 30% to whatever actual miles you are driving depending on the circumstance. The energy graph shows in percent for exactly the same reason you are talking about. Still, even if every Tesla driver switched to the "percent left" choice, I really doubt that would stop all the bitching about range.

I think it would stop a lot of the repetitive posts we see here and on Reddit. The problem is, 1 mile is a defined measurement. So if your car says you can go 100 miles, but you can only go 80, people are like wtf??? A percentage is not defined...so if today I use 15% battery to get to work, and tomorrow I use 20% battery to get to work, you'd hopefully realize "Ah, today was colder and I drove faster! That makes sense."

Just like on your phone. If my phone dies in 6 hours instead of 10, I think "well, I was using it a lot today!!"

 

[Dithermaster]

Still, even if every Tesla driver switched to the "percent left" choice, I really doubt that would stop all the bitching about range.

Well, it might. My P85+ had been on percent mode when the whole "my range went down!" debacle happened. Many people saw their indicated mileage drop overnight with a software update. I have no idea if it affected my car because the % stayed the same (of course). I did switch it over to miles for a while to see what it indicated. It seemed fine but since I didn't have the daily viewing, I don't know if it changed or by how much. Even with the new car I use percent. It makes more sense to me than "fake miles" that I have to do math on based on consumption to figure out "real miles". I'd rather just do that with percent.

 

[brur]

I was using miles, and I saw what appeared to be heaps of mileage disappearing. Of course, it works on your mind. Even when you think you understand it's just a figure based on XYZ. I've gone zen or zoned out on the matter now. My psychiatrist thinks I'm making good progress.
I have Elon to thank for my present state of acceptance.