this post was submitted on 05 Jan 2024

124 points (94.3% liked)

RealTesla

487 readers

1 users here now

Posts must be about Tesla, EV, or AV
Meta Posts must be pre-approved.
Shitposts are limited
No Elon Worship
All Links must include the original title of the Content
Sites behind Paywalls must have text included.
Don't be an asshole
No Image Posts

founded 2 years ago

MODERATORS

dragontamer

[email protected]

124

Tesla lowers Model Y, S, and X range estimations following exaggeration complaints (www.theverge.com)

submitted 11 months ago by dragontamer to c/realtesla

29 comments fedilink hide all child comments

Not the Cybertruck story, but perhaps more important.

The Justice Department has been probing Tesla for their exaggerated range claims, and suddenly Tesla has decided to reduce their estimates on these already-released cars.

all 30 comments

sorted by: hot top controversial new old

[–] Blue_Morpho 29 points 11 months ago* (last edited 11 months ago) (4 children)

They need to update the actual software, not only the published estimates. My nephew was left stranded at 2am because the built in navigation which directs you to super charging stations based on charge remaining said he had 5 miles of charge to reach a station 1 mile away. He didn't plan the charging stop. The Tesla navigation software told him at the previous charge station where to drive to reach the next charge without running out.

The "5 miles range" turned into 0 miles over the next few seconds. He had to push his car the rest of the way to reach the supercharger station.

This despite Tesla's claim that you can drive a little longer even when you reach "0 miles" range.

[–] AA5B 6 points 11 months ago* (last edited 11 months ago) (1 children)

Im not sure I agree this is a concern. All my life I’ve been told that range is just an estimate and it’s especially variable toward the end. Granted that’s gas and analog measurements, but battery estimates also aren’t exact. I would never drive an ICE car down to an estimated 5 miles remaining, I know not to trust that estimate so much.

While an EV should be more accurate, are we sure this isn’t a case of more precision than accuracy? Maybe the only problem is the false confidence of being overly precise

[–] Blue_Morpho 9 points 11 months ago (1 children)

I would never drive an ICE car down to an estimated 5 miles remaining, I know not to trust that estimate so much.

You keep missing the point that it was Tesla navigation that told him he could reach the next nearest Supercharging station.

HE DIDN'T HAVE A CHOICE.

As to driving an ICE with 5 miles remaining, ICE cars have 1 gallon minimum in the tank when they say 0 miles left. My Sienna had 3 gallons. That's 15-45 miles under worst conditions. Tesla used to go 20+ miles at 0 range but that must have changed in their quest to advertise maximum range.

[–] [email protected] 0 points 11 months ago (1 children)

Sorry, not expert on Teslas, what you mean he didn't have a choice?

[–] Blue_Morpho 9 points 11 months ago (2 children)

You are driving across country based on Tesla saying it has Supercharging stations across the route so you can make a trip. The Tesla navigation says its fine. At each Supercharging stop along the way, it says everything is fine. At a Supercharging stop, the Telsa says you are charged fully and can easily reach the next Supercharging station. But after you start driving to the next station, the charge level drops rapidly until you are stranded.

He didn't leave his house with the Tesla saying he had a 5 mile range. He left the Supercharging station during a cross country trip with the Tesla saying it had a 260 mile range when that wasn't true. He didn't choose to leave with a low charge. The Telsa lied to him about its actual range.

[–] [email protected] 3 points 11 months ago

Ok I see, tbh as I said no idea of all this system.

[–] dragontamer 2 points 11 months ago* (last edited 11 months ago)

But after you start driving to the next station, the charge level drops rapidly until you are stranded

To add to this: above 50% charge, it has been proven that Tesla computers lie to you and purposefully overestimate your range.

The computers adjust to reflect closer to reality by the 50% mark. But above 50% Tesla computers are there to sell cars, not there to tell the truth or serve as a real estimate.

[–] dlpkl 4 points 11 months ago (3 children)

I didn't think you'd be able to push an EV. Do Tesla's have some sort of tow or neutral mode?

[–] [email protected] 4 points 11 months ago* (last edited 11 months ago) (1 children)

"Neutral" on an EV doesn't mean the same thing as on a IC car. The motors are still physically connected to the wheel, so turning the wheels will create an electric current.

That current will flow into the batteries from the "wrong end". Normally, charge controllers and battery management systems will ensure every cell of battery gets a balanced charge, but those systems don't work when you tow or push an EV. You can end up ruining the battery. Another issue is that power and friction cause heat, and when the car isn't running, that might cause overheating.

Now, the good news is that you're very unlikely to push an electric car fast enough or far enough for either of those to become a real issue. Electric cars are still big and heavy.

[–] dlpkl 2 points 11 months ago

Yeah that's what I thought. So not only are you fighting the usual weight and friction, but also resistance from the motors.

[–] Blue_Morpho 4 points 11 months ago

Dunno but he did it while his girlfriend steered. It is a model Y. He's a big guy who works out. He said it took everything in him to get it over a tiny hill at the exit ramp.

[–] [email protected] 2 points 11 months ago

This sounds like Microsoft’s update time estimations.

[+] [email protected] -9 points 11 months ago* (last edited 11 months ago) (4 children)

The car cannot take into consideration wind and pavement roughness so well as to indicate that perfectly, nor battery wear I think. I'm not saying Tesla good, I'm saying nephew stupid. People run out of fuel all the time for the same reason: trusting estimate without a bit of common sense added to it.

Fyi: I drive a PS2 and it's estimate system doesn't take into consideration current weather, but the Google Maps one in Android Auto does, so the latter gives me vastly closer estimates.

Edit: Jesus the amount of people who think you can precisely measure the amount of power left in a battery. Try to differentate what I was talking about from Tesla or any other manufacturer. Elon is a pissant, but trusting estimates to be accurate is not smart, especially in EVs.

[–] dragontamer 13 points 11 months ago* (last edited 11 months ago) (1 children)

Except Tesla literally lies to you with the range estimates. Internal documents show that the range estimates are false, and Tesla staff KNOWS its false.

https://www.motortrend.com/news/tesla-ev-lying-range-much-worse-elon-musk/

EDIT: The Cybertruck range estimates are proof of this lying culture. Cybertruck is NO WHERE close to the estimates.

[–] DoomBot5 4 points 11 months ago

Hell, if you look at it's estimate based on the last 30 miles of consumption, you can see just how vastly different it is from the actual range estimate displayed on either the charge level or the trip.

[–] [email protected] 8 points 11 months ago (2 children)

Wind and pavement won't make 5 miles into zero miles in a few seconds.

[–] AA5B 2 points 11 months ago* (last edited 11 months ago) (1 children)

I once read a description of battery metering where the best you can do is read current over time, and estimate where you are by the slope of the line. That can easily change drastically from one second to the next.

I still think the problem s an overly precise estimate, and the false confidence of someone thinking it is exact. Even digital technology is implemented in analog and some things just don’t have exact answers

[–] [email protected] 4 points 11 months ago (1 children)

In the case of Tesla there is ample evidence that they intentionally lie about range so it doesn't matter how hard it is to guess when we already know they are lying.

[–] AA5B 3 points 11 months ago (1 children)

Not denying that, but I’m also not relying on an estimate so close to zero

[–] [email protected] 2 points 11 months ago (1 children)

At least you have common sense. Seeing other replies is befuddling. Yeah Tesla lies constantly, but literally no EV has bulletproof estimates.. because they can't have that. And trusting an estimate as fact is just so weird to me.

[–] Blue_Morpho 1 points 11 months ago

And trusting an estimate as fact is just so weird to me.

That means Tesla's claim that supercharging stations allow you to make long trips is a lie. When you charge at a Tedla supercharging station, the Tesla tells you where to drive to the next charging station to not run out of charge. If range cannot be estimated such that you can't reach the next station they tell you to drive to, their Supercharging network claims are lies.

[–] BleatingZombie -5 points 11 months ago (1 children)

I completely agree. Also, yes they can. Predict the rate of change based on the amount used over the past hour or so and adjust. Gas cars do this already, so I'm sure electric cars do

[–] dragontamer 8 points 11 months ago* (last edited 11 months ago) (1 children)

Erm, that's not the issue. Gasoline cars have gasoline in the tank, meaning its very easy to estimate how much fuel you have left. In fact its not an estimate at all, its a direct measurement of gasoline in the tank.

The charged electrons inside a Li-ion battery? No good estimate methodology. I'm serious. You can't measure with a dipstick, there's no float-bob, you can't weigh, you can't do jack diddly shit with Li-ion batteries from any physical perspective.

What Li-ion estimations are is something called "coulomb counting". You count the electrons in, and count the electrons out. Guess what? It gets inaccurate when things like internal-leakage changes based off of the temperature changes over the past week and a half of charging. This is why phone Li-ion batteries can shutoff at "10%", because those electrons leaked out in a way that the colomb-counting sensor was unable to detect. (Many, many different sources of leakages).

Its like trying to measure the amount of gasoline in your tank by simply measuring the gasoline in and out of a gas-tank. But what about evaporation? What about a whole slew of other issues? You just can't plug all leaks in gasoline tanks or Li-ion. But you can just weigh the gas-tank to figure out how much gasoline is there, or you can have a float-bob that tells you how far the gasoline is from the bottom of the tank directly. But literally no such technology exists (or ever existed) for Li-ion.

Maybe in the future this could be invented. But this "last bit of charge phantomly disappeared" is almost innate to battery-tech, at least with today's level of technology.

In fact, measuring the "zero point" of any battery is famously one of the most difficult problems in battery tech. The only way that seems to reliably work is by discharging the battery all the way to 0% so you know for sure where it is. But otherwise? Good luck finding it (and you shouldn't discharge to 0% because it damages the battery anyway). So the columb-counting methodology gets less-and-less accurate over time as all of these leakages or other problems add up... and since the 0% point is rarely reached, there's almost no way to calibrate in practice.

[+] BleatingZombie -6 points 11 months ago (2 children)

So ... blame the consumer for their ignorance? Require a class before they can buy an electric car?

Errrmmm I don't know what your point is

[–] dragontamer 7 points 11 months ago* (last edited 11 months ago)

I'm saying that this problem is innate to any and all electric vehicles with today's level of technology. And Elon Musk is a fucking liar for trying to suggest that "battery fuel gauges" can reliably determine the "last ounce" of charge.

The honest thing to do for EV proponents is to make sure people know the tradeoffs, so that no one tries to go down to 0% charge unless they absolutely have to. Or maybe for us all to encourage battery-chargers for cars to discharge the battery fully on some occasions (to reliably figure out the zero-point calibration). I dunno what the best solution is, but liars spreading lies about how EVs work is bad for everyone.

The consumer is obviously being misled by Tesla (and Elon Musk) on this matter. And that's before the shitty range-estimates come into play (literally a conspiracy where Tesla employees have documents that prove that they know about how bad some of these range estimates are, but are working to prevent the public from knowing about it). Some of it is understandable given the difficulty of the problem of battery-estimation. But its always wrong for the company to take an official stance to lie to their customers.

[–] Blue_Morpho 0 points 11 months ago

Require a class before they can buy an electric car?

Tesla has temperature sensors. They could give estimated range at the start of a charge based on current conditions. It was only a few hours between supercharging stations. The temperature didn't change dramatically between 11pm and 2 am.

But they don't do that for marketing. People would be returning their new Tesla's bought in the winter when they started it for the first time and it said 150 mile range instead of 250.

[–] Blue_Morpho 7 points 11 months ago

You missed the part where it was Tesla's navigation software that charged his car and told him where to drive to not run out before the next charge.

If you can't reach the next nearest Supercharging station that the Tesla navigation says you can reach, everything Tesla says about range is a lie.

[–] [email protected] 1 points 11 months ago

Surely they have had other Teslas drive over the same road before the nephew. That data just wasn’t used to update the estimate. Tesla bad.

[–] DoomBot5 9 points 11 months ago

First, are “comfort and functionality improvements” made by Tesla that require more energy, and second is the implementation of revised EPA testing requirements that result in a “higher consumption and a slight decrease in overall range.”

Here is the real answer. They don't give a shit about the complaints, the EPA revised the way they calculate estimates, and this impacted Tesla.