Tesla has a range inflation problem | Hacker News Reader

Tesla has a range inflation problem

techcrunch.com

59 pointsrntn42 comments

It's very easy to see how favorably or unfavorably Tesla's claimed range compares to competitors based on independent tests of multiple EVs in the same conditions:

https://www.youtube.com/watch?v=6LWL90paufE

https://www.youtube.com/watch?v=ynCaTDR4rDQ

https://www.youtube.com/watch?v=eFB6hsYXDiA

https://www.youtube.com/watch?v=fvwOa7TCd1E

Spoiler alert: Tesla models fare about as well, if not better, than their EV cousins, hitting around 80% of the stated range in the wild.

We're not talking about how well they do compared to EPA range estimates; we're talking about the car choosing to put the EPA range estimate on the dash even when the car knows that number is bunk.

Most ICE cars (and other electric cars) use driving history averages to adjust the range estimate on the dash, and a touch of pessimism. They're still not perfect, but better than claiming EPA range for someone who you know never gets close to it.

Tesla owner, not saying they didn’t do this, but to consider…

The range in an EV is impacted by more variables than a gasoline vehicle. I’ve seen so many sensationalized headlines the last couple a weeks, in particular one about how the heat wave is killing electric cars.

Look, as with any big change, people have to learn how the new thing works and that takes time. I don’t get the same range on a 95 degree day as I do on a 75. The rain impacts it, charging my phones impacts it, running the AC impacts it. I believe the number provided in the vehicle is fairly accurate and considers a handful of factors. But others like your driving style also have a huge impact and there is no way to account for that.

Can the software get better and incorporate more variables in its estimate? Yes. Do consumers also just assume the number displayed is perfect? Also yes.

We had this same problem when GPS navigation became available. There were stories of people who followed the guidance before their common sense and ended up in strange places. Now we know to use it as a guide and check for common sense.

I find it funny that a side effect of ICE cars being so inefficient is that their range is much more predictable. If ~50% of your petrol turns into heat no matter what you do then using the AC doesn't make a lot of difference.

I don't think they have to do with inefficiency.

Predictability is there because combustion engines operate the same regardless of weather conditions.

AC / energy consumption anxiety doesn't exist because of said predictability as well as your ability to refill 1000km worth of fuel in a couple minutes.

The only difference if ICE were 100% efficient would be a slightly smaller gas tank.

I don't see how this makes any sense. An AC compressor is an extra load regardless of whether the motor is powered by gas or electricity. If it requires 10 Joules to run the AC for 1 minute, you need 20 more Joules worth of fuel. If turning on the AC didn't decrease gas mileage, then the efficiency of the engine somehow magically increased when it starts turning the compressor.

This same reasoning should also apply to charging your phone, cranking the bass, etc, because that represents an additional load on the alternator which must be powered by the engine.

Of course, turning on the heater in gas vs electric car is a different story, because the gas car is using the heat energy you mentioned that would otherwise be wasted.

Because the drivetrain is several times more efficient, other loads have an outsized impact on mileage. The Model 3 is rated by EPA to be 131 MPGe.

So yeah -- if your drivetrain is getting 131 MPG, your A/C that consumes the same amount of power is going to have a much larger proportional impact.

Come on, tesla makes self driving cars and now they have trouble estimating range?

If you tell the car where you're going the estimate is deadly accurate.

Favoring optimistic predictions isn't a reasonable response to large margins on complex probabilistic models. "Under-promise and over-deliver."

Choosing the side of the margin that favors your marketing at the risk of stranding the customer on the side of the road is the unscrupulous choice.

> The rain impacts it, charging my phones impacts it, running the AC impacts it.

That doesn't sound too good.

If this is the case, why is Tesla the outlier? Ford, Hyundai, GM etc. all seem to be able to accurately estimate range. In fact, as the article mentions, Hyundai is slightly inaccurate in the opposite direction (though not by the same margin as Tesla over-estimates).

If this is simply people getting used to a new technology, wouldn't it be a similar story across the board?

related with 600 comments "Tesla created secret team to suppress thousands of driving range complaints" https://news.ycombinator.com/item?id=36891642

Hunh. Tesla lies to it's customers. Who'd of thunk it?

Reuters' Steve Stecklow writing Tesla clickbait again in bad faith, range can change 10-30% based on how fast you go and weather: https://www.tesla.com/support/range

This is blogspam, the actual article is here:

https://www.reuters.com/investigates/special-report/tesla-ba...

This is a more complicated problem than it seems because even determining the remaining battery percentage is non-trivial with lithium-ion batteries. Apple has an entire team dedicated to this.

Predicting the remaining range is another step of complexity. I'm not saying it can't be done, but they should definitely sandbag the estimate.

Some of the features using energy when parked, like sentry mode and cabin temperature maintenance, probably consume power and add to customer frustration as well.

I have a work friend that constantly complains that his fuel economy isn't what he expects in his hybrid toyota. It has taken me long time to get through to him that fuel economy varies based on activities and environment... (cold weather/running the heater/mountain driving/towing items etc)

I don't know why but he seems to be outraged that something as simple as running the heater in the car in winter will alter the fuel economy of the car.

Really depends, if I blitz along at 80, then yeah. If I stick to the speed limit, it’s not a bad estimate, on mostly slow local roads, we get substantially better. It’s a great demonstration of drag’s nonlinear nature (the cube of the speed?).

> The National Maximum Speed Limit (NMSL) was a provision of the federal government of the United States 1974 Emergency Highway Energy Conservation Act that effectively prohibited speed limits higher than 55 miles per hour (89 km/h). It was drafted in response to oil price spikes and supply disruptions during the 1973 oil crisis and remained the law until 1995.

I've always been under the impression 55 was the optimum speed for (ICE) fuel efficiency.

Does this change at all for electrics (given regenerative braking)?

The best MPG you can get on the highway is just the slowest you can safely go without confusing other drivers, making people angry, or getting pulled over by the police. Drag is proportional to velocity squared, so every increase in mi/hr wastes more energy pushing air out of the way of the car.

To answer your regenerative braking question: braking is inefficient by default, so just avoid tapping the brakes in situations where you could coast.

With regenerative braking, you'll never get back 100% the energy because of thermodynamics. There's always wasted energy any time energy is moved, transformed, stored, etc.

Go too slow though and mileage goes down. At 60 mph my car gets around 25 mpg. At 1 mph it would get around 3 mpg.

That's because my car never consumes less than around 0.3-0.4 gallons per hour when the engine is running.

Modern cars with variable valve timing and other goodies can achieve very good combustion under low load. Looking around, it seems ~30km/h would be the sweet spot.

It would be slightly higher if you run the AC on high, increasing total consumption by ~10% and moving the dial a bit.

No it's a tradeoff between speed and efficiency. If you went 45mph, you'd get even better gas mileage. But people already couldn't drive 55.

Active discussion today here: https://news.ycombinator.com/item?id=36891642

How good are range estimates on ICE cars and other electric cars. How do other electric cars handle cold weather estimates?

Range estimates are shit on ICE cars too. They're called "estimates" for a reason. You can't predict the future. I'm not a Tesla fan but this whole thing is nonsense.

Has or had? The Reuters article opens with: "About a decade ago.."

All EVs have this problem.

That said - drive conservatively and you can meet or exceed the predicted range.

(experience owning several EVs, driving others)

They also lied about C02 emissions. My country manufactures 30% of electricity from coal. That is hardly zero emissions!

Even if it was 100% produced from coal, an EV still pollutes less because emissions are easier to reduce at a plant than in a vehicle.

And petrol engines are very inefficient at 35% efficiency.

Coal Plants are even less

It's “CO2” (or “CO₂”, to get fancy), not “C02”. (Oh, not zero.)

I've always been concerned about the accuracy of the degradation of the range/the range in general.

There is no real concrete way to tell if they are lying because the source raw data is obfuscated.

Always has. Nothing new under the sun https://lemmy.world/post/1319083

i didn’t realize buying a tesla was a moral failing. i guess i’ll but from an ethical company like GM instead

> The Korea Fair Trade Commission also found that Tesla failed to inform owners that cold weather can reduce its cars’ range. The commission cited tests showing Tesla cars lost up to 50.5% of the company’s claimed ranges in cold weather.

Ouch. I've had trouble with cameras not working for long in the bitter Korean cold.

I don't understand how this became a problem, much less one that was discoverable.

If they're still selling extended-range batteries as DLC, just give the customer the extra capacity for free but spread it out over the service life of the battery (doesn't Apple do this to keep battery capacity constant?). They won't be buying the upgrade for a 10 year old battery anyway, so might as well use the extra material for something and avoid this sort of PR.

Or dynamically give/take away exactly enough of the unpaid-for reserve capacity to be able to consistently deliver whatever the promised range was. Unlock the remaining extra capacity as the upgrade. By the time people start noticing the extra range isn't working as advertised, you can gaslight everyone over inconsistent usage patterns across aging cars.

Big news, automobile manufacturer inflates gas mileage! Really!

If anyone here has had a car that actually got the stated mileage in real world conditions, please let me know.

I live on top of a hill and every single car I've ever owned, including, Honda, Toyota, Ford, Nissan, Dodge, Volvo that I recall right now has gotten oh, about 30% lower mileage than advertised. due to that pesky law of gravity that requires energy to be expended going up and down.

But even when I lived in leveller locales, cars never got the EPA rated mileage, except once, when I drove a Honda with a CVCC engine on a road trip through level countryside at about 45 miles per hour all day. I got terrific mileage. I never did it again.

Your comment is just publicly broadcasting your ineptitude at efficient driving. I regularly beat EPA estimates in both the EV I have and the ICE car I had.

You can read more about EPA mileage testing: https://nepis.epa.gov/Exe/ZyPURL.cgi?Dockey=P100IENB.TXT I've personally found it to be fair and reflects the real world quite well.

> If anyone here has had a car that actually got the stated mileage in real world conditions, please let me know.

I owned a 2016 Subaru BRZ, stickered at 24 mpg city/30 mpg highway, and over the 3 years I owned it, I averaged 32 mpg.

My previous car was a 2000 Suzuki Esteem, stickered at I think 27/32. I averaged 28 city, 35 highway.

I now have a Model 3 Performance, estimated 300 mile range. I get 230-270 miles depending on the weather.

> If anyone here has had a car that actually got the stated mileage in real world conditions, please let me know.

Depends on driving style. I drive a manual Ford Fiesta (with worse specced fuel efficiency than the automatic) living in the foothills of the Rocky Mountains, and it's abnormal for me not to get better fuel economy than the automatic's spec.

Reply on news.ycombinator.com