The DOJ Tesla probe has expanded to include EV driving ranges

[karlos22]

The advertised range is of course rubbish, just like it is with gas vehicles.
I have never paid any attention to it.

 

[mongo]

Manufacturers submit the original data, EPA tests 20% of new cars every year.
Regarding Kia/ Hyundai values:

8. What happened with Hyundai and Kia?
Each year, EPA tests a subset of the new vehicle models at our National Vehicle and Fuel Emissions Laboratory (NVFEL) in Ann Arbor, Michigan to verify that the fuel economy estimates provided by auto manufacturers are accurate.
In 2011 and 2012, EPA began performing an audit program of manufacturers’ coastdown tests. Coastdown testing is used to develop the dynamometer inputs for each vehicle model, so that the laboratory tests accurately replicate its tire rolling resistance, friction due to bearings and brakes, and aerodynamics. EPA audit tested multiple vehicle models, including the model year 2012 Hyundai Elantra. Discrepancies between EPA coastdown test results and information provided to EPA by Hyundai resulted in an ongoing investigation into the data for other Hyundai
and Kia vehicles.
Hyundai Motor America and Kia Motors America lowered their fuel economy (mpg) estimates for the majority of their model year 2012 and 2013 models to be consistent with EPA test results. The mileage for most vehicles was reduced by one to two mpg. The largest adjustment was six mpg highway for the Kia Soul.

 

[alexgr]

This reminds me of the Kia mpg suit, where the owners sued them after EPA caught them for faulty testing and misrepresenting their MPG on the sticker. I can't remember the outcome. Guess I should go look around and see. Only remember as a co-worker had a Soul and she was all giddy for what they were to receive in the settlement.

This case is different from Kia and others as the MPG posted on Tesla sticker is the same as on the EPA site.

 

[alexgr]

Yes, so my average range is 222mi...even though Ive driven the car like an absolute madman at times, in the winter, etc. Still averaged all out to 222.

Interesting, I wonder would you get 420 range if driving completely relaxed?

 

[mongo]

This case is different from Kia and others as the MPG posted on Tesla sticker is the same as on the EPA site.

So was Kia's... The data comes from the manufacturer unless the EPA tests it themselves later and forces a change.

 

[alexgr]

Isn't 74kwh/333wh/mi = 222 mi?

333 Wh/mi is not very good. For both 19M3LR and 22MYLR I have the same efficiency of 278 Wh/mi, so Tesla seems to be 20% more efficient than Kia? But it is only 3% more efficient (still more efficient!) than Kia if we take 260 miles range as the most efficient average drive in eco mode.

 

[gt2690b]

333 Wh/mi is not very good. For both 19M3LR and 22MYLR I have the same efficiency of 278 Wh/mi, so Tesla seems to be 20% more efficient than Kia? But it is only 3% more efficient (still more efficient!) than Kia if we take 260 miles range as the most efficient average drive in eco mode.

even thats pretty terrible for a 19M3LR

 

[opasenBG]

A screenshot (taken on 10/31/2023) from my 2022 Model3 Long Range OEM tires (purchased Dec.2022) w/ aero wheel covers removed on day 1.

If usable capacity on the battery is 75kWh* that puts it at ~302mi range or ~85% of the advertised 358mi* range. With aero covers off it takes about 3%* hit so technically I should be around 88% had I left them on.

I'm not taking sides here so don't "yell" at me. Just sharing my real world numbers.

* All of the above specs are based on top-page results from google search so can't comment on how accurate they are.

 

[sleepydoc]

It’s not a “problem”. Interstates are for traveling long distances at a relatively high rate of speed. It’s normal for people to drive at that rate of speed.

I did not buy an EV to drive a slower route than I would with an ICE when the interstate is readily accessible.

As I’ve said multiple times, my issue is not with higher energy consumption at higher highway speeds. My issue is with competitors being able to more accurately predict and hit/exceed the stated EPA driving range under the speed conditions of typical Americans versus Tesla.

If anything, I’d expect Tesla to be BETTER, because they’ve been at this EV game much longer.

I’ve owned Toyota, Mazda, Hyundai, Volkswagen, Subaru and Audi ICE vehicles in my 27 years of driving. They’ve all met or exceeded their EPA estimates on the highway when driving at around 70 mph. The VW Jetta Sportwagen TDI I had was especially good at hitting 43mpg+ at those speeds versus its 39mpg rating.

Our Model Y and 3 are the only vehicles to come up short by 20 to 25 percent.

It’s a problem if you want optimal economy and insist on driving 70+ mph. You can’t beat physics - wind resistance drops precipitously below 60 mph. I’ve seen clear degradation in mileage above 60 MPH in all the vehicles I’ve owned, no matter what the drive train.

Don’t forget that VW committed (and was convicted of) outright fraud with their TDI engines. The only way they achieved their mileage was by disabling the pollution control systems. (I had a TDI wagon for 8 years and profited nicely from the settlement.)

 

[sleepydoc]

A lot of this reminds me of computer processors. First there was clock speed. Then they started making benchmarks. Then the processor companies started specifically designing their processors so they were optimized for the benchmarks. Were they committing fraud? no, the benchmarks were accurate. They just weren’t necessarily representative of how people would use the processors.

From what I’ve seen, Tesla isn’t committing fraud, they are using the benchmark protocol that yields the best numbers. Assuming that this is the case, the issue is not with Tesla, but rather with the protocols that the EPA publishes, and requires the manufacturers to use if someone has actual evidence that Tesla is not being truthful and/or is committing fraud, please post it. Otherwise it seems the real issue is that the way people drive does not match the protocols that the EPA uses. THAT’S NOT TESLA’S FAULT, THAT’S THE EPA’S FAULT.

 

[Big Earl]

More apples to oranges testing:
"In short, this is the approximate number of miles that a vehicle can travel in combined city and highway driving (using a mix of 55% highway and 45% city driving) before needing to be recharged, according to the EPA's testing methodology."

"Edmunds begins with full battery charge and drives an electric vehicle on a mix of city and highway roads (approximately 60% city, 40% highway) until the battery is almost entirely empty. (We target 10 miles of remaining range for safety.) The miles traveled and the indicated remaining range are added together for the Edmunds total tested range figure. "

The big sore thumb here is where they're getting their remaining range figures from. It's well-known that Tesla's "range" display uses rated range - a number derived from the EPA rating and the remaining usable kWh in the battery. If the car is rated at 250 Wh per mile, every 4 miles of displayed "range" is 1 kWh of energy. The calculation on this display is fixed and does not change based on recent driving.

Almost all other EVs use a "guessometer" that changes its remaining range calculation based on recent driving - usually the previous 10 miles of your current drive. This does not provide an accurate measurement of remaining energy capacity, as it can be gamed. Example: if you save your city driving for the end of your test, the computer will adjust the "guessometer" upward based on your newly-efficient driving, and display a higher remaining range than it would had you continued driving on the interstate.

The red flag for me is that Consumer Reports only tests about half of the battery capacity on their 142-mile test. This opens the system up for the gaming that I describe above. If they drove the full range, they wouldn't be able to game the results with guessometer vs energy meter.

 

[Big Earl]

Interesting point. Wondering how the EPA got it so wrong, and why the numbers aren't skewed consistently across all the car makers? Not implying anything, I'm just thinking it through.

The EPA allows manufacturers to choose between a couple of different testing procedures (I believe 2-cycle and 5-cycle), and adjust the range downward if they want to. As a consumer, I have no idea which procedure is more accurate, and I don't know whether or not a manufacturer has arbitrarily lowered their figures to represent a more realistic highway range. The end result is that it's basically impossible to compare apples to apples.

I do know that my Teslas can get the rated range if I drive similar to the EPA test procedures in fair weather. I also know that the EPA test procedures are not at all representative of long distance, highway road trips.

I'd like to see more standardization here. Disallow arbitrary adjustments to reported range by manufacturers, and make them all follow the same procedure.

 

[Big Earl]

I think most on here would be happy if Tesla included an "estimated" range option similar to what they have on their energy screen that one could set the car to

the estimated could be based on a trip (or last 1000 miles) and/or lifetime average efficiency for that driver.. that is something they could easily add in an OTA update

Having owned two cars with range guessometers, I can unequivocally say that I do not prefer it over Tesla's approach (rated range display plus navigation estimation of % on arrival). The problem with guessometers is they only tell you what your estimated range will be based on recent driving while not taking into account potential changes down the road. You might be going along thinking you have enough range, then hit an elevation increase that you didn't know about, only to watch the guesstimation plummet. The same can be said for bad weather - the guessometer won't know anything about it, but Tesla's nav takes weather and elevation into account

 

[mongo]

From what I’ve seen, Tesla isn’t committing fraud, they are using the benchmark protocol that yields the best numbers.

The EPA allows manufacturers to choose between a couple of different testing procedures (I believe 2-cycle and 5-cycle), and adjust the range downward if they want to. As a consumer, I have no idea which procedure is more accurate, and I don't know whether or not a manufacturer has arbitrarily lowered their figures to represent a more realistic highway range. The end result is that it's basically impossible to compare apples to apples.

I do know that my Teslas can get the rated range if I drive similar to the EPA test procedures in fair weather. I also know that the EPA test procedures are not at all representative of long distance, highway road trips.

I'd like to see more standardization here. Disallow arbitrary adjustments to reported range by manufacturers, and make them all follow the same procedure.

ICE cars run the 5 cycle test to determine MPG, then they multiply that number by the manufacturer's reported tank size to get range.

EVs run the 5 cycle to get a Wh/mile rating, then that is converted to MPGe at 33.7kWh/gallon

Here's where things go wacky, range and MPG are separate, decoupled, tests for EVs:

EVs run the city cycle till they die and highway till they die (with results blended) or a mixed test (city, highway, city, constant speed, city, highway, city, constant speed) until they die to determine range.

Both range types commonly use a 0.7 adjustment factor, but there are other methods (or were as of 2017).

Not much for Tesla to tweak.

 

[GBMaryland]

Good god. CR's "test" isn't indicative of anything. Nerfed regen (Why? "fairness" isn't a good answer), extrapolated results from a guessometer (If car A has a less accurate guess than car B, it'll get a better range result). As usual, CR constructs a brain dead methodology to show what it wants to show. Usually it does this via ranking weights into categories that no one cares about, but in this case, they made up a testing methodology which isn't indicative of anything real world and used that.

I though regen couldn’t be turned off?

One of my biggest complaints is Tesla not allowing that on my MYP…. Especially after coming from a M3LR that I could turn off regen.

so, in order to turn off region, they had to edit settings in the vehicle, either through the maintenance menus or by installing a third-party device?

 

[Big Earl]

I though regen couldn’t be turned off?

One of my biggest complaints is Tesla not allowing that on my MYP…. Especially after coming from a M3LR that I could turn off regen.

so, in order to turn off region, they had to edit settings in the vehicle, either through the maintenance menus or by installing a third-party device?

I'm curious how they're doing that, as well, since new Teslas do not have adjustable regen settings. The previously available setting was low, not off.

 

[gt2690b]

I'm curious how they're doing that, as well, since new Teslas do not have adjustable regen settings. The previously available setting was low, not off.

SEXY buttons lol

 

[Big Earl]

SEXY buttons lol

Yeah that's the only way I can think of. If so, shame on them for testing a vehicle under conditions that aren't available to most users.

 

[ionsphere]

Having owned two cars with range guessometers, I can unequivocally say that I do not prefer it over Tesla's approach (rated range display plus navigation estimation of % on arrival). The problem with guessometers is they only tell you what your estimated range will be based on recent driving while not taking into account potential changes down the road. You might be going along thinking you have enough range, then hit an elevation increase that you didn't know about, only to watch the guesstimation plummet. The same can be said for bad weather - the guessometer won't know anything about it, but Tesla's nav takes weather and elevation into account

Ok, these are good points you're making, although I'm not 100% sure what a guessometer is, lol.
One thing I just figured out today, having only had my car for a little over a month, is that when the trial period of the premium connectivity ran out a few days ago, I lost my ability to look through the cameras on the app. I had thought something was wrong with my phone key not connecting, but the friendly guys at the service center steered me right. They also explained that not only was I losing things like being able to watch and listen to streaming channels(not that important to me since I can just Bluetooth it in the car for listening to music and podcasts which is mostly what I use), but also that the navigation would no longer take into account live traffic updates in my routing. I'm wondering if what you're proposing would also not work without the added subscription per month?

 

[Big Earl]

Ok, these are good points you're making, although I'm not 100% sure what a guessometer is, lol.
One thing I just figured out today, having only had my car for a little over a month, is that when the trial period of the premium connectivity ran out a few days ago, I lost my ability to look through the cameras on the app. I had thought something was wrong with my phone key not connecting, but the friendly guys at the service center steered me right. They also explained that not only was I losing things like being able to watch and listen to streaming channels(not that important to me since I can just Bluetooth it in the car for listening to music and podcasts which is mostly what I use), but also that the navigation would no longer take into account live traffic updates in my routing. I'm wondering if what you're proposing would also not work without the added subscription per month?

A guess-o-meter is an estimated range display. It attempts to guess at how far you’ll be able to drive based on recent driving. This term was lovingly coined by Nissan LEAF owners back in 2011, and is often abbreviated GOM.

Smart route planning and range estimations still work even without previous connectivity.