EPA fuel efficiency rating

[llamafilm]

Just for fun, I looked up the last few years of Tesla 3 LR at the EPA website (www.fueleconomy.gov) and these numbers don't make sense to me.
The 2022 model shows 260Wh/mi with 358mi total range. With simple math that equates to a battery capacity of 93kWh.
Similarly, the 2021 model shows 250Wh/mi with 353mi total range, which equates to 88kWh battery capacity.

From what I've read, the 2021 battery should be 75kWh and the newer ones should be 82kWh. What gives?

 

[Kimmi]

My 2021 LR was rated for 353 miles and 254kWh/mile. I don’t know capacity when new, but with 30K miles Stats shows 74 as maximum battery capacity.

 

[D Good]

What's interesting is that my 2021 was rated for 353 miles new, but I have the old 75kwh battery. Now the cars have an 82 kWh battery and are only rated for 333 miles. The car switched from the old Atom to the new Ryzen processor which did lower range, but shouldn't a significantly larger battery make up for that fact? Also, just 2 years prior, a very similar car, 2019 Model 3 LR, was rated for just 310 miles. What gives? I think it goes to show that these cars simply cannot go 353 miles on a single charge. With a 75kwh battery, the car would need to have an efficiency of 212 wh/mile. While this can be accomplished, you have to be going like 30mph with no wind, and this is using ALL of the 75kwh, something you should never and could never do. In short, Tesla inflates range numbers. Driving on the highway, I get around 210 miles of range. Abysmal.

 

[ucmndd]

I seem to recall EPA ratings include ALL energy that goes into the car, including the ~10% charging losses, etc.

 

[llamafilm]

What's interesting is that my 2021 was rated for 353 miles new, but I have the old 75kwh battery. Now the cars have an 82 kWh battery and are only rated for 333 miles.

That's a good point. Sounds like the EPA website is wrong then, and it should show 333 for the current model?

I seem to recall EPA ratings include ALL energy that goes into the car, including the ~10% charging losses, etc.

You might be onto something here, but I think charge efficiency should be better than that. Teslafi shows my charge efficiency is 94% over 2 years. The EPA numbers here are about 13% too high.

 

[E90alex]

If you calculate the range based on known battery capacity and the rated consumption, you get a range figure that’s actually more realistic to what you get in the real world.

 

[yesimon]

The MPGe calculations and Range calculations are not exactly the same. The range on the sticker is generally 55% city+45% highway (50-60 mph) driving mix with a penalty (adjustment factor of 30%). Meanwhile consumption (Wh/mi) does not have the adjustment factor, and yes also includes charging losses from the wall to battery. Charging losses are not factored into the range figure because that would make no sense. Additionally Tesla uses the 5 cycle EPA test versus 2 cycle EPA test used by most other manufacturers for a lower adjustment factor. I'll just quote from the link the most important bits.

This [range test] procedure uses the same EPA city cycle (20-mph average speed over 7.5 miles with 18 stops) and highway cycle (48-mph average speed, 10.3 miles) used for fuel-economy tests of all light-duty vehicles. Both cycles feature extremely gentle driving; the most aggressive acceleration would be equivalent to an 18-second 60-mph time. Between city and highway loops, the test calls for a steady speed of 55 or 65 mph to deplete the battery. The EPA knows these low-speed tests aren't representative of the real world, so every EV's window-sticker range is the product of an adjustment factor that yields a more realistic consumer-facing figure.


This is where it gets interesting. The default adjustment factor reduces the window-sticker range by 30 percent. So a car that achieves 300 miles of range during the city-cycle dynamometer test ends up with a 210-mile city rating. However, the EPA allows automakers the option to run three additional drive cycles and use those results to earn a more favorable adjustment factor. Currently, only Tesla and Audi employ this strategy for their EVs, and Tesla scores the most advantageous results, with adjustments that range from 29.5 percent on the Model 3 Standard Range Plus to 24.4 percent on the Model Y Performance. If Tesla had used the standard adjustment factor of 30 percent, the Model Y Performance's window-sticker range would drop to 292 miles. But because Tesla takes advantage of the EPA's alternate methodology, the company can instead claim a 315-mile range.

 

[llamafilm]

@yesimon if I understand correctly, you're saying actual tested range of the 2022 M3 LR is 511 miles, and the EPA rates it at 358 after a 30% penalty factor? That doesn't make sense to me; even in the most favorable driving conditions.

 

[AlanSubie4Life]

the 2021 battery should be 75kWh and the newer ones should be 82kWh. What gives?

2021 model year is when packs started switching to 82.1kWh (actually closer to 81kWh usually). Before that they were 77.8kWh. The first 2021 models with 353 miles (not 358) of range had 77.8kWh packs but started switching to the big packs mid-2021. (Performance 2021 always had the new pack type.)

Teslafi shows my charge efficiency is 94% over 2 years

I would not trust TeslaFi. It's not correct. (You can dig into the TeslaFi documentation to determine why. I think one reason is that it may report the kWh added in the car for "energy to the pack" which is not correct - it's well known that this number is 4.7% too high (1/0.955). But there may be other reasons; I don't use TeslaFi so I have no idea at all. TeslaFi is just a tool; it's not intended to be accurate.)

The answer to your question is as above - this is a wall-to-wheels number. Roundtrip losses are around 10%. The EPA results show about 11% (they provide charging event energy, and the energy metered during a discharge of the entire pack to empty - this gives you the losses). Note these losses depend on the charging setup but at least 240V/32A is assumed for these results.

And yes, the EPA tests get some very high numbers for city and highway, since they run them at very low speeds ( the highway cycle (HWFET) is quite slow for example, average speed is 48mph).

It's a complicated topic which you can dig into if you wish. All the documents are here; currently the website is down (this has been happening frequently lately): Basic Search | Document Index System | US EPA

You'll find the charge event energy and the discharge energy there. Your calculated numbers are pretty close to the charge event energy, by the way.

You can also go here, this is less useful though. However, it does allow you to derive the scaling factor used (ratio of adjusted to unadjusted economy).

Download Fuel Economy Data

Fuel economy Fuel Economy Datafiles (1978-present), Fuel Economy Guides (1996-present)

www.fueleconomy.gov

This has all been extensively documented on TMC in the past - you just have to search with the right keywords, and do some research.

Currently Tesla is using an adjustment factor of about 0.747 for the Model 3 as I recall. Before the heat pump it was closer to 0.703 (but there were a couple of exceptions).

Just for fun,

Not that fun, tbh.

 

[yesimon]

@yesimon if I understand correctly, you're saying actual tested range of the 2022 M3 LR is 511 miles, and the EPA rates it at 358 after a 30% penalty factor? That doesn't make sense to me; even in the most favorable driving conditions.

Yes, 25% penalty factor, driving below 60 mph on flat ground, no A/C, running until the wheels stop moving (using the entire battery buffer below 0%)