If the official EPA range is....

[Hmp10]

I'd say the guy that follows the mandated protocol has more integrity than the guy that sand bags but that's just me.

You're focusing on adherence to a testing protocol and forgetting what the real purpose of range ratings should be: to give buyers realistic guidance on how a car goes in the mix of driving most people do.

There is a long enough history of EVs falling short of their range ratings and enough accumulated data on the variance by brand for testing protocols to be adjusted to bring everyone closer to realistic estimates.

The fact is that a Porsche range rating is more realistic than a Tesla range rating. Instead of arguing that Porsche should have adhered to an arbitrary protocol, I think the argument should be to change the protocol to bring it more into line with what Porsche did to obtain a more realistic mileage range.

You are more interested in method and compliance. I am more interested in ultimate purpose and outcome.

Sponsored

 

[ajdelange]

Edmunds testing recently found that the Model Y comes closest to delivering its EPA-rated range in real-world driving of any Tesla to date. So, at least for the non-performance models with which Tesla plays all kinds of numbers games, they're getting more frank.

Tesla isn't playing any numbers games though clearly that's what you want them to do. Tesla runs the full set of tests and reports the number and the EPA approves it and issues it. Let's say your Tesla's EPA range is 300 mi just to make the calculations easier. I think you want Tesla to report 83% of that (249 miles) because you experience 83% "efficiency" in most of your driving as does the average S driver. But to ask them to do that ill serves me who averages 118% "efficiency". I wan't them to publish 354 miles.

I know you aren't a techie and that most of the relevant concepts are foreign to you but what you consistently seem to miss is that not everyone drives in the same way and under the same conditions as everyone else such that only way we can compare cars on what even comes close to an apples to apples basis is to promulgate some standardized set of conditions, test all cars against that set of conditions and predict performance under real conditions based on deviations from the standard. We recognize that there are some short comings to this approach but we really can't do better. So this is what SAE and EPA have done. If a manufacturer such as Tesla chooses to follow the test procedure well and come up with a good number but Porsche doesn't I think it incumbent on the government to refuse to approve the Porsche number. For whatever reason they haven't chosen to do this. Porsche's numbers are clearly unrealistic. If their EPA number represents performance at highway speed then it clearly doesn't represent the performance against the EPA profile set as that is not the intent of the EPA profile set. If you cannot understand this you will never grasp the big picture.

 

[Trandall]

For some reason I feel the need to summarize the EPA vs. real world debate so here goes:
The average consumer is largely oblivious to what makes things tick and wrongly assumes a EPA range test should closely resemble their unique use case. Since some people drive primarily in San Francisco Bay area, some in Minnesota, and others in Arkansas this is impractical. The EPA/ SAE or whomever instead has a set of "standardized" repeatable tests that can be helpful as a point of reference and unfortunately not much else. To further complicate matters Manufacturers can select from different tests and apply a correction factor at there discretion, subject to review by testing regulator. The result is customers read numbers from different manufacturers that have used different test and expect them to match their use case. To get an accurate idea of range what should happen is customers reference their own use case to the EPA result and they will have a custome repeatable result they are looking for. The sad part is a significant portion of the population is either incapable or unwilling to do this. So to conclude the proper solution to this probably needs to start in 2nd grade math class and go from there.

 

[Hmp10]

I know you aren't a techie and that most of the relevant concepts are foreign to you but what you consistently seem to miss is that not everyone drives in the same way and under the same conditions . . . .

I know you're dismissive of Edmunds testing. However, they explain their range testing method clearly, and it has been applied consistently over a series of reports.

Until just a few months ago, Teslas consistently fell shorter of their EPA ranges in Edmunds testing than all other brands. That suddenly changed in April with the Model Y.

If Tesla had been using the same EPA protocols all along and Edmunds has been using its own testing protocols consistently across all these tests, why is the Model Y suddenly the first Tesla that is almost reaching to its EPA range in Edmunds testing?

It would seem that either Tesla used different methods from its other models for getting an EPA range assigned to the Model Y, or Edmunds changed its testing methods just for the Model Y.

 

[SeaGeo]

You list your profession as engineer. As an engineer you should understand why this is the case and be perfectly comfortable with it.

I am. Just noting the "why" for you.

As noted by others, Porsche (and others) take responsibility to try and make sure their EPA rating is either generally reasonable or conservative in the use case where range matters for the customers. Tesla does the opposite and tries to show as much range as possible with the EPA testing available to them for marketing purposes. The issue people take is that it results in an inconsistency in the EPA ratings relative to real world performance at highway speeds (where people generally actually care about range). And in those conditions, Tesla seems to be an outlier in how poorly they perform relative to their EPA ratings. Other manufacturers do tend to do better. So when you get a Mach e for example, or an ID.4, most people will get close to the EPA combined range (which is the one prominently displayed that most users expect to get), and tesla's generally won't.

Which makes it a failure in the EPA's processes and communication that Tesla is taking advantage of.

 

[ajdelange]

You're focusing on adherence to a testing protocol and forgetting what the real purpose of range ratings should be: to give buyers realistic guidance on how a car goes in the mix of driving most people do.

That is exactly what the EPA profiles are designed around but that is not really what they are intended to do. They are intended to serve as a standard for comparison. Now you may think that should be to give buyers an idea as to how the car goes but that is
a)not what they are for
b)and impossible goal to reach as there is too much variance among driving habits and conditions. What profile would you propose be used to achieve your goal?

There is a long enough history of EVs falling short of their range ratings and enough accumulated data on the variance by brand for testing protocols to be adjusted to bring everyone closer to realistic estimates.

Again I will ask what profile would you propose to bring the retired businessman in Florida AND the working logger in Saskatchwan closer to realistic estimates.

The fact is that a Porsche range rating is more realistic than a Tesla range rating. Instead of arguing that Porsche should have adhered to an arbitrary protocol, I think the argument should be to change the protocol to bring it more into line with what Porsche did to obtain a more realistic mileage range.

What Porsche did was blow the tests or shamelessly sandbag although I did not think the rules were such that they would be allowed to do so.

You are more interested in method and compliance. I am more interested in ultimate purpose and outcome.

As an engineer I understand that the current system, warts and all, is as good as it is going to get and I understand how to work with it to get the information I need. What you seem to want is something you can't have in a single number. The EPA number, as it stands, represents nominal range under a certain set of driving conditions. As the results don't give you values you are happy with it is incumbent on you to come up with a test protocol that gives you a number you would like better. You could just change the rules to say "Use the current protocol but multiply the result by 0.83 but as noted that's going to make me unhappy. I'd wan't changed to multiply the result by 1.18 because that's what i average overall WRT the EPA rating for my car. In case anyone thinks I am serious about this I want it left as is and I want all the manufacturers to be forced to do the complete suite and I want closer supervision by EPA to be sure they are doing the testing correctly. As things are now I can't use the EPA ratings to compare Porsche to Tesla which is what the EPA rating is intended to do. I repeat that it is not there to give people a realistic idea as to how far the car will go on a charge though I must say that with my Teslas it has been pretty close (2.5% more in town, 6% less on the freeway).

 

[Hmp10]

To further complicate matters Manufacturers can select from different tests and apply a correction factor at there discretion, subject to review by testing regulator. grade math class and go from there.

Yes. This is the reason the whole argument about the reliability of EPA testing for relative comparisons across brands breaks down.

It's why I think the brand that selects the tests and correction factor that yield the most conservative range rating has more marketing integrity than the brand that makes the choices that yield the most optimistic EPA rating.

 

[ajdelange]

So to conclude the proper solution to this probably needs to start in 2nd grade math class and go from there.

A lot of it has to do with the mechanism of FUD. You may have noticed a certain animus against Tesla here. And we all know that Elon is a scoundrel whose family owned a diamond mine, whose cars kill people with their autopilots and burst into flame regularly. Once these things get out there it is nearly impossible to scotch them. So of course Tesla cheats on its range tests. Who can deny that? Some smartass who happens to have been an engineer all his life? If you don't understand the basic concepts how can he convince you?

 

[ajdelange]

Yes. This is the reason the whole argument about the reliability of EPA testing for relative comparisons across brands breaks down.

It's why I think the brand that selects the tests and correction factor that yield the most conservative range rating has more marketing integrity than the brand that makes the choices that yield the most optimistic EPA rating.

You are saying that Tesla's robust testing should be thrown out in order that ressults compare to Porsche's flawed testing because the latter better represents performance under conditions that don't represent what EPA wants tested. Are you sure you want to say that?

 

[ajdelange]

I am. Just noting the "why" for you.

Roger that.

 

[ajdelange]

I know you're dismissive of Edmunds testing. However, they explain their range testing method clearly, and it has been applied consistently over a series of reports.

I doubt the consistency and also the number of replications. I've done a few tests in my day and I have a basic understanding of how to design and run a test series.

It would seem that either Tesla used different methods from its other models for getting an EPA range assigned to the Model Y, or Edmunds changed its testing methods just for the Model Y.

Those are two possibilities. Another is that the Y has smaller A*Cd so that drag does not emerge as fast but the most likely is that Edmund "changed" its testing conditions, perhaps quite unintentionally. There is a reason these tests are done in lab rather than outdoors on a track.

 

[Hmp10]

You may have noticed a certain animus against Tesla here.

I own a Tesla and bought my brother one. I also hold a significant amount of Tesla stock. Several weeks ago I went to the Tesla website to order a Model X Plaid. (I stopped when I noticed that the model refresh failed to add a center armrest, beverage holders, or storage pockets to the second row.)

My animus is against manipulative marketing practices. For instance, why does Tesla publish 0-60 times using rollout only for their performance models, while not factoring rollout into the numbers they publish for the standard models, thereby exaggerating the performance difference between the standard and performance models (which just happen to be far more expensive)?

I remember a discussion some months ago when Lucid published range numbers the Air attained when FEV tested the car using EPA protocols. You argued ad nauseum about all the stratagems Lucid and/or FEV might have used to produce a range number that embarrassed Tesla.

So animus can show up anywhere.

 

[ajdelange]

As I recall in that discussion I pointed out a few things that Rawlinson said that were certainly if not untrue definitely intended to make his product look as good as possible and I distinctly remember saying that I find that perfectly acceptable within in limits and I certainly would never assert that Musk doesn't do plenty of that. There is definitely a bit of the snake oil salesman in both those guys.

I also remember trying to get you to understand that when you send your prototype to a lab to see what its EPA rating might turn out to be you have more flexibility than when the test is being run "for the record" in order to obtain EPA approval.

Now Scaringe does not seem to be a snake oil salesman at all(remember this is a RIvian forum). I certainly feel no animus against Peter Rawlinson or his company. In fact I was delighted to see a large space across from the local Apple store has been taken by them and I certainly hope that they will be a huge success (full disclosure: I am long CCIV).

Bringing this back to Rivian: They are a class act and I expect them to do the full set of EPA tests. I therefore expect their "real world" numbers to be as much lower than the EPA rating as Tesla's are and perhaps even more so as A*Cd will be larger than Tesla's sedans and SUVs. Of course this is speculation on my part.

 

[Gshenderson]

FYI - if you drive down the east coast on I-95, you will find many places where the speed limit exceeds 65. In Virginia, we have 70 mph limit highways, so driving 75 mph is pretty average on those roads.

In large swaths of the high prairie and Rocky Mountain states, the speed limit is 80mph!

 

[ajdelange]

As noted by others, Porsche (and others) take responsibility to try and make sure their EPA rating is either generally reasonable or conservative in the use case where range matters for the customers. Tesla does the opposite and tries to show as much range as possible with the EPA testing available to them for marketing purposes. The issue people take is that it results in an inconsistency in the EPA ratings relative to real world performance at highway speeds (where people generally actually care about range).

I'm afraid we have to add you to the list of those who don't understand what the EPA rating's purpose is. It is not intended to reflect real world performance at highway speed. It is intended to reflect performance against 5 profiles which blend highway and city driving. As the average speed over these profiles is appreciably lower than that of a pure freeway trip it is to be expected that there will be additional burden on the battery when driving at highway speed relative to when driving the EPA profiles. Thus it is irresponsible for Porsche to sandbag (IMO) and responsible of Tesla to report what EPA wants them to report.

So when you get a Mach e for example, or an ID.4, most people will get close to the EPA combined range (which is the one prominently displayed that most users expect to get), and tesla's generally won't.

As an engineer you should understand that the highway range has to be less than the combined range and that, therfore, a Monroney sticker that displays the highway range is not displaying the EPA range. Users should NOT expect to get the EPA range on the highway. The problem seems to be that many do.

This problem could be solved by redefining the test requirement to test against highway conditions but which highway conditions? Out west where the speed limit is 75 or around one of the metropolitan areas where it is 55 or 65?

Sponsored