Small turbo engines don't deliver(consumer reports)

[enzo1]

[onewhippedpuppy]

Here's the article: Consumer Reports finds small turbo engines don't deliver on fuel economy claims

Lots of holes in their argument, which is essentially that Consumer Reports did not obtain the EPA MPG in their testing with turbocharged 4-cylinder engines. The EPA testing is strictly defined and government regulated to provide an apples to apples comparison of every vehicle. Consumer Reports testing is? I seriously doubt their testing is anywhere near as rigorous and repeatable as the EPA testing, pretty convenient that they in no way define their MPG test procedures. As we all know there are many factors that contribute to gas mileage, their data seems coincidental at best.

Having had several turbocharged 4-cylinders that delivered great performance and great MPG, I know it is certainly possible. My 1993 Saab 9000 Aero had 220 HP stock, and even after equipping it with a manual boost controller I could get 35 MPG on the highway from its 2.3L 4-cylinder. Great car......

[gprsh924]

Wouldn't a more valuable comparison be taking 2 ford fusions, one with the turbo and one with a conventional engine and comparing 0-60 times and fuel economy?

[notfarnow]

pretty sure he said fuel "ecomony" at the start.

[jyl]

They do provide some information about their fuel economy test procedures. Car Testing | Auto Test Center - Consumer Reports

My recollection of my time driving 4 cyl turbo cars was that they got good mileage if you dribbled along at low rpm off boost, but if you were on boost a lot the mileage was a lot worse. Maybe the CU tests involve more acceleration than the EPA tests. Which is more realistic? I don't know but I seldom hear anyone praising the EPA tests for realistically simulating real-world driving. In fact, when talking about hybrid mpg (e.g. Prius), people here quickly discount the EPA tested mileage.

[McLovin]

In the article, it says the EPA figures are from laboratory simulations.

IMO, the Consumer Reports data are valid data points.

Fuel economy of course depends on a lot of variables, that are always changing, but their testing methods seems sound, and their results are valid and useful data points.

[fluque]

Originally Posted by jyl

My recollection of my time driving 4 cyl turbo cars was that they got good mileage if you dribbled along at low rpm off boost, but if you were on boost a lot the mileage was a lot worse. Maybe the CU tests involve more acceleration than the EPA tests. Which is more realistic? I don't know but I seldom hear anyone praising the EPA tests for realistically simulating real-world driving.

You are spot on. I drive a small yet peppy 2.0 lt turbo car and mileage varies greatly depending on my driving habits, much more than other fuel injected vehicles I've owned.

IMO the question is how should they test these cars to get a reasonable result. When I just bought my car I was on the boost most of the time (enjoying it), and now much less. The issue at hand is that the same car/driver may get very different results at different points.

[BReif61]

Originally Posted by gprsh924

Wouldn't a more valuable comparison be taking 2 ford fusions, one with the turbo and one with a conventional engine and comparing 0-60 times and fuel economy?

There you go thinking logically.

[Jferr006]

We have a turbo Sonic and it gets much better fuel economy than claimed. Driving habits must have a pretty decent impact on the results. I also thought it would have made more sense to compare a model's turbo vs. non turbo models to prove that point rather than, "hey theres another more efficient car out there w/o a turbo".

[svandamme]

My 1.6L Turbo diesel Bmw uses about 1 Liter/100km less then the 2.0L BMW it replaced.
I lost a bit of acceleration and some top speed but in general i still drive like i stole it...

[onewhippedpuppy]

Originally Posted by jyl

They do provide some information about their fuel economy test procedures. Car Testing | Auto Test Center - Consumer Reports

My recollection of my time driving 4 cyl turbo cars was that they got good mileage if you dribbled along at low rpm off boost, but if you were on boost a lot the mileage was a lot worse. Maybe the CU tests involve more acceleration than the EPA tests. Which is more realistic? I don't know but I seldom hear anyone praising the EPA tests for realistically simulating real-world driving. In fact, when talking about hybrid mpg (e.g. Prius), people here quickly discount the EPA tested mileage.

I didn't see this, in my opinion it still leaves a lot of room for error.


From the CR site:
Fuel economy

We perform our own fuel-economy tests, independent of the government's often-quoted EPA figures and the manufacturers' claims. Using a precise fuel-flow measuring device spliced into the fuel line, we run three separate circuits. One is on a public highway at a steady 65 mph. That circuit is run in both directions to counteract any wind effect. A second is a stop-and-go simulated city-driving test done at our track. The third is a 150-mile "one-day trip" using several drivers taking turns around a 30-mile loop of public roads that include a highway section, secondary roads, and rural byways. CR's overall fuel-economy numbers are derived from those three fuel consumption tests


Public highway at 65 MPH - adjusted for temperature, seasonal fuel blend?
Stop and go simulated city test - how do you maintain consistency in acceleration, braking, etc?
Day trip - how do you maintain consistency for acceleration, braking, etc? How do you adjust for differing traffic patterns, weather, fuel blend?

[onewhippedpuppy]

I totally agree that many real world results do not agree with the EPA testing. Frankly that's inevitable, because they are testing on a dyno in a controlled lab setting. But at least the EPA testing provides you with a comparison tool between vehicles, where you know each has been tested with the exact same test conditions. As soon as you introduce different people and real roads you introduce error into your testing.

How Vehicles Are Tested

[jyl]

There's obviously more controls to the CU procedure than are described in the single summary paragraph. I think you can assume the testing is designed by engineers who thought of everything you have thought of and more. Is it as repeatable as dyno testing - surely not. However, you don't test airplanes on a dyno, and yet manage to get reasonably valid data.

The thing with the EPA testing is that the auto companies know the procedure in considerable detail. So they could, hypothetically, design to the test. In this day and age, program software to the test. Do they . . .

[afterburn 549]

My Abarth gets over 30 no matter how hard I beat it........and I have a lot of hills around here

[onewhippedpuppy]

Originally Posted by jyl

There's obviously more controls to the CU procedure than are described in the single summary paragraph. I think you can assume the testing is designed by engineers who thought of everything you have thought of and more. Is it as repeatable as dyno testing - surely not. However, you don't test airplanes on a dyno, and yet manage to get reasonably valid data.

The thing with the EPA testing is that the auto companies know the procedure in considerable detail. So they could, hypothetically, design to the test. In this day and age, program software to the test. Do they . . .

For aerospace we repeat the exact same test conditions multiple times on a fully instrumented aircraft, with a team watching the test parameters real-time via telemetry. Test conditions are very rigid, and it's not unusual to throw out more than half of the conditions because they don't meet the criteria. We strive to utilize the same pilot for a given set of data to maintain consistency. Allowable weather conditions are also strictly controlled not only from a go/no-go perspective but also in real time, with actual measured weather conditions being recorded for each performed test condition. Ultimately all of the retained test conditions are normalized for the actual measured weather conditions. So yeah, I'm a little anal about this stuff.

[jyl]

How accurately are you trying to measure a given characteristic?

What I mean is, and this probably isn't the right terminology for aerospace, but suppose you are trying to measure fuel consumption at a given airspeed, altitude, etc. Say your tests return sonething around 6.5 gph, making up a number. Do you test enough to determine that it is 6.3 to 6.8 gph to a 95% confidence?

Same question for other performance metrics, rate of climb or roll or whatever.

[rick-l]

There is an article in Automobile magazine this month about how the EPA mileage is calculated.

From what I remember. They say it is done by the manufacturer. The biggest factor is the coast down from 70 mph which is fudged all over the place. The driver trains for days to get the least acell to match the speed profiles of the EPA course.

They say Kia/Hyundai is paying owners because they fudged the numbers past "the limit".

[onewhippedpuppy]

Originally Posted by jyl

How accurately are you trying to measure a given characteristic?

What I mean is, and this probably isn't the right terminology for aerospace, but suppose you are trying to measure fuel consumption at a given airspeed, altitude, etc. Say your tests return sonething around 6.5 gph, making up a number. Do you test enough to determine that it is 6.3 to 6.8 gph to a 95% confidence?

Same question for other performance metrics, rate of climb or roll or whatever.

It totally depends on the test, the more critical the data the tighter the requirements. Performance data is very intensive, where flight conditions can have very tight tolerances on things such as altitude, angle of attack, stabilized airspeed, etc. Landing data is critical too with things like sink rate, angle of attack, and airspeed. Ultimately it's all difficult for the pilot to hit, so lots of tests are required to get enough data. I'm not one of the data crunching types so I don't know what specific statistical confidence interval they work to, but it's pretty tight and all subject to FAA review. There's a reason why airplanes are expensive and take years to certify.

[afterburn 549]

Its a Time Space Continuum How to measure...what to measure, when to measure

[Tervuren]

Originally Posted by onewhippedpuppy

It totally depends on the test, the more critical the data the tighter the requirements. Performance data is very intensive, where flight conditions can have very tight tolerances on things such as altitude, angle of attack, stabilized airspeed, etc. Landing data is critical too with things like sink rate, angle of attack, and airspeed. Ultimately it's all difficult for the pilot to hit, so lots of tests are required to get enough data. I'm not one of the data crunching types so I don't know what specific statistical confidence interval they work to, but it's pretty tight and all subject to FAA review. There's a reason why airplanes are expensive and take years to certify.

I believe the first live rocket ejection was done by a navy test pilot at low altitude and inverted while testing maximum speed of control in a super sonic dive.

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

Notice the loss of hair on ALL navy jet fighter pilots. Night Carrier landings in the clouds in a jet fighter were not kind to your scalp in those days!