Jump to content


Photo
- - - - -

real world fuel consumption


  • Please log in to reply
146 replies to this topic

#1 mariner

mariner
  • Member

  • 1,367 posts
  • Joined: January 07

Posted 14 March 2013 - 09:00

A lobbying group " transport and enviroment" have produced a report to highlight " cheating " by car mfrs on the EU CO2 drive cycle.

They are claiming that real world fuel consumption is far worse than the EU figures. Of course that is absolutely not news to most people here , or the OEM's but this report is the most detailed analysis I have seen of how and why.

http://www.transport...20v15_final.pdf

They may be guilty of some bias but , personally, I think the report is fairly balanced .

I also learnt that a new world wide CO2 test is in process the " world light duty test cycle" common to EU, USA and Japan.

The report is very detailed and captures many of the known tricks like squeezing down and inertia class ( Chrsyler were doing that in the 1970's! for US CAFE!)

Of course the real point is that if you make the very commercial survival of a car maker dependent on a lab. test you should expect the test procedure to be minutely analysed and optimised. Juat like the F1 rules reading and being coached on to best approach exam questiuons. Smart people, when heavily motivated , find smart solutions.

As an aside here id teh popular journalism summary of teh same report.

http://www.bbc.co.uk...siness-21759258

Advertisement

#2 Greg Locock

Greg Locock
  • Member

  • 4,490 posts
  • Joined: March 03

Posted 14 March 2013 - 09:36

The real point being that the tests should be defined by smart people, not lobbyists and politicians.

As an example, the driveby noise regs. The American test is a damn sight more sensible than the EEC one, which encourages some rather odd emphasis in exhaust tuning which bears little resemblance to the real world, plus of course modern cars on their first exhaust aren't even 5% of the noise problem in cities.

Mind you, a certain American icon had a probem passing US driveby, so they ran an Italian icon past the bloke from the government, and said that obviously if that was acceptabe then the American icon was. He agreed.

In the beeb report:

disconnecting the alternator, thus no energy is used to recharge the battery during the test
the use of special lubricants that are not used in production cars, in order to reduce friction
turning off all electrical gadgets such as the air-conditioning or the radio

The first two are obvious fiddles and I am amazed they aren't covered.

The third should be defined in the test procedure, frankly if it isn't there whose fault is that? Given that a/c was very optional in europe when the rules were written then it's not surprising it wasn't included. My employer does't include an option (such as A/C) in its standard vehicle definition unless it has a 50% take rate. I'm kind of guessing that small european hatchbacks have less than 50% AC.

Incidentally, big hint to greeny boys writing technical reports: Don't be so whiny. That report is much harder for me to take seriously because of its holier than thou tone. Since I agree with the basic premise, that is a pretty bad outcome.

Edited by Greg Locock, 14 March 2013 - 09:59.


#3 Magoo

Magoo
  • Member

  • 2,458 posts
  • Joined: October 10

Posted 14 March 2013 - 09:49

The real point being that the tests should be defined by smart people, not lobbyists and politicians.


Oh technocracy. If only we could get the smart people to ever agree on anything, or to turn in a coherent report on time.


#4 Greg Locock

Greg Locock
  • Member

  • 4,490 posts
  • Joined: March 03

Posted 14 March 2013 - 10:02

Oh technocracy. If only we could get the smart people to ever agree on anything, or to turn in a coherent report on time.

Fair cop.

Actually I'm a bit annoyed with the thrust of that report. There is strong economic pressure on manufacturers to improve their test results as against real world utility and the authors seem inclined to whinge on about the manufacturers doing as they are asked to do. If the EEC issued a directive to improve real world fuel economy, without increasing braking distances, with a/c switched on, hey guess what, we'd do that, if some lazy tax payer funded arse will define a suitable test. But if the only test is an unrepresentative badly drafted procedure then that's what we'll use. Because that's what he have to do. Meanwhile 3 major European OEMs are going bankrupt and all the rest have been sold off, except the German ones.




Edited by Greg Locock, 14 March 2013 - 10:12.


#5 Magoo

Magoo
  • Member

  • 2,458 posts
  • Joined: October 10

Posted 14 March 2013 - 10:06

A lobbying group " transport and enviroment" have produced a report to highlight " cheating " by car mfrs on the EU CO2 drive cycle.

They are claiming that real world fuel consumption is far worse than the EU figures. Of course that is absolutely not news to most people here , or the OEM's but this report is the most detailed analysis I have seen of how and why.


It wouldn't be terribly difficult to identify and correct the problems in the test regime and generate more realistic figures. But then you have the problem of a new set of test figures way out of line with the first set. I suppose the solution there is to steer back the test process toward realism in multiple phases.


#6 Greg Locock

Greg Locock
  • Member

  • 4,490 posts
  • Joined: March 03

Posted 14 March 2013 - 10:27

The report agrees with one of my bugbears -autostop. It is borderline useless in the real world. A car at idle typically burns 0.25 gall/h. when it drives it is burning around 1 gall/hr. So, if you are idling the car for 20 minutes per hour, the improvement is ~10%, but if your drive is 20 minutes per hour idling, you'll need the engine on to run the a/c and so on.

My own usage is more like 10% idling, and it would be ok for the a/c to die in that period, but that means the fuel saving is correspondingly small.

I'll read the rest tomorrow.



#7 Magoo

Magoo
  • Member

  • 2,458 posts
  • Joined: October 10

Posted 14 March 2013 - 10:31

Fair cop.

Actually I'm a bit annoyed with the thrust of that report. There is strong economic pressure on manufacturers to improve their test results as against real world utility and the authors seem inclined to whinge on about the manufacturers doing as they are asked to do. If the EEC issued a directive to improve real world fuel economy, without increasing braking distances, with a/c switched on, hey guess what, we'd do that, if some lazy tax payer funded arse will define a suitable test. But if the only test is an unrepresentative badly drafted procedure then that's what we'll use. Because that's what he have to do. Meanwhile 3 major European OEMs are going bankrupt and all the rest have been sold off, except the German ones.


Do the manufacturers have any ethical responsibilities in this test process or do we just take whatever they can get away with and call it ethical? Not trying to be a smartass here, I'm asking.



#8 just me again

just me again
  • Member

  • 1,122 posts
  • Joined: August 00

Posted 14 March 2013 - 11:50

I think in Denmark the cars get a new car tax discount on 4000kr for every km/l it goes above 16km/l(for petrol cars). and the year by year taxes is also smaller the better test result.

So i think there is no limit's for the manufactures in trying to get the best mileage. In the small car segment ( from the smallest up until the Polo class) i think the test is a matter of life and death.

Bjørn

#9 mariner

mariner
  • Member

  • 1,367 posts
  • Joined: January 07

Posted 14 March 2013 - 18:03

It amazes me that anybody should expect mfrs not to " game" a set of rules which control their actual survival. Today if an OEM doesn't pass the rules on fuel efficiency, emissions and crash safety they will simply go out of busines. Even if you are 100% anti capitalist etc. you can see such failure causing tens of thousands of job losses.

In such circumstances the engineers would be irresponsible NOT to optimise versus the rules if they believe other people are doing so.

The easy bit is correcting the test results to the real world. The harder , but more important , bit is preventing people engineering in things which disproportionally help the test versus real world results - like stop/start.



#10 Magoo

Magoo
  • Member

  • 2,458 posts
  • Joined: October 10

Posted 14 March 2013 - 20:43

Gaming the process to some degree or at least exploiting its characteristics to the fullest doesn't surprise me at all. That's SOP.

However, when we get to stunts on the order of swapping out lubricants, taping up panel gaps, etc, the people pulling this stuff need to take a step back and ask themselves WTF they are doing.


Edited by Magoo, 14 March 2013 - 20:53.


#11 Greg Locock

Greg Locock
  • Member

  • 4,490 posts
  • Joined: March 03

Posted 14 March 2013 - 22:37

I haven't got to the actual cheating bit yet, but knowing the squeaky clean ethical stance of our senior management (or so they tell us (once you can fake sincerity your path is upward)), I can't believe that that would be acceptable now. For instance in braking tests we don't optimise the tire pressures, we set the car up as per the manual. My own experience with certification or sign off tests is limited to driveby noise, which isn't life and death, but is expensive, and roll over stability of SUVs, which is life and death, and is taken very seriously. Fixing it mechanically often isn't expensive but absolutely hammers other attributes such as fuel economy, emissions, ride, NVH, brakes. That's why ESC/ROM is becoming de rigeur.

Driving in to work I came to the same conclusion as you, the procedures need to be modified year by year. I guess you have to give manufacturers the option of quoting the mpg at a given certification date, maybe an umbrella organisation would have to establish year to year equivalence. Yes that is very ugly.


#12 Greg Locock

Greg Locock
  • Member

  • 4,490 posts
  • Joined: March 03

Posted 14 March 2013 - 23:26

Right I skimmed through it. I'm not very convinced by some of the cheating claims, but will start by saying that the graphs A2.3 and A2.4 tell an ugly tale. Fig A2.3 is pretty damning evidence that manipulation is routine, although the AC side of things is huge hit by itself. A2.4 rather bafflingly shows a much lower figure, and less of a trend for the directly comparable NEDC test hot and cold. Note that I'm not disputing that the NEDC doesn't agree with real world, it is a silly test profile and wasn't intended to be used to predict real world fuel consumption. The American one is much better.


Quotes and comments

"The same applies to cars, so a test vehicle will roll much farther when the tyres are pumped up as hard as possible. This practice is not specifically excluded in the test procedure, in spite of the fact that it would be dangerous to drive a car in this condition on a real road."

Actually pumping your tires up above the placard figure is recommended procedure on all advanced driving courses I have been on. Now, they don't give figures for this, I'm guessing they mean 50 psi not 38 psi.

"For example, optimising wheel and tyre specification to increase rolling radius by 5% will change the test result by about 2%"

I'm sure it would. However a 30mm difference in tire diameter will not fit in the wheel arch, typically it will rub at full lock and in jounce.

If changing the wheel alignment inside spec limits alters fuel consumption by 20% I'll eat my hat.

Running in. it seems odd to argue that real world is so important, where the average 'new' car has 15000 km on it, and then whinge that some OEMs run the car in for more than 3000 km

"Clearly a dead flat and dead level test track is the only fair basis on which to conduct a coastdown test. However, procedures only specify that any slope on the track must not exceed 1.5%. Test tracks are often designed with precisely this degree of slope, although it is possible using modern methods to create a test surface that is much more level than this. The test must be completed in twice in opposite directions, but Zallinger and Hausberger (2009) point out that the simple averaging method that is used to combine the two results is technically incorrect and still favours results obtained on a sloping track."

Interesting claim. That one at least seems directly checkable.

Refs

Bizarrely they don't link to the refs. Not very helpful when they are casting aspersions of naughtiness. Here's the ones I found so far

Mock P, German J, Bandivadekar A, and Riemersma I, 2012,

http://www.theicct.o...gpaper_2012.pdf

Smokers R, Kadijk G and Dekker H,

http://www.tno.nl/do...ions_051212.pdf

Edited by Greg Locock, 14 March 2013 - 23:28.


#13 Lee Nicolle

Lee Nicolle
  • Member

  • 5,735 posts
  • Joined: July 08

Posted 15 March 2013 - 04:31

I just glimpsed this but all of this testing and figures really are not real world.The manufacturers like to optimise the numbers, the governments make stupid demands so the whole thing is a debacle.
Drivers diferentiate, conditions do by the hour with wind, rain etc. That itself can make a 20% difference.
Having done the car club economy runs in the past bad driving practices can make good figures.
Though manufacturers IF they are overinflating tyres[max handbook pressures though can be constued ok] and taping gaps etc are not playing the game. Though as Greg says the numbers seem very 'crook' Wheelalignment on a production car will make a very miniscule difference, tyre pressures will even taping will make little difference at highway speeds, high speeds yes.
Like most things you take the 'government figures' with a boulder size grain of salt.

#14 murpia

murpia
  • Member

  • 342 posts
  • Joined: September 06

Posted 15 March 2013 - 16:37

Page 24 describes the 'conspicuous distribution' of the inertia classes of test vehicles.

What about the 'conspicuous distribution' of the published EU CO2 g/km figures? They all seem to end with a 9: 119g/km, 129g/km, 139g/km etc. Just like a shop pricing everything at '$xx.99'...

Regards, Ian

#15 desmo

desmo
  • Tech Forum Host

  • 12,908 posts
  • Joined: January 00

Posted 15 March 2013 - 17:54

Wouldn't it be pretty simple methodologically to get to more realistic figures by looking at actual user data from the now ubiquitous onboard systems, comparing it to claimed figures and applying a fudge based on the differences? Maybe if some manufacturers are significantly more or less accurate in their claimed figures vs. real world data, apply manufacturer specific fudges that take that into account for new cars.

#16 mariner

mariner
  • Member

  • 1,367 posts
  • Joined: January 07

Posted 15 March 2013 - 18:30

Going back to Magoo's ethics question I would rate the " damage to society" in this ascending order.

- engineers gaming/cheating the rules of the tests

- the large and well established disconnect betwen test and real world mpg/CO 2per km

- the activites of the mfrs designing cars to do well in the test not the real world

Maybe engineers shouldn't play the rules but the test centres are just like race scrutineers at the pre test stage. If it turns up with taped over panel gaps just throw it out.
What is standard for the test is harder as there are many options on each model, mfg tolernaces exist and the test has to be simple and repeatable. Ive read Volvo disconnected their running lights for the EU test when only they fitted them. I can live with that but I've also heard that only one tyre size is chosen for rolling restaance and some mfrs arrange for that to be a low grip ecomony tyre - that I do think is very, very wrong as it impacts customer safety.

I think all the OEM's use very skilled drivers to follow the test protocol at minimum throttle They may been traiined just like F1 pit crews but that's not cheating even if it's unrealsitic.

The tests are obviuolsy far from the real world, and very short, so the fuel economy is false. However that can be fixed by applying a std. factor acros the board of , say, +20% reduction. It has to be somewhat arbitary as every car is different. Whether that factor needs to be increased as the test versus real world gap is increasing is , I think, the real issue here because its hurting owners wallets over the long term.

- several recent technologies seem much more aimed at good EU CO2 test results than the real world. Greg has highlighted the obvious one - stop/start but to see what may be happening compare two Fiats , same engineers but very diferent outcomes. The Fiat twin air system is clever and it gives the 500 a CO2 number of 95gms/km, on the Autocar test it did 39 mpg "touring" consumption- over a real world route. Now the 1.4 punto abarth also did 39 mpg touring. However its EU test CO2 is 142 gms/km or 50% worse than the 500 twin air . The cars are from the same mfr, so engineering, but ten years apart so it smells strongly of the newer 500 being designed for the test. Certainly the 500 twin air is now notorious for having a huge real wolrd vs test data economy gap. Given most buyers have to go by the CO2/km data that is realy hurting consumers for the 10-15 year life of the car.

Other test related things include dual mass flywheels necesary to allow 4 cylinder diesels to slog at 1,000- 1,500 rpm to maximise diesel efficiency. Nice but the clutches haven't been lasting long and cost 2-3X the price of a conventional clutch - $1,l00 before fitting for an ordinary car.




#17 Magoo

Magoo
  • Member

  • 2,458 posts
  • Joined: October 10

Posted 15 March 2013 - 21:57

Right I skimmed through it. I'm not very convinced by some of the cheating claims, but will start by saying that the graphs A2.3 and A2.4 tell an ugly tale. Fig A2.3 is pretty damning evidence that manipulation is routine, although the AC side of things is huge hit by itself. A2.4 rather bafflingly shows a much lower figure, and less of a trend for the directly comparable NEDC test hot and cold. Note that I'm not disputing that the NEDC doesn't agree with real world, it is a silly test profile and wasn't intended to be used to predict real world fuel consumption. The American one is much better.


Quotes and comments

"The same applies to cars, so a test vehicle will roll much farther when the tyres are pumped up as hard as possible. This practice is not specifically excluded in the test procedure, in spite of the fact that it would be dangerous to drive a car in this condition on a real road."

Actually pumping your tires up above the placard figure is recommended procedure on all advanced driving courses I have been on. Now, they don't give figures for this, I'm guessing they mean 50 psi not 38 psi.

"For example, optimising wheel and tyre specification to increase rolling radius by 5% will change the test result by about 2%"

I'm sure it would. However a 30mm difference in tire diameter will not fit in the wheel arch, typically it will rub at full lock and in jounce.

If changing the wheel alignment inside spec limits alters fuel consumption by 20% I'll eat my hat.

Running in. it seems odd to argue that real world is so important, where the average 'new' car has 15000 km on it, and then whinge that some OEMs run the car in for more than 3000 km

"Clearly a dead flat and dead level test track is the only fair basis on which to conduct a coastdown test. However, procedures only specify that any slope on the track must not exceed 1.5%. Test tracks are often designed with precisely this degree of slope, although it is possible using modern methods to create a test surface that is much more level than this. The test must be completed in twice in opposite directions, but Zallinger and Hausberger (2009) point out that the simple averaging method that is used to combine the two results is technically incorrect and still favours results obtained on a sloping track."

Interesting claim. That one at least seems directly checkable.

Refs

Bizarrely they don't link to the refs. Not very helpful when they are casting aspersions of naughtiness. Here's the ones I found so far

Mock P, German J, Bandivadekar A, and Riemersma I, 2012,

http://www.theicct.o...gpaper_2012.pdf

Smokers R, Kadijk G and Dekker H,

http://www.tno.nl/do...ions_051212.pdf


Good eye on all that.


In the report, I had a problem with the unsupported claim that the contracting test labs are simply bought out by the automakers and turn a blind eye or actually participate in the shenanigans. Hmmm. Really. A charge of that nature requires evidence of some kind.

And there's an easy solution: the govt hires third-party labs to buy cars off the showroom floor for blind testing and sends the bill to the OEs. There, fixed.

But as you know, these areas are small potatoes, the change in the cushions in the test result discrepancies. The significant skews are built right into the cars.

#18 Magoo

Magoo
  • Member

  • 2,458 posts
  • Joined: October 10

Posted 15 March 2013 - 22:13

Drivers, driving, and road/weather conditions are not a factor in the main driving cycle portion of the test procedure. The actual driving loop is negotiated on a chassis dyno aka rolling road rig indoors with an automated driver.



Posted Image









#19 Magoo

Magoo
  • Member

  • 2,458 posts
  • Joined: October 10

Posted 15 March 2013 - 22:31

And yes, anyone can beat the govt/factory fuel consumption figures. Only a modicum of driving skill is required. Simply study the driving cycle map as above and then drive more slowly than that, accelerating and braking more gently and in general, providing smoother inputs to the vehicle. It's not even terribly difficult. All you need are patience and discipline, and no, you won't be blocking traffic like a grandma. Granted, you probably don't want to drive that way, especially you maniacs, but you could if you wanted to.

For example, if you were to alter your route or time your traffic stops to avoid all the big start/stops in the first 800 seconds of the drive cycle, you will beat the stated fuel consumption result by a ton.

Advertisement

#20 bigleagueslider

bigleagueslider
  • Member

  • 842 posts
  • Joined: March 11

Posted 16 March 2013 - 03:04

The government regulations regarding vehicle emissions or fuel consumption are meaningless in reality, since these standards are set arbitrarily. We should also remember that the auto companies have a far greater number of far more capable engineers designing their vehicles to pass these compliance tests, than the government has engineers writing the standards and regulations. Seriously, why does it matter if your new car only gets 25mpg/hwy instead of the published 26mpg/hwy?

And before any of you criticize the auto companies for taking full advantage of every regulatory "loophole", you should understand just how difficult it is to design an auto chassis and drivetrain capable of meeting modern fuel consumption, emissions and safety standards. A task made even more difficult by the incredibly high levels of reliability and performance that modern auto buyers now demand.



#21 Magoo

Magoo
  • Member

  • 2,458 posts
  • Joined: October 10

Posted 16 March 2013 - 10:48

And before any of you criticize the auto companies for taking full advantage of every regulatory "loophole", you should understand just how difficult it is to design an auto chassis and drivetrain capable of meeting modern fuel consumption, emissions and safety standards. A task made even more difficult by the incredibly high levels of reliability and performance that modern auto buyers now demand.


One ethical difficulty is in defining "loophole." That's a very slippery slope. What is a "loophole" and what is fraud? Cheating doesn't allow us to meet standards. Once we cheat, it is now impossible for us to meet the standards. We've given up on that. Now we are working to some other yardstick...of our own design or perhaps no conscious design at all.

In most any commercial project, engineering or otherwise, there are invariably multiple standards and requirements -- regulatory for one, and also the company's goals and objectives for the program. If it is ethically acceptable to "game the system" and seek out loopholes to evade the regulatory requirements, is it then ok to do the same with the company's own benchmarks?


#22 Canuck

Canuck
  • Member

  • 1,642 posts
  • Joined: March 05

Posted 16 March 2013 - 12:41

Wall Street et al do it every minute. Cheating the system by gaming the regs and finding loopholes along with outright cheating of the regs is de rigueur in business today. How else do companies sitting with <$75 billion in cash (literal cash) get off paying $0 in taxes? Sure - what they did was entirely legal but utterly unethical. Car companies cheating the test are only a symptom of a much wider-spread issue with much more pressing problems.

#23 Magoo

Magoo
  • Member

  • 2,458 posts
  • Joined: October 10

Posted 16 March 2013 - 13:25

Wall Street et al do it every minute. Cheating the system by gaming the regs and finding loopholes along with outright cheating of the regs is de rigueur in business today.


I'm not sure how that's relevant here. Are you saying that if Wall St does it, the rest of the business world can do it too?

I'm sure that throughout the business world, on Wall St. or elsewhere, you can find people acting both ethically and unethically. I think I know how I would like my bridges and airliners designed.

#24 Canuck

Canuck
  • Member

  • 1,642 posts
  • Joined: March 05

Posted 16 March 2013 - 20:30

Im not saying if/then at all - nothing of the sort. I'm saying it's endemic and entrenched. It's not okay in any way yet clearly authorities are signing off on the behaviour, ignoring it, or they're plain negligent in their oversight roles.

It's a disturbing trend that seems to grow more and more prevalent (perhaps we're just more aware of it). Laziness and profit-centric management perhaps.

#25 Magoo

Magoo
  • Member

  • 2,458 posts
  • Joined: October 10

Posted 17 March 2013 - 13:01

Im not saying if/then at all - nothing of the sort.


I didn't think so.

#26 Magoo

Magoo
  • Member

  • 2,458 posts
  • Joined: October 10

Posted 17 March 2013 - 15:15

The government regulations regarding vehicle emissions or fuel consumption are meaningless in reality, since these standards are set arbitrarily.


Testing standards usually are to some degree arbitrary. That doesn't make them meaningless or valueless. Depends on the meanings assigned to the results. This weekend a batch of race engines dyno tested at Maranello, Brixworth, etc, under a totally arbitrary test regime were shipped halfway around the world to AU to be raced under another set of operating conditions altogether. And yet the teams had a very accurate idea of the output and fuel consumption to expect.

In the USA, the CAFE/fuel economy data evolved from emissions test protocols in the 1970s, and as relative or comparative results they do a reasonably good job. When the numbers are treated as absolute values -- as in everything over x mpg is good, and everything under is bad -- then problems arise. Not for nothing do we have the universal disclaimer: "Your mileage may vary." We have two huge variables, operator and environment.

Edited by Magoo, 17 March 2013 - 15:17.


#27 BRG

BRG
  • Member

  • 11,489 posts
  • Joined: September 99

Posted 18 March 2013 - 19:26

And there's an easy solution: the govt hires third-party labs to buy cars off the showroom floor for blind testing and sends the bill to the OEs. There, fixed.

If only. But governments everywhere are busily supporting industry rather than protecting consumers.

The recent horsemeat scandal shows just how much European governments have allowed the food industry to get away with murder (poor Dobbin!) for years rather than looking after OUR interests, the consumers - literally in this case. The motor industry is no different.

#28 mariner

mariner
  • Member

  • 1,367 posts
  • Joined: January 07

Posted 19 March 2013 - 08:39

Part of the problem in these things is what gets reported, who funded it and how much poeple read it - the rgood thing is most readers aren't daft or niave.

This appeared in the UK press yesterday

http://www.bbc.co.uk...onment-21803635

If you skim the article you would think - " great" we'll save the planet AND our own money". However its not an official EU report or research but commissioned by a consortium of european parts suppliers , some unions and a climate change lobby group.

The obvious flaw in the claim of saving motoists money by going to 95gms/km vs 130gms is that the fuel price includes 65% taxesand governments will still have to collect those taxes so the real break even is about 7 years , not 3. Thats before any extra repair costs from all the extra technology.

If you skim the comments its clear this flaw is obvious to most people who thought about it much.

The comments include plenty of rants at the testing as people dont see the EU test figures in actual driving. Also the feelings about the long term repair cost of all the extra systems are keenly felt.

So I feel sort of reassured that a significant proportion of Joe Public is smart enough to be skeptical.

#29 meb58

meb58
  • Member

  • 603 posts
  • Joined: May 09

Posted 19 March 2013 - 13:35

Why not determine mpg in this way; allow all of the manufacturers to peform what ever testing they like based upon government regs. - basically status quo. Call the mpg figures from pre-mpg...literally on the sticker on the window.

Once cars hot the show room, the EPA takes a selction of a given model off the lot and drives them in real world - they would reqire a number of consistent/repreatable public routes for these tests. Then, apply 'final' mpg figures to the car based upon the average mpg from this particular series of tests.

In theory, this procedure should expose any funny business...but it also poses a problem...who would buy a car that has not been perfectly certified before the sale? Although I think that this is a potential stumbling block at the point of sale, manufacturers will begin to develop a reputation for either beating mpg numbers or not...knowing that their final certification is real world.



#30 Magoo

Magoo
  • Member

  • 2,458 posts
  • Joined: October 10

Posted 19 March 2013 - 22:39

There is another interesting issue with the Transport & Environment report. As we know, the nub of the matter is "the gap between official tests and real world fuel efficiency." Ok, so how are the so-called "real world" data obtained? What do we think of the two methods and data blocks in the report? Do we have any right to presume these are more rigorous or representative than the lab-derived data?

#31 Magoo

Magoo
  • Member

  • 2,458 posts
  • Joined: October 10

Posted 19 March 2013 - 22:55

The comments include plenty of rants at the testing as people dont see the EU test figures in actual driving. Also the feelings about the long term repair cost of all the extra systems are keenly felt.


The biggest factor by far in folks missing their fuel consumption ratings is their poor (for fuel economy) driving habits.

#32 GreenMachine

GreenMachine
  • Member

  • 763 posts
  • Joined: March 04

Posted 19 March 2013 - 23:03

There is another interesting issue with the Transport & Environment report. As we know, the nub of the matter is "the gap between official tests and real world fuel efficiency." Ok, so how are the so-called "real world" data obtained? What do we think of the two methods and data blocks in the report? Do we have any right to presume these are more rigorous or representative than the lab-derived data?


Real world data is what I get, the rest of you are doing it wrong. Case closed.

This is the problem with so-called 'real world data' - the real world is a dynamic and not repeatable experience that is definitely not homogenous across the population. We might benefit from a simulation that is more transparent to the user (the buyer), and we might benefit from tests that eliminate the most egregious of the fiddles, but what we NEED is a benchmark that is fixed across time and therefore useable as a basis for comparision and information. It doesn't NEED to be 'real' because it is only a benchmark, and people who want it to be 'real' are chasing a chimera.

#33 Robin Fairservice

Robin Fairservice
  • Member

  • 493 posts
  • Joined: March 07

Posted 19 March 2013 - 23:39

I often wonder what fuel specs they use for the testing. Is there a standard for it? What happens if people have to use fuel with 10% Ethanol, like in Canada. My own experience is about a 10% increase in fuel consumption since this started. With the energy required to make Ethanol I believe that it is a tax break racket for Corn producers and Ethanol manufacturers, with no gains for any users or the environment.

#34 gruntguru

gruntguru
  • Member

  • 5,196 posts
  • Joined: January 09

Posted 20 March 2013 - 01:50

What happens if people have to use fuel with 10% Ethanol, like in Canada. My own experience is about a 10% increase in fuel consumption since this started

You are unlucky. Energy content of E10 is about 97% that of Unleaded gasolene and most vehicles don't suffer any more than the 3% difference. E10 (usually) has a higher octane rating and knock-limited vehicles fitted with knock sensing will take advantage of this by advancing the timing.

Nice attempt at a "real world" comparison. http://www.racq.com...._E10_report.pdf

#35 Magoo

Magoo
  • Member

  • 2,458 posts
  • Joined: October 10

Posted 20 March 2013 - 11:40

I often wonder what fuel specs they use for the testing. Is there a standard for it? What happens if people have to use fuel with 10% Ethanol, like in Canada. My own experience is about a 10% increase in fuel consumption since this started. With the energy required to make Ethanol I believe that it is a tax break racket for Corn producers and Ethanol manufacturers, with no gains for any users or the environment.


The fuel specs are as tightly defined as anything else in drive cycle testing. Don't know about Europe but the EPA FTP uses Tier 2 RFG E0 as its baseline reference fuel blend.

E10 burns measurably cleaner than ethanol-free gasoline formulations, so we can't really claim no benefit to the environment.

#36 mariner

mariner
  • Member

  • 1,367 posts
  • Joined: January 07

Posted 20 March 2013 - 12:26

One of the comparators used in that report to get " real world" fuel consumption was fuel card records from Holland (IIRC).

Fuel card records are a pretty good measure of real world fuel economy ( RWFE - my new acronym!) because they accurately record fuel used and milage run. Typicaly they are used not just to record and allocate costs by companies but to look for any expenses cheating i.e excessive fuel usage.

Of course they don't record driving style but , whether its right or wrong , company cars in the UK increasingly have data logging so average speed can also be checked, or you could just ask the driver to push the average speed buton on the dash.

So, I think measuring RWFE by car model is not really too diffiicult. If you assume the same driver uses the car most of the time you could do it by age etc as well.

As an aside the UK is famous for " white van man" the hordes of commeicial vans ( Transits, Sprinters mainly) which are used by delivery companies and tradespeople. There is no exact equivalent in the USA. They are not, currently, subject to any EU CO2 limits but 1/6th of all UK milage is by "white van man" Stds are being developed by the EU but they face the same problem that led to CAFE giving a big incentive to buy an SUV/pick up way back when - how do you measure , or limit, fuel economy in something which may be used to carry heavy loads?

It can be done but to illustrate the problem a long wheellbase Sprinter can , by MB's own admission, do 115 mph simply because its long for good aero and it has to be powerful enough to cary big loads. In reality most are filled with light parcels and travel at 90 mph on motoways!!

#37 Tony Matthews

Tony Matthews
  • Member

  • 17,498 posts
  • Joined: September 08

Posted 20 March 2013 - 12:45

As an aside the UK is famous for " white van man" the hordes of commeicial vans ( Transits, Sprinters mainly) which are used by delivery companies and tradespeople. They are not, currently, subject to any EU CO2 limits but 1/6th of all UK milage is by "white van man"

I am no longer allowed to drive into Central London in my VW Transporter - well, when I say CL, that includes Heathrow, which is hardly central.

#38 meb58

meb58
  • Member

  • 603 posts
  • Joined: May 09

Posted 20 March 2013 - 19:01

But, there has to a be realistic connection to the real world...call it mpgs or potato chips. Hyndai has been accused, I think accurately, of fiddling with their mpg numbers. The actual mileage in many cases is far below the advertised mpg. My VW golf diesel consistently beats the average. How do you reconcile the difference to the buying public? I understand and appreciate your first sentence...but a lot of folks expect a degree of accuracy. Perhaps this just isn't possible...

Real world data is what I get, the rest of you are doing it wrong. Case closed.

This is the problem with so-called 'real world data' - the real world is a dynamic and not repeatable experience that is definitely not homogenous across the population. We might benefit from a simulation that is more transparent to the user (the buyer), and we might benefit from tests that eliminate the most egregious of the fiddles, but what we NEED is a benchmark that is fixed across time and therefore useable as a basis for comparision and information. It doesn't NEED to be 'real' because it is only a benchmark, and people who want it to be 'real' are chasing a chimera.



#39 GreenMachine

GreenMachine
  • Member

  • 763 posts
  • Joined: March 04

Posted 20 March 2013 - 21:55

But, there has to a be realistic connection to the real world...call it mpgs or potato chips. Hyndai has been accused, I think accurately, of fiddling with their mpg numbers. The actual mileage in many cases is far below the advertised mpg. My VW golf diesel consistently beats the average. How do you reconcile the difference to the buying public? I understand and appreciate your first sentence...but a lot of folks expect a degree of accuracy. Perhaps this just isn't possible...


The question is - what is the info used for? If it is to compare new cars, then it (more or less) works, depending on the extent of the gaming by the compared manufacturers.

If it is used to compare what I get now, with what I will get with my new car, it gets more complicated. I can compare what I get with what the test shows for my current car (if it was tested), and then use that to adjust the results for my new car - but this will only work if there are test results for my current car, and the gaming or other fiddles have not significantly altered the results of the old car's test compared to the new car's test.

We won't get rid of the fiddles, but we can expect that their impact is reduced to negligible, by stricter test procedures - use book specs as applicable, car picked at random from production, etc.

Expectations need to be tempered with realism, that we can't always get exactly what we want, and that sometimes near enough is good enough.

Advertisement

#40 Lee Nicolle

Lee Nicolle
  • Member

  • 5,735 posts
  • Joined: July 08

Posted 20 March 2013 - 23:44

I often wonder what fuel specs they use for the testing. Is there a standard for it? What happens if people have to use fuel with 10% Ethanol, like in Canada. My own experience is about a 10% increase in fuel consumption since this started. With the energy required to make Ethanol I believe that it is a tax break racket for Corn producers and Ethanol manufacturers, with no gains for any users or the environment.

As I said earlier some cars seem to like that stuff. And some do not. On long trips I have got better mileage from E10 than normal 91 unleaded, especially when using the 94 octane variation.
Though I do worry about fuel system problems.Methanol [a similar alcohol fuel eats alloy] Though unleaded itself eats these components, kills efi pumps etc so the point is probably irrelevant.
E85 however has to use special pumps, tank coatings, fuel lines like any alcohol fuel. Plus carbs are different, or efi nozzles. On purpose built roadcars they seem to be down on power and economy.
But the stuff does make good power in motorsport applications, especially on turbo engines or high compression engines because of its higher octane levels. I have been tempted but reputedly it is not compatible with fuel tank foam, Nor normal Holley pumps and carbs.

#41 bigleagueslider

bigleagueslider
  • Member

  • 842 posts
  • Joined: March 11

Posted 21 March 2013 - 03:56

The biggest factor by far in folks missing their fuel consumption ratings is their poor (for fuel economy) driving habits.


Magoo- That's a great point. Many of the comments claim there is "cheating" by the auto companies when their vehicles undergo emissions or fuel economy compliance testing. But how is it "cheating" when the regulatory agencies themselves certify that the vehicles comply with their test requirements/procedures?

As for the comments comparing the auto companies approach to their vehicle's emissions or fuel consumption compliance to how US public corporations minimize their annual income tax liability by taking full advantage of the US tax code, this is actually quite a relevant comparison. In both examples, they are simply taking full advantage of the regulations as written. This is not cheating, and if they were doing something in violation of the regulations they would be penalized for doing so.

#42 Magoo

Magoo
  • Member

  • 2,458 posts
  • Joined: October 10

Posted 21 March 2013 - 08:52

As for the comments comparing the auto companies approach to their vehicle's emissions or fuel consumption compliance to how US public corporations minimize their annual income tax liability by taking full advantage of the US tax code, this is actually quite a relevant comparison. In both examples, they are simply taking full advantage of the regulations as written. This is not cheating, and if they were doing something in violation of the regulations they would be penalized for doing so.


In either case, it is naive in the extreme to assume that a company is only "taking full advantage" of the law. For example, in this case we have manufacturers allegedly taping up panel gaps, swapping out lubricants, and disabling alternators and electrical accessories. Just because these practices slip through the enforcement does not make them ethical or legal. Unless one is a moral cripple, anyone who commits these acts has to know they are wrong. There is no gray area here--anyone who sees one is literally deluding himself. It's also extremely bad judgement. This is not an SCCA club race. Millions of dollars and thousands of jobs are at stake. The potential upside is marginal at best, while the potential downside is enormous for the reputation and good will of the company.

EDIT: I can't tell you how repulsed I am by the idea that a practice is legal or proper because we haven't been caught, and then when we do get caught, placing the blame with the regulatory agency for allowing us to get away with it to that point. That's really screwed up.

Edited by Magoo, 21 March 2013 - 08:55.


#43 meb58

meb58
  • Member

  • 603 posts
  • Joined: May 09

Posted 21 March 2013 - 17:08

Indeed!

EDIT: I can't tell you how repulsed I am by the idea that a practice is legal or proper because we haven't been caught, and then when we do get caught, placing the blame with the regulatory agency for allowing us to get away with it to that point. That's really screwed up.
[/quote]


#44 Magoo

Magoo
  • Member

  • 2,458 posts
  • Joined: October 10

Posted 21 March 2013 - 23:31

Let me repeat that I am skeptical about some of the accusations in the report--especially those involving rigging the cars. I would like to see some proof. You don't want to believe people would be that dumb, given the risk/benefit ratio.

#45 Canuck

Canuck
  • Member

  • 1,642 posts
  • Joined: March 05

Posted 22 March 2013 - 01:47

But that's applying logic to human behaviour rather than emotion and ego which are the usual drivers. People are Not logical and repeatedly act against their own best interests for short term reward.

Edited by Canuck, 22 March 2013 - 01:48.


#46 kikiturbo2

kikiturbo2
  • Member

  • 627 posts
  • Joined: December 04

Posted 22 March 2013 - 16:16

I would just like to add that there have been numerous well documented instances of real world cars posting better than stated fuel consumption figures, so I do not buy that "manufacturers rigging cars" idea at all..

What manufacturers do is that they do tend to optimize the cars for the official tests and I do not mean the individual cars to be tested, but rather the whole production..
In reality, the people bit$%&ng about fuel consumption tend not to know how to drive economically..

#47 malbear

malbear
  • Member

  • 242 posts
  • Joined: September 02

Posted 22 March 2013 - 20:31

I would just like to add that there have been numerous well documented instances of real world cars posting better than stated fuel consumption figures, so I do not buy that "manufacturers rigging cars" idea at all..

What manufacturers do is that they do tend to optimize the cars for the official tests and I do not mean the individual cars to be tested, but rather the whole production..
In reality, the people bit$%&ng about fuel consumption tend not to know how to drive economically..

ecenomic driving just keep the engine in the 2000 to 3000 rev range in every gear , be light on the brakes , use anticipation lots, flow driving without sudden changes.
fun normal everyday driving :rotfl: 4000 to 5000 in every gear, late hard braking lots of horne .
outrite HOON 7000 to limmiter or valve bounce in every gear tire squeel braking and using the next cars bumper to slow down. pass enerything except the fuel station? :clap:

#48 Greg Locock

Greg Locock
  • Member

  • 4,490 posts
  • Joined: March 03

Posted 22 March 2013 - 21:02

Thinking about it they don't seem to have documented cases of rigging cars for the coastdown test, the language seems to be more like 'if they got a 5% reduction in rolling resistance by changing tire diameter then it would reduce the co2 by this much', and the same sort of thing for aero.

Having said that the increasing discrepancy between real world and EEC results in that report is (to me) pretty convincing evidence that manufacturers are developing vehicles to pass the test, as a matter of priority over real world fuel consumption. And I'm sure one method used is that they meet the letter, not the spirit, of the test procedure.

Incidentally, I suspect very few cars are tested by independent labs, most will be done in-house, some at a competitor's facility.

Despite all the doom and gloom, my new Falcon has ecoboost, goes like stink thanks to a seamless auto box and 170 kW, and gets 8.0 l/100 km (29 mpg) so far on E10, something like a 30% improvement over the last time I had a Falcon in the early 90s. Or to put it another way, a big fat 5 seater uses about 20% more fuel than a Fiesta.



#49 GreenMachine

GreenMachine
  • Member

  • 763 posts
  • Joined: March 04

Posted 22 March 2013 - 21:48

... thanks to a seamless auto box and ...


Really? There was a thread about these here a while back ...  ;)

#50 bigleagueslider

bigleagueslider
  • Member

  • 842 posts
  • Joined: March 11

Posted 23 March 2013 - 02:02

......I can't tell you how repulsed I am by the idea that a practice is legal or proper because we haven't been caught, and then when we do get caught, placing the blame with the regulatory agency for allowing us to get away with it to that point. That's really screwed up.....


Magoo- I did not state that intentional acts in violation of laws or regulations are legal, proper, ethical or moral, in any circumstance. What I said was that I have no issue with auto companies taking full advantage of the regulations as written to get their product designs certified.

I am also a bit puzzled as to why slight variations in fuel consumption versus the published numbers seem to be such a serious issue with some posters. Do they seriously feel it is a criminal offense if their new Toyota Corolla only gets 35mpg hwy instead of the claimed 36mpg hwy? Auto fuel economy and emissions standards are completely arbitrary numbers set by government bureaucrats. And just how is anyone seriously harmed if their new car's fuel economy is 3% lower than the published value?

In the bigger picture, my personal standard of living is impacted far more by mismanaged/misguided government social programs that end up costing me thousands of dollars in higher taxes each year, than any small discrepancies in CAFE compliance due to "cheating" by the auto OEMs.