Study: Some car models consuming around 50% more fuel than official results • Hacker News Board

Re: Study: Some car models consuming around 50% more fuel than official results

#251

Post by jlarocco » Wed, Sep 30, 2015, 7:09 PM UTC

bonzini wrote: post #118

vkjv wrote: post #86

Earlier quoted context omitted.

I thought the same. In my anecdotal experience, drivers tend to get much lower than the manufacturer mileage because they optimize for things other than fuel economy (e.g., speed). My dad taught me lots of tips on conserving fuel and thankfully I've exceeded the sticker numbers on every car I've owned.

No matter how careful I am, it's hard for me to get lower than 130% of the fuel consumption reported by the manufacturer. Perhaps I can do 120% of the reported consumption at 90 km/h, but I can hardly get lower than 140% on the highway (130 km/h) and in city traffic.

My Subaru Outback has been a nice surprise in this area.

The combined highway/city mileage on the sticker was 24 mpg, and I've been getting 27 or 28 mpg for the entire 2 years I've owned it. For a while I was commuting between Denver and Boulder (about 45 miles each way) five days a week and taking it up to the mountains for hiking or skiing every weekend, so it's been over a wide variety of terrain.

I'm using the numbers on the in-car, computer, though, so maybe I should double check...

Re: Study: Some car models consuming around 50% more fuel than official results

#252

Post by aianus » Wed, Sep 30, 2015, 7:16 PM UTC

mturmon wrote: post #249

aianus wrote: post #210

Earlier quoted context omitted.

> Modern cars do better in rolling hills than on perfectly flat land, since they can cut fuel entirely on the downhill This is physically impossible. Even if the car was perfectly efficient at overcoming the force of gravity on the uphill portions (which it is not), you cannot do better than flat terrain (unless by hilly you mean 'net downhill').

I think you should reconsider "physically impossible". Sure, in some idealized situations, the statement you're responding to would be wrong, grievously wrong. But the actual situation has two novel features: (1) the engine uses up power just by being on; (2) you must maintain a minimum speed in order not to block traffic. If you put these constraints in, you can see that it's possible for a pulsed-acceleration strat…

If you're maintaining a minimum speed in order to not block traffic, you would have to downshift to climb the hill which would kill your efficiency.

But you raise good points, maybe I jumped the gun on 'physically impossible'.

Re: Study: Some car models consuming around 50% more fuel than official results

#253

Post by TylerE » Wed, Sep 30, 2015, 7:20 PM UTC

aianus wrote: post #252

mturmon wrote: post #249

Earlier quoted context omitted.

I think you should reconsider "physically impossible". Sure, in some idealized situations, the statement you're responding to would be wrong, grievously wrong. But the actual situation has two novel features: (1) the engine uses up power just by being on; (2) you must maintain a minimum speed in order not to block traffic. If you put these constraints in, you can see that it's possible for a pulsed-acceleration strat…

If you're maintaining a minimum speed in order to not block traffic, you would have to downshift to climb the hill which would kill your efficiency. But you raise good points, maybe I jumped the gun on 'physically impossible'.

Not at all, in many vehicles. 6th gear in my car is quite long (Turns 1950RPM at 75MPH) and I have no problem maintaining speed on highway grades.

Even if you do downshift that doesn't necessarily mean you're being less efficient either. Being a bit higher up the rev range may actually put you in a better spot on the efficiency curve for a given power output.

http://ecomodder.com/wiki/index.php/Brake_Specific_Fuel_Cons...

Re: Study: Some car models consuming around 50% more fuel than official results

#254

Post by henningo » Wed, Sep 30, 2015, 7:23 PM UTC

ucsdrake wrote: post #237

henningo wrote: post #221

I think it is a bit unfair to just blame the manufacturers. The legislation could likely be changed to make for a more robust test approach. After all, it is engineers we are talking about here, if they are given a task to optimize an engine's emissions and fuel consumption to a specific drive cycle they will do it. But then of course you have the ethical aspects which I won't go into. A computer science analogy to t…

"A computer science analogy to this would be if I gave you an uncompressed image and you had to develop a compression algorithm that made the image as small as possible, you could likely come up with a really good solution. But your algorithm most likely wouldn't do as well on any other image." That analogy doesn't hold. It's more akin to developing two compression algorithms, one for the general case, and a specific…

I agree that the analogy doesn't hold for the recent VW debacle (where the calibration was changed during certification testing), but it holds for the industry in large and what has been going on for the past 10-20 years, which is what the paper is about.

The vehicle manufacturers optimize the engine calibration to the drive cycle they are trying to beat. That is why a US-spec BMW has a different engine tune than a Euro-spec BMW for example, the drive cycles are different.

Re: Study: Some car models consuming around 50% more fuel than official results

#255

Post by abalone » Wed, Sep 30, 2015, 7:25 PM UTC

vkjv wrote: post #86

mattlutze wrote: post #58

Earlier quoted context omitted.

The source of the data is actually www.spritmonitor.de, which is a self-reporting mileage tracking site (like www.fuelly.com). With ~501k vehicles tracked and 8.885B kilometers tracked, the site represents ~17k kilometers per car of user-reported mileage information. Is this the best, or a reliable source, of information to start the reckoning/witch hunt to come? I'm not sure.

I thought the same. In my anecdotal experience, drivers tend to get much lower than the manufacturer mileage because they optimize for things other than fuel economy (e.g., speed). My dad taught me lots of tips on conserving fuel and thankfully I've exceeded the sticker numbers on every car I've owned.

Except some models are significantly more off the mark than others. Maybe some owner populations have worse driving habits than others, but it's worth investigating.

Re: Study: Some car models consuming around 50% more fuel than official results

#256

Post by Karunamon » Wed, Sep 30, 2015, 7:27 PM UTC

kbenson wrote: post #214

Karunamon wrote: post #178

Earlier quoted context omitted.

The execs get all the credit when it goes right, they should get all the blame when it goes wrong. Lock a couple of these people up, and suddenly they'll start caring a lot more about the legality of what their underlings are doing...

You can't just lock up the execs without evidence, that's unjust. You can't just investigate, the nature of large power structures is that some underling will take the fall. Breaking the rules for economic advantage should be met with new rules to force economic disadvantage. Put a 2% tax on total revenue for the next X years, possibly with the addition of a fee based on perceived past benefit. Honest players in the…

The evidence is that X underling, which Y exec is directly responsible for the business conduct of, demonstrably broke the law in Z way. Vicarious liability I beleive is the term I'm looking for.

Again - with taking the credit comes taking the responsibility. I'd like to see the law amended as such.

Re: Study: Some car models consuming around 50% more fuel than official results

#257

Post by takeda » Wed, Sep 30, 2015, 7:38 PM UTC

axx wrote: post #43

eterm wrote: post #20

Earlier quoted context omitted.

You are obfuscating the issue here. The issue isn't that we can't "represent real street results", the issue is that manufacturers took deliberate steps to alter car behaviour with the intention of obtaining better test results than would be present outside the tests.

And that's a surprise to whom? It's the same with smartphone and notebook batteries. Tell me, when was the last time, you actually got 9 hours of battery live (in case of a MBP)? Don't get me wrong, i see it the same way as you are, but this is an issue that you can find everywhere not just in car manufacturing. For example, a lot of power supply companies try all kinds of tricks to game the efficiency test systems.

Perhaps you did not read the article and responding to the title.

The article points out that 1/3 of improvement in past 3 years to meet regulation requirements was real improvement and 2/3 was just manipulation.

It doesn't matter that people drive cars differently than they are tested in cars, to illustrate it they highlighted that due to those manipulations that difference now can be as high as 50% which is beyond claims that people drive cars differently than when they are tested in labs.

Re: Study: Some car models consuming around 50% more fuel than official results

#258

Post by rebootthesystem » Wed, Sep 30, 2015, 7:40 PM UTC

mikeash wrote: post #206

rebootthesystem wrote: post #98

It's called "real world driving". Look, this is no different from Tesla claiming a 280 miles range that nobody can achieve. Not saying it is right. What I am saying is: We need better tests.

It's not actually very hard to achieve rated range in a Tesla. You might need to refrain from flooring it at every opportunity and you might want to cut down from going 10MPH over the speed limit, but it's not a challenge.

I think you missed my point: real world driving.

No vehicle meets manufacturer specifications in this regard, electric or IC.

To be clear, I am not saying this as a negative. I am simply stating it as a fact.

Re: Study: Some car models consuming around 50% more fuel than official results

#259

Post by kbenson » Wed, Sep 30, 2015, 7:42 PM UTC

rnovak wrote: post #247

kbenson wrote: post #235

Earlier quoted context omitted.

It's irrelevant because unless both a) people changed their driving habits on average over that time frame to cause more emissions and b) that change was primarily limited to drivers of GM vehicles, then driver habits do nothing to explain the data. > So drawing the conclusion that manufacturers are "lying" from the discrepancies between those two data sets is ridiculous. Perhaps you should pay attention to actually…

The whole controversy is over manufacturers reporting false numbers, i.e. lying. Saying a manufacturer deserves extra scrutiny can absolutely be inferred to be an accusation of them reporting false numbers, i.e. lying. My reading comprehension isn't at fault here. Secondly, it's still irrelevant. You would have to prove that an improvement at throttle position 0.10 over 5 years equates directly to an improvement at t…

> Saying a manufacturer deserves extra scrutiny can absolutely be inferred to be an accusation of them reporting false numbers, i.e. lying.

> My reading comprehension isn't at fault here

Except that I very clearly said "I can't discount the possibility of something else is going on, such as an influx of a different type of car for GM over that period, I haven't read yet how they normalized the study and accounted for things like that."

> Maybe before accusing people of "not paying attention", you should look at your own language and see if it's ambiguous?

You ignored clear text from me saying that it didn't necessarily indicate GM was at fault. What am I supposed to infer from that?

As to the matter at hand, you still seem to be under the impression that driver habits can account for a change in the numbers we are talking about, when that would mean that GM drivers specifically drive very differently than all other car drivers in the report. We are talking about number across the industry, categorized by manufacturer. Everything listed on the graph in question shows automakers overreporting gains, which could be driver habits. GM specifically shows vastly different numbers from event every other manufacturer, not just the reported numbers. I fail to see how driver habit accounts for a specific manufacturer being so out of bounds with it's own reported numbers (keep in mind engines are often shared across multiple manufacturers).

Have you looked at the graph on page 13 yet?

Re: Study: Some car models consuming around 50% more fuel than official results

#260

Post by mikeash » Wed, Sep 30, 2015, 7:45 PM UTC

rebootthesystem wrote: post #258

mikeash wrote: post #206

Earlier quoted context omitted.

It's not actually very hard to achieve rated range in a Tesla. You might need to refrain from flooring it at every opportunity and you might want to cut down from going 10MPH over the speed limit, but it's not a challenge.

I think you missed my point: real world driving. No vehicle meets manufacturer specifications in this regard, electric or IC. To be clear, I am not saying this as a negative. I am simply stating it as a fact.

You missed my point, which is that it's easy to achieve rated range in a Tesla with real-world driving.

I have not taken care at all to drive my Model S in an efficient manner, and I still have an average of 315Wh/mile over about 8,000 miles of driving so far, where 300Wh/mile is rated efficiency.