Mythbusters: Is a bike greener than a car?

[jamesqf]

[QUOTE=sendler;385681]

Originally Posted by jamesqf

Yes, me for one. QUOTE]

No fuel log either?

No. Why, when the car itself keeps a perfectly good record of mpg? Which is accurate to within limits of gas pump & tire wear affecting the odometer, 'cause I checked.

PS: Should perhaps add that these days I get most of my gas from the cheap stations where I just stick a $20 bill in the reader (about 2/3 of a tank), so I never bother keeping records, let alone bothering to input them here.

[sendler]

My computer always reads about 5% high compared to the tank average. You guys really think the modern gas pumps are only accurate to +-2%?

[IamIan]

Originally Posted by sendler

My computer always reads about 5% high compared to the tank average. You guys really think the modern gas pumps are only accurate to +-2%?

Do you have testing data that shows pumps are better than that?

Not just what it takes to pass ... which varies state to state ... some hold to less than 1% +/- and inspect every year ... others are not nearly as strict.

But also the rate of failing those tests ... pumps failing to pass inspection is not all that rare ... more in some places than others.

What method do you use to determine 5% ? ... what level of accuracy / significant figures , did you have for the distance measurement and the volume of fuel measurement ?

[sendler]

Wiki says US pumps are certified to .3%
.
In the United States, the National Institute of Standards and Technology (NIST) specifies the accuracy of the measurements in Handbook 44. Table 3.30 specifies the accuracy at 0.3% meaning that a 10-US-gallon (37.9 L; 8.3 imp gal) purchase could vary between 9.97 US gal (37.7 L; 8.3 imp gal) and 10.03 US gal (38.0 L; 8.4 imp gal) as to the actual amounts at the delivery temperature of the gasoline.
.
Why would you be so surprised to find that the computer read out in the Insight would be wrong? My computer says 58 mpg when the tank fill says 55.5. It is just a computation based on samples of engine load via the MAP sensor. Not nearly as sophisticated as an MPGuino which counts injection on time. But then that has to be calibrated by comparing to... actual tank fills.

[IamIan]

Originally Posted by sendler

Wiki says US pumps are certified to .3%

I didn't find that in a search ... can you point me to it , or the source of the claim?

Originally Posted by sendler

In the United States, the National Institute of Standards and Technology (NIST) specifies the accuracy of the measurements

In the U.S. ... NIST can specify anything they like ... but they don't hold the authority ... in the U.S. each of the 50 states have their own Department of
weights and measures ... and each of those have their own standards for testing methods ... enforcement policies ... etc... some are more precise and more strict than others... some have more failure rates than others.

Originally Posted by sendler

Why would you be so surprised to find that the computer read out in the Insight would be wrong?

I'm not ... As I already wrote previously in Post #77.

I think all measurements have a +/- margin or error.

And even if I assumed your 5% reported error rate was itself 100% accurate ... the ~76mpg I've had for the last 1,400+ Miles still gives me more than 72mpg ... which as I wrote previously ... even with dash margin for error , I'm still getting ~70mpg or better without needing to drive "pretty slow"... jamesqf's 71.4 over 120k miles would still be about ~68 MPG .. etc.

Originally Posted by sendler

My computer says 58 mpg when the tank fill says 55.5. It is just a computation based on samples of engine load via the MAP sensor. Not nearly as sophisticated as an MPGuino which counts injection on time. But then that has to be calibrated by comparing to... actual tank fills.

But the questions I asked in post #93 remains.

What was the method for that 55.5?

What level of +/- error margin was the distance measurement?
What level of +/- error margin was the Gallon measurement?

How much of tire wear is included? ... how much volume change with temperature is included? ... etc ... etc.


- - - - - -

Kansas for example reported they gas pumps in 2006-2007 was :
"the compliance rate for gas pump accuracy was 96 percent"

that's +/- 4%

Link

[sendler]

Fuel dispenser - Wikipedia, the free encyclopedia
.
What is your average speed for your 70+ mpg runs? Saying +- 5 mph speed limit doesn't tell us anything. You must not be on the 65mph highway much.

[jamesqf]

Originally Posted by sendler

Why would you be so surprised to find that the computer read out in the Insight would be wrong? My computer says 58 mpg when the tank fill says 55.5. It is just a computation based on samples of engine load via the MAP sensor.

Why do you think that? That's pretty much the way the (quite inaccurate) ScanGauge does the computation, but why would Honda engineers not use the much more accurate injector time/volume method?

Further, as I said before, I checked the accuracy of the Insight's fuel consumption gauge back when I first got it. There is a measured 5 mile stretch of flat highway near here, with markers set up specifically so drivers can check speedometer & odometer accuracy. My Insight was dead on - less than 1/10 mile off over the stretch. (It also matches map distances over ~250 mile trips, etc.) I recorded fillups over several months, compared mpg computed from gallons & miles with what the gauge said, and they were the same to within the 0.2-0.3 mpg accuracy of the gauge's display.

Further, as IamIan points out, even if the gauge is only accurate to 5%, that's not a heck of a lot of difference.

[IamIan]

Originally Posted by sendler

What is your average speed for your 70+ mpg runs? Saying +- 5 mph speed limit doesn't tell us anything. You must not be on the 65mph highway much.

Speed changes over the route ... Being within +/- 5MPH of posted limit tells us exactly what is needed to be known in relation to the claim of "pretty slow".

No matter what the speed limit is ... me staying within +/- 5 MPH from that posted speed limit ... is not "Pretty Slow".

The only way around that "pretty slow" claim is if you claim that only doing 5mph faster than the posted speed limit is itself "pretty slow"

My route contains a mix of many conditions ... suburbia ~35mph , highway ~65mph , bumper to bumper ~5mph , etc... where it is posted for 35 , I stay +/-5 from that ... where it is posted 55 I stay +/- 5 of that ... etc... posted speed limit is not , "pretty slow"

Originally Posted by sendler

Fuel dispenser - Wikipedia, the free encyclopedia

NIST as posted before is not the regulatory agency ... they are not the ones who do the testing ... not the ones who set the standards to be meet ... not the ones who issue fines ... etc... their standards are not relevant.

The regulatory Authority is under the department of weights and measures ... each state of the 50 states in the U.S. has their own department of weights and measures.

- - - - - - -

Here's another example of some of the accuracy issue from gas pumps.

Link

Arizona May 2007 Had
6 pumps found off up to 200 Cubic Inches out of 231 Cubic Inches in a Gallon
~86% error

6 other up to 32 Cubic Inches off ( ~13% Error )

8 others up to 17 Cubic Inches off ( ~7% error )

(Arizona had at the time a state requirement to be within 6 cubic Inches)
~2.6% error ( if the pump was 2.5% off they passed and got a sticker no fine )

more interesting is that at that time Arizona had a upper end cap limit on fines ... $5,000 per pump ... $50,000 per owner.

So if a owner owns many gas stations ... he can only be fined a maximum of $50,000 per year ... no matter how often his pumps are wrong , or how often they are wrong , or by how much.

[jamesqf]

Originally Posted by sendler

What is your average speed for your 70+ mpg runs? Saying +- 5 mph speed limit doesn't tell us anything. You must not be on the 65mph highway much.

In my case, the 70+ mpg is not "runs", it's the total over 120K miles since I bought the car. It's true that I don't do all that much freeway driving (bar a couple of years when I was doing I5/I580/I680 into San Jose every other week or so. It does include a lot of mountain driving, going over 8000+ ft passes in snowstorms, fording creeks on dirt roads in the Toiyabe Mountains, and so on.

[sendler]

Originally Posted by IamIan

Speed changes over the route ... Being within +/- 5MPH of posted limit tells us exactly what is needed to be known in relation to the claim of "pretty slow".

Saying that you are running the speed limit doesn't tell us anything. 50 mph is pretty slow regardless of what the speed limit is compared to interstate highway driving.
.
That is why I was wondering about what your average speed was in order to get results that are so far above the average for the Gen1 Insight which is about 60 mpgUS. My car came from Colorado with 65,000 miles on it and the first owner had a lifetime of 56 mpg. I always state my mileage as "on my 40 mile, 80% highway commute at 65 mph" so people have an idea of what the conditions are. I hypermile religiously and always follow a big truck to get my average up to 66mpg in the Insight. I can hit 43 with my Honda Fit. My CBR250R is up to 95 mpgUS on the same commute. My PCX150 has to stay on the back highways instead of the interstate and gets 102 mpgUS with 80% of the trip at 52 mph with zero drafting.