On 07/21/2017 6:30 PM, Mad Roger wrote:
....

Am I correct to understand that you are saying if you go only 300 miles on
one tank, then the fill-level inaccuracy is (say) plus or minus 1 gallon
per tank; but if you go 3,000 miles (obviously on multiple tanks), that the
fill-level inaccuracy is one tenth of that plus or minus one gallon per
tank?

No, but the 1 gal error in the ~10X gallons _TOTAL_ over the 3000 mi is
only 1/10th of the (approx) tenth of that on each individual tank.

And it's only the total that matters in the overall average mileage...it
didn't matter on the intermediary tanks at all; they could be top-offs,
half-fills, full-fills; totally immaterial. All that matters is the
consistency in that last fill being to the same level as the first and
that one discrepancy is the only "error" that exists in that portion of
the total process...all the rest cancel identically.

--

On Sat, 22 Jul 2017 11:43:15 -0600,
rbowman wrote:

It's a given that the speedometers on Japanese bikes are 10% high. The
theory is the manufacturers have no control over what tires will be
fitted in the future and are covering themselves over any legal
problems. "But the speedometer said I was doing 65..." On either of my
Suzukis that means I'm doing 60 and well under the posted speed limit.

The Harley speedometer is accurate. Obviously if the Japanese wanted
accurate speedometers they could do it.

The main point for bringing up the otherwise unrelated issues of
speedometer accuracy and repeatability is that some people here seemed to
think if they ran a calculation more times, that the "accuracy" somehow
(magically?) gets better because (magically?) their reading is low as much
as it is high.

Fact is...
The accuracy might get better.
It might not get better.

In the case of the speedometer, it will never get better because, often,
they always read high (on a given set of wheels and gears).

So just assuming that the gauge reads randomly below the actual speed as
much as it reads randomly high above the actual speed isn't going to get
better accuracy even if you run a billion speedometer runs.

It will always be wrong (by a certain amount high).

On Sat, 22 Jul 2017 00:46:50 -0400,
rickman wrote:

So my odometer is accurate and precise.

I understand you because you're exactly the type of person that I had in
mind when I asked the question in the first place.

I don't know what you mean. I have checked my odometer against the markers
on the highway as well as against my GPS (I think the highway markers are
more accurate than the GPS). It is spot on with the current tires to 1% or
better.

Does your tripmeter have a decimal place and digits after that decimal
place?

My speedometer is mechanical and so has a separate calibration factor.

The speedometer example was only brought in to point out that the vain hope
that averages result in better "accuracy" is patently false.

Mom-and-pop type of people actually believe that a speedometer reads even
close to accurately - and worse - some here propose the vain notion that
the more readings they take, somehow (magically?) the more accurate the
results will be.

A speedometer that reads high isn't going to result in more accurate
calculations even if you do a billion test runs.

+ A pumpmeter of 20.25 gallons is likely relatively accurate & precise

Of course it is. States inspect them at some point.

You don't seem to understand what accuracy and precision even mean.
Haven't you taken even one science lab course?

+ Matching fuel level in the tank isn't even close to accurate nor precise

I don't agree. I let the pump click off and then continue to pump for a
number of more clicks until it cuts off immediately.

I'm not at all surprised about your concept of the fuel-level estimation,
and, in fact, you're exactly the mom-and-pop type person I was talking
about when I opened the thread.

I understand you.

I always need to run
at least another fifteen miles before I am home so that is better part of a
gallon burned so I don't need to worry about the gas warming up and running
out of the tank. I believe this makes for very consistent fill ups.

I'm sure you do believe that.

My MPG results pretty well show the consistency of my measures.

I'm sure your MPG results support any theory you want them to support.
I believe you.

You know what happens when you assume...

You don't know how funny that statement was to me when I just read it now.

I see less than 19 or even 19.5 MPG.

I bet you see that decimal place even though it's not in the tripmeter
estimation nor in the filllevel estimation.

You see, I understand you because you're the type of person I had in mind
when I asked the question.

I think the consistency of my MPG readings show how well each of these can
be measured.

I'm sure you do.

As you say, the pump is going to be dead on.

Whoa! I never said the pump was "dead on" and anyone reading this thread
who thinks I think the pump is "dead on" would have completely
misunderstood everything else I said.

All I said was that the inaccuracies and imprecisions in the pump reading
are likely better than the otherwise astoundingly huge imprecision in the
fuel-fill level estimation and in the lesser inaccuracy of the tripmeter
estimation.

Other than scale
error which can be calibrated out the odometer will be very good.

Define "very good" please.

Filling your tank can be good as well.

I'm sure you believe that filling the tank is "accurate" since you
calculate 19.5 miles per gallon and not something like 19.5 rounded up to
20 and then the error taken into account such that it's more likely
anywhere between 19 and 21 mpg than it is 19.5 mpg.

It's not like they design gas tanks to have air pockets.

Actually, they do have air pockets.
Those air pockets change in size based on temperature & pressure & fill
level.

Even the fuel changes in density based on those parameters.

You don't need to know any of this specifically.

Of course I don't. 19.5 mpg is all I need to know.
And if I change "something" which results in 19.7mpg, then of course, that
something was the cause. I understand. I really do.

Why do you care which of the three has what specific degrees of accuracy and
precision?

I care because when I do a calculation, my assumption is that 19.5mpg is
actually something closer to 19 to 21 mpg than it is to 19.5.

If the "change" I'm measuring is within that margin of error, then I can't
say anything about what that "change" was.

And, more importantly, neither can you.
Which is the entire point after all.

On 07/21/2017 8:30 PM, dpb wrote:
On 07/21/2017 7:02 PM, dpb wrote:
...

+ How accurate& precise is a reading of 20.25 gallons on a gas pump?

NIST tolerance is 6 in^3 in a 5 gal measure. AFAIK that's what all state
W&M departments use for their tolerance. A NIST document of 20,000
tested meters showed 0-mean normally distributed discrepancies at about
90% bounds on the +/-6 number. The 6/5gal -- ~0.5%

....

And remember that is the "shut 'er down" tolerance, not the average...as
noted, the most probable based on the NIST sample was in the +/-0 bin (1).

I didn't quite recognize what the figure was yet when first looked at it
and had closed the link when I realized the significance so don't have
the actual numbers at hand...but the +/-6 number was quite a way out on
the tails of the distribution altho I don't know just precisely the
tails percentages.

And, actually while the report used "normal" in discussing the
distribution, it really wasn't normal as in bell-shaped, it was
symmetric and zero-mean, but the tail in each direction dropped off more
as hyperbolic than a normal--hence the tail percentages would actually
by somewhat lower than a real normal of same mean, standard deviation.


I got curious myself on what the numbers revealed and looked at the NIST
numbers again.

I computed an empirical cdf and compared it to normal...statistics from
the 20,036 observations are below:

[h,s]=cdfplot(x);
s
s =
min: -50
max: 146
mean: -0.0788
std: 3.7681
median: 0
mode: 0


I then compared to normal on the same plot and as outlined above
N(mean,std) is too long-tailed on both ends in comparison. It turns out
that N(mean,std/1.5) is pretty close on both tails to about the +/- 6 point.


Anyway, from the above it's simple enough to get some pretty good
estimates of what pump volume errors one might expect...the table below
is from the empirical cdf NIST data...

P error(in^3)/5Gal error(%)
0.001 -22 -1.82
0.005 -9 -0.78
0.010 -8 -0.69
0.025 -6 -0.52
0.050 -5 -0.43
0.250 -2 -0.17
0.500 0 0
0.750 2 0.17
0.900 4 0.34
0.950 5 0.43
0.975 6 0.52
0.990 7 0.60
0.995 10 0.86
0.999 22 1.82

From the above, one can conclude the pump metering error small for all
except the extreme outlier pumps.

--

On 07/22/2017 6:42 PM, Mad Roger wrote:
....

The main point for bringing up the otherwise unrelated issues of
speedometer accuracy and repeatability is that some people here seemed to
think if they ran a calculation more times, that the "accuracy" somehow
(magically?) gets better because (magically?) their reading is low as much
as it is high.
....

NO! That is not at all what any of those people said. You're simply
repeating the same contention which is ok as far as it goes in that a
point estimate is not the same as averaging or using other techniques to
increase the precision of the measurement.

They (and I in particular) only point out that averaging muddles out the
differences in intermediate filling levels.

The alternative measurement increases the denominator in the computation
at the expense of no additional error in the measurement of the quantity
used; the errors in the intermediary quantities in levels cancel
identically because all the fuel gets used and the final level
discrepancy is only the one but is is now related to the total quantity
instead of the single.

There has been nobody I've seen who's claimed a bias error will be
anything but that -- but in this case that one is a simple calibration
and correction that will not add appreciable error if made. And even if
not won't have any bearing on differences in performance over various
driving conditions for the same vehicle in terms of seeing changes in
that vehicle's relative performance. That it is off by whatever that
percentage in mileage is from the actual is too obvious to belabor and
nobody here has made any claim to the contrary.

--

On Sat, 22 Jul 2017 23:42:24 -0000 (UTC), Mad Roger
wrote:

On Sat, 22 Jul 2017 11:43:15 -0600,
rbowman wrote:

It's a given that the speedometers on Japanese bikes are 10% high. The
theory is the manufacturers have no control over what tires will be
fitted in the future and are covering themselves over any legal
problems. "But the speedometer said I was doing 65..." On either of my
Suzukis that means I'm doing 60 and well under the posted speed limit.

The Harley speedometer is accurate. Obviously if the Japanese wanted
accurate speedometers they could do it.

The main point for bringing up the otherwise unrelated issues of
speedometer accuracy and repeatability is that some people here seemed to
think if they ran a calculation more times, that the "accuracy" somehow
(magically?) gets better because (magically?) their reading is low as much
as it is high.

Fact is...
The accuracy might get better.
It might not get better.

In the case of the speedometer, it will never get better because, often,
they always read high (on a given set of wheels and gears).

So just assuming that the gauge reads randomly below the actual speed as
much as it reads randomly high above the actual speed isn't going to get
better accuracy even if you run a billion speedometer runs.

It will always be wrong (by a certain amount high).
and exactly how is that germaine to the issue at hand? The speedo
means NOTHING. All we care about is the ODO - which will ALWAYS be
consistent, even in inaccuracy - so can be easily compensayed

On Sat, 22 Jul 2017 23:42:25 -0000 (UTC), Mad Roger
wrote:

On Sat, 22 Jul 2017 00:46:50 -0400,
rickman wrote:

So my odometer is accurate and precise.

I understand you because you're exactly the type of person that I had in
mind when I asked the question in the first place.

I don't know what you mean. I have checked my odometer against the markers
on the highway as well as against my GPS (I think the highway markers are
more accurate than the GPS). It is spot on with the current tires to 1% or
better.

Does your tripmeter have a decimal place and digits after that decimal
place?

My speedometer is mechanical and so has a separate calibration factor.

The speedometer example was only brought in to point out that the vain hope
that averages result in better "accuracy" is patently false.

Mom-and-pop type of people actually believe that a speedometer reads even
close to accurately - and worse - some here propose the vain notion that
the more readings they take, somehow (magically?) the more accurate the
results will be.

A speedometer that reads high isn't going to result in more accurate
calculations even if you do a billion test runs.

+ A pumpmeter of 20.25 gallons is likely relatively accurate & precise

Of course it is. States inspect them at some point.

You don't seem to understand what accuracy and precision even mean.
Haven't you taken even one science lab course?

+ Matching fuel level in the tank isn't even close to accurate nor precise

I don't agree. I let the pump click off and then continue to pump for a
number of more clicks until it cuts off immediately.

I'm not at all surprised about your concept of the fuel-level estimation,
and, in fact, you're exactly the mom-and-pop type person I was talking
about when I opened the thread.

I understand you.

I always need to run
at least another fifteen miles before I am home so that is better part of a
gallon burned so I don't need to worry about the gas warming up and running
out of the tank. I believe this makes for very consistent fill ups.

I'm sure you do believe that.

My MPG results pretty well show the consistency of my measures.

I'm sure your MPG results support any theory you want them to support.
I believe you.

You know what happens when you assume...

You don't know how funny that statement was to me when I just read it now.

I see less than 19 or even 19.5 MPG.

I bet you see that decimal place even though it's not in the tripmeter
estimation nor in the filllevel estimation.

You see, I understand you because you're the type of person I had in mind
when I asked the question.

I think the consistency of my MPG readings show how well each of these can
be measured.

I'm sure you do.

As you say, the pump is going to be dead on.

Whoa! I never said the pump was "dead on" and anyone reading this thread
who thinks I think the pump is "dead on" would have completely
misunderstood everything else I said.

All I said was that the inaccuracies and imprecisions in the pump reading
are likely better than the otherwise astoundingly huge imprecision in the
fuel-fill level estimation and in the lesser inaccuracy of the tripmeter
estimation.

Other than scale
error which can be calibrated out the odometer will be very good.

Define "very good" please.

Filling your tank can be good as well.

I'm sure you believe that filling the tank is "accurate" since you
calculate 19.5 miles per gallon and not something like 19.5 rounded up to
20 and then the error taken into account such that it's more likely
anywhere between 19 and 21 mpg than it is 19.5 mpg.

It's not like they design gas tanks to have air pockets.

Actually, they do have air pockets.
Those air pockets change in size based on temperature & pressure & fill
level.

Even the fuel changes in density based on those parameters.

You don't need to know any of this specifically.

Of course I don't. 19.5 mpg is all I need to know.
And if I change "something" which results in 19.7mpg, then of course, that
something was the cause. I understand. I really do.

Why do you care which of the three has what specific degrees of accuracy and
precision?

I care because when I do a calculation, my assumption is that 19.5mpg is
actually something closer to 19 to 21 mpg than it is to 19.5.

If the "change" I'm measuring is within that margin of error, then I can't
say anything about what that "change" was.

And, more importantly, neither can you.
Which is the entire point after all.
The man is right You are wrong. You ASS U ME too much - and at the
risk of insulting the few GOOD engineers on the list, you OBVIOUISLY
are an "engineer", but not one I'd hire for a job. The job would come
in WAY over budget, WAY late, and would need to be completely redone
by techitians and technologists at great cost, or to save time and
money, completely decommissioned and scrapped - starting over with
someone who knew what thet were doing, and how to do it - engineer or
not.

a whole lot of crap snipped

I care because when I do a calculation, my assumption is that 19.5mpg is
actually something closer to 19 to 21 mpg than it is to 19.5.

If the "change" I'm measuring is within that margin of error, then I can't
say anything about what that "change" was.

And, more importantly, neither can you.
Which is the entire point after all.
Roger, me lad - you wouldn't happen to be a britiah trained engineer,
now, would you?? In what discipline of engineering?

On Saturday, July 22, 2017 at 9:10:45 PM UTC-4, wrote:
On Sat, 22 Jul 2017 23:42:25 -0000 (UTC), Mad Roger
wrote:

On Sat, 22 Jul 2017 00:46:50 -0400,
rickman wrote:

So my odometer is accurate and precise.

I understand you because you're exactly the type of person that I had in
mind when I asked the question in the first place.

I don't know what you mean. I have checked my odometer against the markers
on the highway as well as against my GPS (I think the highway markers are
more accurate than the GPS). It is spot on with the current tires to 1% or
better.

Does your tripmeter have a decimal place and digits after that decimal
place?

My speedometer is mechanical and so has a separate calibration factor.

The speedometer example was only brought in to point out that the vain hope
that averages result in better "accuracy" is patently false.

Mom-and-pop type of people actually believe that a speedometer reads even
close to accurately - and worse - some here propose the vain notion that
the more readings they take, somehow (magically?) the more accurate the
results will be.

A speedometer that reads high isn't going to result in more accurate
calculations even if you do a billion test runs.

+ A pumpmeter of 20.25 gallons is likely relatively accurate & precise

Of course it is. States inspect them at some point.

You don't seem to understand what accuracy and precision even mean.
Haven't you taken even one science lab course?

+ Matching fuel level in the tank isn't even close to accurate nor precise

I don't agree. I let the pump click off and then continue to pump for a
number of more clicks until it cuts off immediately.

I'm not at all surprised about your concept of the fuel-level estimation,
and, in fact, you're exactly the mom-and-pop type person I was talking
about when I opened the thread.

I understand you.

I always need to run
at least another fifteen miles before I am home so that is better part of a
gallon burned so I don't need to worry about the gas warming up and running
out of the tank. I believe this makes for very consistent fill ups.

I'm sure you do believe that.

My MPG results pretty well show the consistency of my measures.

I'm sure your MPG results support any theory you want them to support.
I believe you.

You know what happens when you assume...

You don't know how funny that statement was to me when I just read it now.

I see less than 19 or even 19.5 MPG.

I bet you see that decimal place even though it's not in the tripmeter
estimation nor in the filllevel estimation.

You see, I understand you because you're the type of person I had in mind
when I asked the question.

I think the consistency of my MPG readings show how well each of these can
be measured.

I'm sure you do.

As you say, the pump is going to be dead on.

Whoa! I never said the pump was "dead on" and anyone reading this thread
who thinks I think the pump is "dead on" would have completely
misunderstood everything else I said.

All I said was that the inaccuracies and imprecisions in the pump reading
are likely better than the otherwise astoundingly huge imprecision in the
fuel-fill level estimation and in the lesser inaccuracy of the tripmeter
estimation.

Other than scale
error which can be calibrated out the odometer will be very good.

Define "very good" please.

Filling your tank can be good as well.

I'm sure you believe that filling the tank is "accurate" since you
calculate 19.5 miles per gallon and not something like 19.5 rounded up to
20 and then the error taken into account such that it's more likely
anywhere between 19 and 21 mpg than it is 19.5 mpg.

It's not like they design gas tanks to have air pockets.

Actually, they do have air pockets.
Those air pockets change in size based on temperature & pressure & fill
level.

Even the fuel changes in density based on those parameters.

You don't need to know any of this specifically.

Of course I don't. 19.5 mpg is all I need to know.
And if I change "something" which results in 19.7mpg, then of course, that
something was the cause. I understand. I really do.

Why do you care which of the three has what specific degrees of accuracy and
precision?

I care because when I do a calculation, my assumption is that 19.5mpg is
actually something closer to 19 to 21 mpg than it is to 19.5.

If the "change" I'm measuring is within that margin of error, then I can't
say anything about what that "change" was.

And, more importantly, neither can you.
Which is the entire point after all.
The man is right You are wrong. You ASS U ME too much - and at the
risk of insulting the few GOOD engineers on the list, you OBVIOUISLY
are an "engineer", but not one I'd hire for a job. The job would come
in WAY over budget, WAY late, and would need to be completely redone
by techitians and technologists at great cost, or to save time and
money, completely decommissioned and scrapped - starting over with
someone who knew what thet were doing, and how to do it - engineer or
not.

Oh no..... Just yesterday Clare had found a new BFF, they were
ready to get a room together. Now they're bickering big time.

ROFL.

On Friday, July 21, 2017 at 9:37:31 PM UTC-4, dpb wrote:
On 07/21/2017 6:48 PM, trader_4 wrote:
...

No, he's been saying the same thing for many posts now. If you fill
up the tank 10 times, any inaccuracy from not exactly filling the tank
to the same level at the first time and last time is reduced by an
order of magnitude, because it only matters on the last fill. All the
other 9 fills, you have the gas pump reading.

I'm glad I "only" had to explain reactor power trip setpoint uncertainty
analysis/error propogation to the NRC staff statisticians and then sit
through the ACRS (Advisory Committee on Reactor Safeguards) hearings,
not usenet.

--



ROFL

On Saturday, July 22, 2017 at 8:56:47 PM UTC-4, dpb wrote:
On 07/22/2017 6:42 PM, Mad Roger wrote:
...

The main point for bringing up the otherwise unrelated issues of
speedometer accuracy and repeatability is that some people here seemed to
think if they ran a calculation more times, that the "accuracy" somehow
(magically?) gets better because (magically?) their reading is low as much
as it is high.
...

NO! That is not at all what any of those people said. You're simply
repeating the same contention which is ok as far as it goes in that a
point estimate is not the same as averaging or using other techniques to
increase the precision of the measurement.

They (and I in particular) only point out that averaging muddles out the
differences in intermediate filling levels.

+1

I'm not sure I;d even call it averaging, it's a different test
method. Two ways of doing it:

1 - Fill the tank once, drive until it's near empty, fill it again.
Your accuracy is greatly affected by your ability or inability
to fill it to exactly the same level. If you're off by a gallon
on a 15 gallon tank, it;s 7%.

2 - Fill it at the beginning, drive it a much longer distance,
through 10 tanks worth of gas where you have the pump reading
on all of those, then fill it the last time to as close to the
original fill as possible.

Method 2 reduces the inaccuracy due to not filling it to exactly
the same level by an order of magnitude. If you're off by a gallon
between the first and last fill, it's an error of ~0.7%.
You still have whatever the accuracy of the pumps are to deal with,
but I agree I'd trust that the pumps are going to be a lot
closer to the 0.7% accuracy than 7%. I'd think they are better
than 0.7%. And like you said, in between
the first and last fill, it doesn't matter if you fill it all the
way or only half full, etc.

On Sat, 22 Jul 2017 21:03:45 -0400,
wrote:

and exactly how is that germaine to the issue at hand? The speedo
means NOTHING. All we care about is the ODO - which will ALWAYS be
consistent, even in inaccuracy - so can be easily compensayed

While the odometer has a decimal place, the tripmeter generally does not.

So anyone using the tripmeter/pumpmeter calculation has no scientific right
to include the decimal place in the mpg calculation.

It's (mathematically) impossible to calculate 19.5 mpg when the tripmeter
reading doesn't have a decimal place.

Likewise, it's (mathematically) impossible to calculate 19.5 mpg when the
fill-level estimation isn't accurately known to some vague concept of less
than a gallon.

The only reading, if the three required, that is reasonably accurate
(someone quoted some figures already for the pumps, which I appreciate), is
the pumpmeter itself.

But that pumpmeter reading is completely dependent on the fill-level
estimation, which isn't known to any reasonable degree of accuracy.

I agree the way to get around the horrible fill-level inaccuracy is to
average over numerous tankfulls, which helps greatly, but doesn't eliminate
the two major inaccuraciesa which make it impossible for a mom-and-pop
tripmeter/pumpmeter calculation to have a decimal place in the result.

On Sat, 22 Jul 2017 19:56:37 -0500,
dpb wrote:

NO! That is not at all what any of those people said.

OK. Maybe I got it wrong.

Still, some people claim 19.5 miles per gallon, which I posit is an
impossible level of precision given the tripmeter/pumpmeter method, even
when taken over 10 tank fills.

Most tripmeters don't even have a decimal place, so, you can't include
decimal points in the calculation.

Worse, the fill-level estimation is crude at best, where again, there is no
decimal point.

It's a mathematical fact that if your measurements don't have decimal
points in them, then your answer can't have them either.

Anyone quoting MPG with a decimal point has to first get those decimal
points into the measurements!

On Sat, 22 Jul 2017 19:29:49 -0700 (PDT),
trader_4 wrote:

1 - Fill the tank once, drive until it's near empty, fill it again.
Your accuracy is greatly affected by your ability or inability
to fill it to exactly the same level. If you're off by a gallon
on a 15 gallon tank, it;s 7%.

2 - Fill it at the beginning, drive it a much longer distance,
through 10 tanks worth of gas where you have the pump reading
on all of those, then fill it the last time to as close to the
original fill as possible.

Method 2 reduces the inaccuracy due to not filling it to exactly
the same level by an order of magnitude. If you're off by a gallon
between the first and last fill, it's an error of ~0.7%.
You still have whatever the accuracy of the pumps are to deal with,
but I agree I'd trust that the pumps are going to be a lot
closer to the 0.7% accuracy than 7%. I'd think they are better
than 0.7%. And like you said, in between
the first and last fill, it doesn't matter if you fill it all the
way or only half full, etc.

I agree that averaging the "fill level" estimate is a great way to reduce
that huge error of guessing where the last fill level was.

I never disagreed with that, although I may not have realized it in the
very beginning. So, we have to assume a 10-tank fill when we state what our
innacuracies are.

I'm ok with assuming a 10-tank fill.

But remember, your inaccuracy is no better than your worst measurement, so,
what do we do about a tripmeter that has no decimal point?

I will wager that most people use the tripmeter and not the odometer.

If the tripmeter/pumpmeter has no decimal point in the numerator, you can't
possibly get a decimal point in the resulting calculation.

That's how math works.
Isn't it?

So it's not "me" you'd argue against.
It's math you have to argue against, since I'm just the messenger.

On Sat, 22 Jul 2017 19:44:26 -0500,
dpb wrote:

I got curious myself on what the numbers revealed and looked at the NIST
numbers again.

I computed an empirical cdf and compared it to normal...statistics from
the 20,036 observations are below:

[2 quoted lines suppressed]
s =
min: -50
max: 146
mean: -0.0788
std: 3.7681
median: 0
mode: 0
[1 quoted line suppressed]

I then compared to normal on the same plot and as outlined above
N(mean,std) is too long-tailed on both ends in comparison. It turns out
that N(mean,std/1.5) is pretty close on both tails to about the +/- 6 point.


Anyway, from the above it's simple enough to get some pretty good
estimates of what pump volume errors one might expect...the table below
is from the empirical cdf NIST data...

P error(in^3)/5Gal error(%)
0.001 -22 -1.82
0.005 -9 -0.78
0.010 -8 -0.69
0.025 -6 -0.52
0.050 -5 -0.43
0.250 -2 -0.17
0.500 0 0
0.750 2 0.17
0.900 4 0.34
0.950 5 0.43
0.975 6 0.52
0.990 7 0.60
0.995 10 0.86
0.999 22 1.82

From the above, one can conclude the pump metering error small for all
except the extreme outlier pumps.

I love that you are the only one quoting actual numbers and not pulling
them out of your butt to answer the question!

But your numbers confuse me because they seem to be in cubic inches.
You also mentioned that metric pumps are more accurate, but that's
impossible, simply because the pump is as accurate as the pump can get,
which, we can assume, is a mechanical thing (and not a metric thing).

All you're saying is that a liter is four times smaller than a gallon so
the error is four times less for a given liter versus a given gallon but
that's not saying it's more accurate. It's just saying the volume is less
so the resulting error is less.

Anyways, can you just summarize what the error is for a typical USA pump in
gallons?

For a typical 20-gallon fill, how many gallons off can reality be, plus or
minus from the indicated reading on the pumpmeter?

On Saturday, July 22, 2017 at 10:31:27 PM UTC-4, Mad Roger wrote:
On Sat, 22 Jul 2017 19:56:37 -0500,
dpb wrote:

NO! That is not at all what any of those people said.

OK. Maybe I got it wrong.

Still, some people claim 19.5 miles per gallon, which I posit is an
impossible level of precision given the tripmeter/pumpmeter method, even
when taken over 10 tank fills.

Most tripmeters don't even have a decimal place, so, you can't include
decimal points in the calculation.

Math challenged too, I see. I have 3 apples. I take one away.
What percentage is left? According to the above, I can't determine
that what's left is 66.66%

You certainly can include decimal points, the accuracy is affected
by the total distance traveled. If you travel 10 miles, then yes
the fact that you don't have tenths has a big effect. If you
drive 100 miles it matters by an order of magnitude less. If
you travel 10,000 miles, three orders of magnitude less and you
have several decimals of precision.




Worse, the fill-level estimation is crude at best, where again, there is no
decimal point.


No, worse is that it's been explained to you by what?, 3 people now,
that by filling the tank many times on a longer trip, the fill level
issue only matters on the last fill and the inaccuracy due to that
is greatly reduced by all the other fills being measured by the gas
pump, which is highly accurate.



It's a mathematical fact that if your measurements don't have decimal
points in them, then your answer can't have them either.


See the above examples.


Anyone quoting MPG with a decimal point has to first get those decimal
points into the measurements!

Just stop already.

On Saturday, July 22, 2017 at 10:40:14 PM UTC-4, Mad Roger wrote:
On Sat, 22 Jul 2017 19:44:26 -0500,
dpb wrote:

I got curious myself on what the numbers revealed and looked at the NIST
numbers again.

I computed an empirical cdf and compared it to normal...statistics from
the 20,036 observations are below:

[2 quoted lines suppressed]
s =
min: -50
max: 146
mean: -0.0788
std: 3.7681
median: 0
mode: 0
[1 quoted line suppressed]

I then compared to normal on the same plot and as outlined above
N(mean,std) is too long-tailed on both ends in comparison. It turns out
that N(mean,std/1.5) is pretty close on both tails to about the +/- 6 point.


Anyway, from the above it's simple enough to get some pretty good
estimates of what pump volume errors one might expect...the table below
is from the empirical cdf NIST data...

P error(in^3)/5Gal error(%)
0.001 -22 -1.82
0.005 -9 -0.78
0.010 -8 -0.69
0.025 -6 -0.52
0.050 -5 -0.43
0.250 -2 -0.17
0.500 0 0
0.750 2 0.17
0.900 4 0.34
0.950 5 0.43
0.975 6 0.52
0.990 7 0.60
0.995 10 0.86
0.999 22 1.82

From the above, one can conclude the pump metering error small for all
except the extreme outlier pumps.

I love that you are the only one quoting actual numbers and not pulling
them out of your butt to answer the question!

But your numbers confuse me because they seem to be in cubic inches.

You told us the other day you were some kind of scientist, yet
cubic inches confuse you?

"NIST tolerance is 6 in^3 in a 5 gal measure"

231 cubic inches in a gallon. 6/(231*5) = .005 or 0.5%

You really should just stop already.




You also mentioned that metric pumps are more accurate, but that's
impossible, simply because the pump is as accurate as the pump can get,
which, we can assume, is a mechanical thing (and not a metric thing).

All you're saying is that a liter is four times smaller than a gallon so
the error is four times less for a given liter versus a given gallon but
that's not saying it's more accurate. It's just saying the volume is less
so the resulting error is less.

Anyways, can you just summarize what the error is for a typical USA pump in
gallons?

For a typical 20-gallon fill, how many gallons off can reality be, plus or
minus from the indicated reading on the pumpmeter?

You told us the other day you're some kind of scientist. DPB told
you the NIST

On 07/22/2017 08:31 PM, Mad Roger wrote:
Most tripmeters don't even have a decimal place, so, you can't include
decimal points in the calculation.

I cannot address 'most' tripmeters only those on the five vehicles I
own. They all read to tenths of a mile.

Worse, the fill-level estimation is crude at best, where again, there is no
decimal point.

I don't usually pay that much attention but with this thread in my mind
when I fueled the bike today I noticed the gallons on the pump had not
one but three decimal places. I cannot attest to the accuracy. I don't
know what criteria the inspector uses when he places his seal on the pump.

It's a mathematical fact that if your measurements don't have decimal
points in them, then your answer can't have them either.

Anyone quoting MPG with a decimal point has to first get those decimal
points into the measurements!

Like I said, I have all sorts of decimal places available.

On 07/22/2017 08:40 PM, Mad Roger wrote:
But remember, your inaccuracy is no better than your worst measurement, so,
what do we do about a tripmeter that has no decimal point?

I will wager that most people use the tripmeter and not the odometer.


I'll go out on a limb here -- where do we find a trip meter with no
decimal place? In fact, before trip meters became ubiquitous, the
odometer showed tenths.

https://i.ebayimg.com/images/g/wVoAA...gCa/s-l300.jpg

This is similar to the panel in my '86 F150. Early in the game both the
odometer and trip meter were mechanical and read tenths.

Think about it. In a newer car where the odometer only shows miles, how
do you follow directions like 'turn right four tenths of a mile past
mile marker 37' ?

On 07/22/2017 08:31 PM, Mad Roger wrote:
While the odometer has a decimal place, the tripmeter generally does not.

Please post specific car models where that is the case.

On 07/22/2017 11:52 AM, Ed Pawlowski wrote:
On 7/22/2017 1:38 PM, rbowman wrote:



I should look at the instantaneous readouts versus mph to see if the
mpg falls off gradually or if there is an efficiency sweet spot around
65-70. Except for around the cities the interstate speed limit in this
and some of the adjoining states is 80. Drive 65 at your own risk.


I tried that one day on a flat stretch so there would be little
variance. This was on my regular trip to work. Speed limit is 65. One
day I did 70, the next 65, then at 60 is was dicey, the next day I tried
55 for about 30 seconds and decided not to risk my life.

I forget the details, but 60 was better than 70 by a couple of mpg.
Problem is, I prefer driving 75. If I could get away with it I'd go 85+
but don't want to pay the fines.

At under 70 my car usually is in the 35 mpg + range; at 80, it is more
like 32. I get even better mileage in Oregon with its 55 mph speed
limit. I also get bored out of my mind. There isn't a whole lot of
anything between Ontario and Bend but I figure as soon as I get up to a
decent speed a OSP cruiser will materialize from the sagebrush.

That stupid speed limit is the least of Oregon's problems.

On Sun, 23 Jul 2017 02:31:23 -0000 (UTC), Mad Roger
wrote:

On Sat, 22 Jul 2017 21:03:45 -0400,
wrote:

and exactly how is that germaine to the issue at hand? The speedo
means NOTHING. All we care about is the ODO - which will ALWAYS be
consistent, even in inaccuracy - so can be easily compensayed

While the odometer has a decimal place, the tripmeter generally does not.
Backwards, my boy. Every car I have ever owned with a trip odo has
tenths on the trip - many (or at least a few) with trips do not have
tenths on the main ODO

So anyone using the tripmeter/pumpmeter calculation has no scientific right
to include the decimal place in the mpg calculation.

Get a LIFE!!

It's (mathematically) impossible to calculate 19.5 mpg when the tripmeter
reading doesn't have a decimal place.

But it DOES - and you don't have to use the trip odo if you don't want
to. Use a pen and paper (you remeber those?) and write down the start
mileage when you do the first fill up. Keep track of all fuel added
AFTER that mileage, and at the last fillup, total all the fuel added,
write down the current mileage, and subtract (as an engineer you DO
know how to subtract, right??( - the difference is the total mileage
covered. Devide that number, complete with tenths,by your total fuel
consumed, to the tenth of a unit, and you have your accurate fuel
consumption, to the tenth of a MPG ( or in the case of a car
calibrated in Km, to the tenth of a Km/liter - which is a WHOLE LOT
more accurate.

Likewise, it's (mathematically) impossible to calculate 19.5 mpg when the
fill-level estimation isn't accurately known to some vague concept of less
than a gallon.

But if you are as smart as the average fifrg grader you KNOW it is
possible to get a LOT closer than that. Fill till the fuel is visible
in the filler neck, at the lead free gas restrictor plate, and you are
accurate to within about half a cup full in the vast majority of
cases.

The only reading, if the three required, that is reasonably accurate
(someone quoted some figures already for the pumps, which I appreciate), is
the pumpmeter itself.

But that pumpmeter reading is completely dependent on the fill-level
estimation, which isn't known to any reasonable degree of accuracy.

Bull crap

I agree the way to get around the horrible fill-level inaccuracy is to
average over numerous tankfulls, which helps greatly, but doesn't eliminate
the two major inaccuraciesa which make it impossible for a mom-and-pop
tripmeter/pumpmeter calculation to have a decimal place in the result.
Bull Crap.

On 07/22/2017 07:14 PM, wrote:
a whole lot of crap snipped

I care because when I do a calculation, my assumption is that 19.5mpg is
actually something closer to 19 to 21 mpg than it is to 19.5.

If the "change" I'm measuring is within that margin of error, then I can't
say anything about what that "change" was.

And, more importantly, neither can you.
Which is the entire point after all.
Roger, me lad - you wouldn't happen to be a britiah trained engineer,
now, would you?? In what discipline of engineering?


Engineering Management, I'm thinking.

On Sun, 23 Jul 2017 02:31:24 -0000 (UTC), Mad Roger
wrote:

On Sat, 22 Jul 2017 19:56:37 -0500,
dpb wrote:

NO! That is not at all what any of those people said.

OK. Maybe I got it wrong.
Darn right you did.

Still, some people claim 19.5 miles per gallon, which I posit is an
impossible level of precision given the tripmeter/pumpmeter method, even
when taken over 10 tank fills.
Bull crap

Most tripmeters don't even have a decimal place, so, you can't include
decimal points in the calculation.
Bull crap

Worse, the fill-level estimation is crude at best, where again, there is no
decimal point.
Bull crap

It's a mathematical fact that if your measurements don't have decimal
points in them, then your answer can't have them either.
So? Both can and do.

Anyone quoting MPG with a decimal point has to first get those decimal
points into the measurements!
WEhich any grade 5 student could do (at least when I was in grade 5)

On Sun, 23 Jul 2017 02:40:09 -0000 (UTC), Mad Roger
wrote:

On Sat, 22 Jul 2017 19:29:49 -0700 (PDT),
trader_4 wrote:

1 - Fill the tank once, drive until it's near empty, fill it again.
Your accuracy is greatly affected by your ability or inability
to fill it to exactly the same level. If you're off by a gallon
on a 15 gallon tank, it;s 7%.

2 - Fill it at the beginning, drive it a much longer distance,
through 10 tanks worth of gas where you have the pump reading
on all of those, then fill it the last time to as close to the
original fill as possible.

Method 2 reduces the inaccuracy due to not filling it to exactly
the same level by an order of magnitude. If you're off by a gallon
between the first and last fill, it's an error of ~0.7%.
You still have whatever the accuracy of the pumps are to deal with,
but I agree I'd trust that the pumps are going to be a lot
closer to the 0.7% accuracy than 7%. I'd think they are better
than 0.7%. And like you said, in between
the first and last fill, it doesn't matter if you fill it all the
way or only half full, etc.

I agree that averaging the "fill level" estimate is a great way to reduce
that huge error of guessing where the last fill level was.

I never disagreed with that, although I may not have realized it in the
very beginning. So, we have to assume a 10-tank fill when we state what our
innacuracies are.

I'm ok with assuming a 10-tank fill.

But remember, your inaccuracy is no better than your worst measurement, so,
what do we do about a tripmeter that has no decimal point?
You open your eyes, and get over it. What car do you have with no
tenths on either the main or trip odo???? Use your head and write down
the start mileage, accurate top the tenth on the main ODO if you
happen to have the rasre car with no tenths on the tripmeter.

I will wager that most people use the tripmeter and not the odometer.
You would lose your bet

If the tripmeter/pumpmeter has no decimal point in the numerator, you can't
possibly get a decimal point in the resulting calculation.

That's how math works.
Isn't it?

If yopu are an engineer too stupid to find a way around it, I suppose
so.
So it's not "me" you'd argue against.
It's math you have to argue against, since I'm just the messenger.
And a deaf, dunb and blind messanger. A pedantic child - - -

On 07/22/2017 08:40 PM, Mad Roger wrote:
For a typical 20-gallon fill, how many gallons off can reality be, plus or
minus from the indicated reading on the pumpmeter?

http://www.npr.org/templates/story/s...toryId=5304258

The article talks about Washington but most states have a similar
protocol. Pump 5 gallons of gas. 1 gallon is 231 cubic inches, so that
is 1155 cubic inches. The volume must be within 6 cubic inches or
roughly 0.5%. I'll let you do the math for 20 gallons.

http://billingsgazette.com/news/stat...5351736b4.html

Montana uses the same test. Note that he estimates 2 to 3% of the pumps
fail and have to be repaired and also says with normal wear the pumps
tend to dispense more than stated but some may dispense less. That's
where averaging over a number of tanks comes in unless you fill up at
the same pump at the same station every time. I certainly don't.

On Sun, 23 Jul 2017 02:40:10 -0000 (UTC), Mad Roger
wrote:

On Sat, 22 Jul 2017 19:44:26 -0500,
dpb wrote:

I got curious myself on what the numbers revealed and looked at the NIST
numbers again.

I computed an empirical cdf and compared it to normal...statistics from
the 20,036 observations are below:

[2 quoted lines suppressed]
s =
min: -50
max: 146
mean: -0.0788
std: 3.7681
median: 0
mode: 0
[1 quoted line suppressed]

I then compared to normal on the same plot and as outlined above
N(mean,std) is too long-tailed on both ends in comparison. It turns out
that N(mean,std/1.5) is pretty close on both tails to about the +/- 6 point.


Anyway, from the above it's simple enough to get some pretty good
estimates of what pump volume errors one might expect...the table below
is from the empirical cdf NIST data...

P error(in^3)/5Gal error(%)
0.001 -22 -1.82
0.005 -9 -0.78
0.010 -8 -0.69
0.025 -6 -0.52
0.050 -5 -0.43
0.250 -2 -0.17
0.500 0 0
0.750 2 0.17
0.900 4 0.34
0.950 5 0.43
0.975 6 0.52
0.990 7 0.60
0.995 10 0.86
0.999 22 1.82

From the above, one can conclude the pump metering error small for all
except the extreme outlier pumps.

I love that you are the only one quoting actual numbers and not pulling
them out of your butt to answer the question!

But your numbers confuse me because they seem to be in cubic inches.
You are the engineer, son of physics majors - figure it out!!!
You also mentioned that metric pumps are more accurate, but that's
impossible, simply because the pump is as accurate as the pump can get,
which, we can assume, is a mechanical thing (and not a metric thing).
You fail to grasp the simple fact that a tenth of a liter is a whole
lot less than a tenth of a gallon???? Accuracy of READING the pump is
therefore about 4 times more accurate with a metric pump, because your
read error of +.1/-0 units is based on the much smaller unit.

All you're saying is that a liter is four times smaller than a gallon so
the error is four times less for a given liter versus a given gallon but
that's not saying it's more accurate. It's just saying the volume is less
so the resulting error is less.

and your engineer's understanding of accuracy does not equate to a
smaller error?????????

Anyways, can you just summarize what the error is for a typical USA pump in
gallons?
As good as Less than 1/10 of a percent according to the information
quoted, with a very few as bad as 1.82%. An american gallon is 128
fluid ounces, so 1.82% of 128 ounces is 2.23 ounces maximum error,
+/1, with most being within .5%, or 0.64 ounces per gallon

For a typical 20-gallon fill, how many gallons off can reality be, plus or
minus from the indicated reading on the pumpmeter?
The poorest pump checked in that data would be +/- 44.6 oz per 20
gallon tank - the average about +/- 12 ounces.
ASS U MEing the error is randomly distributed,around zero, your
chances of the error being anywhere CLOSE to even the 12 ounces is so
small as to be virtually insignificant unless you always used the same
pump - in which case it is totally immaterial if used for comparative
purposes.

For an engineer, you sure have a poor grasp of the concepts.

On Sat, 22 Jul 2017 22:12:26 -0600, rbowman
wrote:

At under 70 my car usually is in the 35 mpg + range; at 80, it is more
like 32. I get even better mileage in Oregon with its 55 mph speed
limit. I also get bored out of my mind. There isn't a whole lot of
anything between Ontario and Bend but I figure as soon as I get up to a
decent speed a OSP cruiser will materialize from the sagebrush.

That stupid speed limit is the least of Oregon's problems.

Where is their limit 55?

On 07/22/2017 10:45 PM, Bill Vanek wrote:
On Sat, 22 Jul 2017 22:12:26 -0600, rbowman
wrote:

At under 70 my car usually is in the 35 mpg + range; at 80, it is more
like 32. I get even better mileage in Oregon with its 55 mph speed
limit. I also get bored out of my mind. There isn't a whole lot of
anything between Ontario and Bend but I figure as soon as I get up to a
decent speed a OSP cruiser will materialize from the sagebrush.

That stupid speed limit is the least of Oregon's problems.

Where is their limit 55?


The last time I was there US20, US395, and other 2 lane roads in eastern
Oregon. Apparently the raised it to 65 in March of 2016 but are rolling
it back in some places.

http://www.oregonlive.com/commuting/...eed_limit.html

but according to this the limit is now 70 on rural roads:

http://www.speed-limits.com/oregon.htm

70 on Rt. 20 would make a lot more sense if that is indeed what it is
now. I'm not planning to check it out personally though.

On Sat, 22 Jul 2017 21:10:41 -0400, wrote:

On Sat, 22 Jul 2017 23:42:25 -0000 (UTC), Mad Roger
wrote:

On Sat, 22 Jul 2017 00:46:50 -0400,
rickman wrote:

So my odometer is accurate and precise.

I understand you because you're exactly the type of person that I had in
mind when I asked the question in the first place.

I don't know what you mean. I have checked my odometer against the markers
on the highway as well as against my GPS (I think the highway markers are
more accurate than the GPS). It is spot on with the current tires to 1% or
better.

Does your tripmeter have a decimal place and digits after that decimal
place?

snip

Which is the entire point after all.
The man is right You are wrong. You ASS U ME too much - and at the
risk of insulting the few GOOD engineers on the list, you OBVIOUISLY
are an "engineer", but not one I'd hire for a job. The job would come
in WAY over budget, WAY late, and would need to be completely redone
by techitians and technologists at great cost, or to save time and
money, completely decommissioned and scrapped - starting over with
someone who knew what thet were doing, and how to do it - engineer or
not.

You know, this guy has a hard-on against "non-engineers" measuring their MPG.
Rickman above told him he uses his odometer, then he goes on about tripmeters.
I answered his main complaints in another post. That exchange went like this:

"+ Tripmeter accuracy is what in the average car over a 300-mile tank?
+ Owners ability to "match" the previous level of fuel is what?
+ Gas station pump reading accuracy is what?

I never used the tripmeter for MPG, because I never bothered testing them with mile
markers.
Matching gas level is trivial - and it only has to done at the beginning and end of the
trip.
Gas station pumps - I assume they are accurate, and can't control that anyway.
I'm confident that my measurements are accurate to within .1 MPG."

His response to me totally ignored those responses, and he posed the same questions again!
Then, for some reason, he stated talking about speedometers.
He's a troll.

On 07/23/2017 1:12 AM, rbowman wrote:
On 07/22/2017 10:45 PM, Bill Vanek wrote:
On Sat, 22 Jul 2017 22:12:26 -0600, rbowman
wrote:

....

That stupid speed limit is the least of Oregon's problems.

Where is their limit 55?


The last time I was there US20, US395, and other 2 lane roads in eastern
Oregon. Apparently the raised it to 65 in March of 2016 but are rolling
it back in some places.
....


They're not the only seemingly bizarre place--between Clayton and
Springer is 100 mi of open country with either 55 (or _maybe_ 60) that
makes no common sense at all...

--

On 07/22/2017 9:40 PM, Mad Roger wrote:
On Sat, 22 Jul 2017 19:44:26 -0500,
dpb wrote:

I got curious myself on what the numbers revealed and looked at the NIST
numbers again.

I computed an empirical cdf and compared it to normal...statistics from
the 20,036 observations are below:

[2 quoted lines suppressed]
s =
min: -50
max: 146
mean: -0.0788
std: 3.7681
median: 0
mode: 0
....

Anyway, from the above it's simple enough to get some pretty good
estimates of what pump volume errors one might expect...the table below
is from the empirical cdf NIST data...

P error(in^3)/5Gal error(%)
0.001 -22 -1.82
0.005 -9 -0.78
0.010 -8 -0.69
0.025 -6 -0.52
0.050 -5 -0.43
0.250 -2 -0.17
0.500 0 0
0.750 2 0.17
0.900 4 0.34
0.950 5 0.43
0.975 6 0.52
0.990 7 0.60
0.995 10 0.86
0.999 22 1.82

From the above, one can conclude the pump metering error small for all
except the extreme outlier pumps.

....

But your numbers confuse me because they seem to be in cubic inches.

Well, yes, as said before the NIST standard for compliance testing is a
metering error of 6 cu in in 5 gal so the reported data are the
observed errors in a 5 gal test...

You also mentioned that metric pumps are more accurate, ...

_I_ said nothing about metric anywhere during the thread. Another
respondent pointed out that a liter, being smaller than a gallon, when
metered to the same tenth of a unit as the gallon will be a smaller
absolute error than in gallons. Seems fairly obvious...

Anyways, can you just summarize what the error is for a typical USA pump in
gallons?

The above data showed that the most probable error was 0 (actually less
than +/-1 ci since data are reported to nearest whole number). I
mentioned multiple times already and its in the table the distribution
was symmetrical and the mode and median were both 0. The mean is just
under 0.1 ci (-0.08) so there'd be a good place to start for just a
random pump taken from the population of pumps.

For a typical 20-gallon fill, how many gallons off can reality be, plus or
minus from the indicated reading on the pumpmeter?

Take your choice of how conservative you want to be or how likely it is
to be of that magnitude--that's why I gave the ecdf data--you can choose
the appropriate number for the particular use.

20*0.018 -- 0.36 gal would encompass 99.9% of all observed pumps on
either the over- or under-dispensing side; the likelihood finding an
operational pump of that drastic a metering error would be only 0.1%
though, so not likely.

OTOH the most probable pump taken from random would be 1 and so using
that as a bound, 1 ci-- 0.004329 US gal * 20 -- 0.0866 gal. Or, iow,
about what the 0.1 gal pump readout would indicate.

Of course, like any probability, what a particular realization will be
is totally dependent upon the actual pump used but the (sizable)
sampling of operating pumps taken during routine weights and measures
compliance checks shows that in general they work pretty well with a few
that have issues.

An interesting sidelight on that was the summary table of percentage
failures (exceeding the 6 cu in threshold) by pump manufacturer. There
were 4 with 100% compliance, another for in the high 80-90% range,
another 4/5 in the low 80% and then one laggard at 73%. I'd not have
guessed there were so many manufacturers but it appears there's a
price/performance element there as is so often the case...

(The manufacturers were anonymous so no way to use the data to go find a
station with one of the compliant pumps, unfortunately ). It did
note that W&M compliance checks could be much more effective at a given
cost/manpower level if used stratified sampling by vendor...


--

On Sunday, July 23, 2017 at 12:19:41 AM UTC-4, wrote:
On Sun, 23 Jul 2017 02:31:24 -0000 (UTC), Mad Roger
wrote:

On Sat, 22 Jul 2017 19:56:37 -0500,
dpb wrote:

NO! That is not at all what any of those people said.

OK. Maybe I got it wrong.
Darn right you did.

Still, some people claim 19.5 miles per gallon, which I posit is an
impossible level of precision given the tripmeter/pumpmeter method, even
when taken over 10 tank fills.
Bull crap

Most tripmeters don't even have a decimal place, so, you can't include
decimal points in the calculation.
Bull crap

Worse, the fill-level estimation is crude at best, where again, there is no
decimal point.
Bull crap

It's a mathematical fact that if your measurements don't have decimal
points in them, then your answer can't have them either.
So? Both can and do.

Anyone quoting MPG with a decimal point has to first get those decimal
points into the measurements!
WEhich any grade 5 student could do (at least when I was in grade 5)

Oh my, Clare and his new buddy aren't getting along well at all.

On Sunday, July 23, 2017 at 12:39:12 AM UTC-4, wrote:


But your numbers confuse me because they seem to be in cubic inches.
You are the engineer, son of physics majors - figure it out!!!
You also mentioned that metric pumps are more accurate, but that's
impossible, simply because the pump is as accurate as the pump can get,
which, we can assume, is a mechanical thing (and not a metric thing).
You fail to grasp the simple fact that a tenth of a liter is a whole
lot less than a tenth of a gallon???? Accuracy of READING the pump is
therefore about 4 times more accurate with a metric pump, because your
read error of +.1/-0 units is based on the much smaller unit.


Wrong. If a pump has an accuracy of 1%, then it has an accuracy of
1%. Whether what you draw is measured in liters or gallons.




All you're saying is that a liter is four times smaller than a gallon so
the error is four times less for a given liter versus a given gallon but
that's not saying it's more accurate. It's just saying the volume is less
so the resulting error is less.

and your engineer's understanding of accuracy does not equate to a
smaller error?????????

Per above, Mad Roger is right on that one. Do you really think a pump
that reads out in liters is going to be more accurate than the same
pump that reads out in gallons? It's just a math conversion going to the
display.



For an engineer, you sure have a poor grasp of the concepts.

I don't recall Mad Roger claiming to be an engineer. I think he
said he was a "scientist" with vague references to biology or
life sciences.

PS: Why can't you trim posts?

On Sunday, July 23, 2017 at 7:20:18 AM UTC-4, Vic Smith wrote:
On Sat, 22 Jul 2017 21:10:41 -0400, wrote:

On Sat, 22 Jul 2017 23:42:25 -0000 (UTC), Mad Roger
wrote:

On Sat, 22 Jul 2017 00:46:50 -0400,
rickman wrote:

So my odometer is accurate and precise.

I understand you because you're exactly the type of person that I had in
mind when I asked the question in the first place.

I don't know what you mean. I have checked my odometer against the markers
on the highway as well as against my GPS (I think the highway markers are
more accurate than the GPS). It is spot on with the current tires to 1% or
better.

Does your tripmeter have a decimal place and digits after that decimal
place?

snip

Which is the entire point after all.
The man is right You are wrong. You ASS U ME too much - and at the
risk of insulting the few GOOD engineers on the list, you OBVIOUISLY
are an "engineer", but not one I'd hire for a job. The job would come
in WAY over budget, WAY late, and would need to be completely redone
by techitians and technologists at great cost, or to save time and
money, completely decommissioned and scrapped - starting over with
someone who knew what thet were doing, and how to do it - engineer or
not.

You know, this guy has a hard-on against "non-engineers" measuring their MPG.
Rickman above told him he uses his odometer, then he goes on about tripmeters.
I answered his main complaints in another post. That exchange went like this:

"+ Tripmeter accuracy is what in the average car over a 300-mile tank?
+ Owners ability to "match" the previous level of fuel is what?
+ Gas station pump reading accuracy is what?

I never used the tripmeter for MPG, because I never bothered testing them with mile
markers.
Matching gas level is trivial - and it only has to done at the beginning and end of the
trip.
Gas station pumps - I assume they are accurate, and can't control that anyway.
I'm confident that my measurements are accurate to within .1 MPG."

His response to me totally ignored those responses, and he posed the same questions again!
Then, for some reason, he stated talking about speedometers.
He's a troll.

I noted the same thing in his other thread about wheel sizes. Right from
the start Mad Roger made the wrong assumption that torque at the wheel
translated directly into fuel usage. He was conflating torque with
work/energy. So, I carefully explained it, gave him examples of lifting
a 100lb rock up 2ft using a 4 ft lever or a 10 ft lever. While the
torque is different, the work/energy used is exactly the same. Gave
him another example of a bicycle going uphill with different gear ratios.
Did he acknowledge that, discuss it? No, just like with your example,
instead he obfuscated by then bringing up wind resistance and continued
to troll on. He also claimed to be a "scientist", then he says that
DPB's post about fuel pump accuracy was confusing because he used
cubic inches and gallons?

We don't agree on a lot things, but we sure agree on this!

On 07/22/2017 11:39 PM, wrote:
On Sun, 23 Jul 2017 02:40:10 -0000 (UTC), Mad Roger
wrote:

On Sat, 22 Jul 2017 19:44:26 -0500,
dpb wrote:

I got curious myself on what the numbers revealed and looked at the NIST
numbers again.

I computed an empirical cdf and compared it to normal...statistics from
the 20,036 observations are below:

,,,

min: -50
max: 146
mean: -0.0788
std: 3.7681
median: 0
mode: 0
....

For a typical 20-gallon fill, how many gallons off can reality be, plus or
minus from the indicated reading on the pumpmeter?
The poorest pump checked in that data would be +/- 44.6 oz per 20
gallon tank - the average about +/- 12 ounces.
....

Actually, the extrema limits in the sample are much worse: -50 and +146
in the 5 gal test collection. I simply computed the table for typical
exceedance limits; didn't include the endmost points there. The 0.999
point, for example, is the int(20036*0.999)-th observation or 20015 so
there were 21 more observations above that value. The last two,
however, were really, really outliers that skew things quite a lot. The
last five observations were

NIST(end-5:end,
ans =
28 1
29 1
30 4
56 1
127 3
146 1


It's quite an oddity that there were 3 observations at 127; illustrating
again that "random data aren't" or more correctly that one can always
find patterns visually even in random data.

Of course, if the pump is putting out 5.6 gal/5 gal reading, you'll only
have "pumped" 20*5/5.6 -- ~17.8 gal when the tank is full already...

Hmmm...I had noticed the outlier on the positive end and looked at it;
wonder what the LH tail looks like now...

NIST(1:5,
ans =
-50 1
-45 1
-36 1
-35 4
-32 1


Pretty similar pattern, just not as extreme. There are two outliers
separated quite a bit from the bulk of the rest of the observations, but
they're only 10 cu in differential to nearest whereas there's 70+ on the
other end.

Interesting that from a customer viewpoint you're just about as likely
to get more than you're paying for as under and on the extreme ends by a
lot more than by what you get shorted.

I didn't try to find what Canadian limits are -- I presume they must be
somewhat more stringent in order to match, more or less, the size of the
measurement interval? You know?

--

On Sun, 23 Jul 2017 10:04:30 -0500, dpb wrote:

On 07/22/2017 9:40 PM, Mad Roger wrote:
On Sat, 22 Jul 2017 19:44:26 -0500,
dpb wrote:

I got curious myself on what the numbers revealed and looked at the NIST
numbers again.

I computed an empirical cdf and compared it to normal...statistics from
the 20,036 observations are below:

[2 quoted lines suppressed]
s =
min: -50
max: 146
mean: -0.0788
std: 3.7681
median: 0
mode: 0
...

Anyway, from the above it's simple enough to get some pretty good
estimates of what pump volume errors one might expect...the table below
is from the empirical cdf NIST data...

P error(in^3)/5Gal error(%)
0.001 -22 -1.82
0.005 -9 -0.78
0.010 -8 -0.69
0.025 -6 -0.52
0.050 -5 -0.43
0.250 -2 -0.17
0.500 0 0
0.750 2 0.17
0.900 4 0.34
0.950 5 0.43
0.975 6 0.52
0.990 7 0.60
0.995 10 0.86
0.999 22 1.82

From the above, one can conclude the pump metering error small for all
except the extreme outlier pumps.

...

But your numbers confuse me because they seem to be in cubic inches.

Well, yes, as said before the NIST standard for compliance testing is a
metering error of 6 cu in in 5 gal so the reported data are the
observed errors in a 5 gal test...

For the mathematically and engineering challenged, that is 3.325
fluid ounces - or less than half a cup - or 3.25 in 640 - or an error
of less than 0.5%
You also mentioned that metric pumps are more accurate, ...

_I_ said nothing about metric anywhere during the thread. Another
respondent pointed out that a liter, being smaller than a gallon, when
metered to the same tenth of a unit as the gallon will be a smaller
absolute error than in gallons. Seems fairly obvious...

Anyways, can you just summarize what the error is for a typical USA pump in
gallons?

The above data showed that the most probable error was 0 (actually less
than +/-1 ci since data are reported to nearest whole number). I
mentioned multiple times already and its in the table the distribution
was symmetrical and the mode and median were both 0. The mean is just
under 0.1 ci (-0.08) so there'd be a good place to start for just a
random pump taken from the population of pumps.

For a typical 20-gallon fill, how many gallons off can reality be, plus or
minus from the indicated reading on the pumpmeter?

Take your choice of how conservative you want to be or how likely it is
to be of that magnitude--that's why I gave the ecdf data--you can choose
the appropriate number for the particular use.

20*0.018 -- 0.36 gal would encompass 99.9% of all observed pumps on
either the over- or under-dispensing side; the likelihood finding an
operational pump of that drastic a metering error would be only 0.1%
though, so not likely.

OTOH the most probable pump taken from random would be 1 and so using
that as a bound, 1 ci-- 0.004329 US gal * 20 -- 0.0866 gal. Or, iow,
about what the 0.1 gal pump readout would indicate.

Of course, like any probability, what a particular realization will be
is totally dependent upon the actual pump used but the (sizable)
sampling of operating pumps taken during routine weights and measures
compliance checks shows that in general they work pretty well with a few
that have issues.

An interesting sidelight on that was the summary table of percentage
failures (exceeding the 6 cu in threshold) by pump manufacturer. There
were 4 with 100% compliance, another for in the high 80-90% range,
another 4/5 in the low 80% and then one laggard at 73%. I'd not have
guessed there were so many manufacturers but it appears there's a
price/performance element there as is so often the case...

(The manufacturers were anonymous so no way to use the data to go find a
station with one of the compliant pumps, unfortunately ). It did
note that W&M compliance checks could be much more effective at a given
cost/manpower level if used stratified sampling by vendor...

On Sun, 23 Jul 2017 11:21:22 -0500, dpb wrote:

On 07/22/2017 11:39 PM, wrote:
On Sun, 23 Jul 2017 02:40:10 -0000 (UTC), Mad Roger
wrote:

On Sat, 22 Jul 2017 19:44:26 -0500,
dpb wrote:

I got curious myself on what the numbers revealed and looked at the NIST
numbers again.

I computed an empirical cdf and compared it to normal...statistics from
the 20,036 observations are below:

,,,

min: -50
max: 146
mean: -0.0788
std: 3.7681
median: 0
mode: 0
...

For a typical 20-gallon fill, how many gallons off can reality be, plus or
minus from the indicated reading on the pumpmeter?
The poorest pump checked in that data would be +/- 44.6 oz per 20
gallon tank - the average about +/- 12 ounces.
...

Actually, the extrema limits in the sample are much worse: -50 and +146
in the 5 gal test collection. I simply computed the table for typical
exceedance limits; didn't include the endmost points there. The 0.999
point, for example, is the int(20036*0.999)-th observation or 20015 so
there were 21 more observations above that value. The last two,
however, were really, really outliers that skew things quite a lot. The
last five observations were

NIST(end-5:end,
ans =
28 1
29 1
30 4
56 1
127 3
146 1


It's quite an oddity that there were 3 observations at 127; illustrating
again that "random data aren't" or more correctly that one can always
find patterns visually even in random data.

Of course, if the pump is putting out 5.6 gal/5 gal reading, you'll only
have "pumped" 20*5/5.6 -- ~17.8 gal when the tank is full already...

Hmmm...I had noticed the outlier on the positive end and looked at it;
wonder what the LH tail looks like now...

NIST(1:5,
ans =
-50 1
-45 1
-36 1
-35 4
-32 1


Pretty similar pattern, just not as extreme. There are two outliers
separated quite a bit from the bulk of the rest of the observations, but
they're only 10 cu in differential to nearest whereas there's 70+ on the
other end.

Interesting that from a customer viewpoint you're just about as likely
to get more than you're paying for as under and on the extreme ends by a
lot more than by what you get shorted.

I didn't try to find what Canadian limits are -- I presume they must be
somewhat more stringent in order to match, more or less, the size of the
measurement interval? You know?
I do not know the requirements or test results today, but I DO know
back when we had mechanical meters our pumps (at the stations where I
worked)were never out by more than a couple oz in the 5 gallon
calibration, and the new electronic metering pumps (deployed when we
switched from imperial Gallons to Liters) were "significantly more
accurate" in metering. The accuracy changed a bit with delivery speed
on the mechanical pumps - can't remember if the change was that the
read higher or lower with reduced pump speed, but the variance was
quite low. The electronic meters were supposedly less susceptible to
volume arrors based on fuel velocity.

The last years of my automotive career were not involved with gasoline
retailing in any way - from a short time after the switch to Metric.

On 07/23/2017 12:32 PM, wrote:
The accuracy changed a bit with delivery speed
on the mechanical pumps - can't remember if the change was that the
read higher or lower with reduced pump speed, but the variance was
quite low. The electronic meters were supposedly less susceptible to
volume arrors based on fuel velocity.

I haven't noticed if it is on all pumps but at least some of the ones
I've used have a sticker reinforcing that the gasoline is dispensed by
volume, not energy content. I've had a couple of times when I've filled
a motorcycle tank with cool gasoline from underground storage tanks on a
hot summer day and suddenly found I'd gotten more volume than I'd paid
for -- and more volume than the tank would hold.

On 07/23/2017 1:32 PM, wrote:
On Sun, 23 Jul 2017 11:21:22 -0500, wrote:
....

I didn't try to find what Canadian limits are -- I presume they must be
somewhat more stringent in order to match, more or less, the size of the
measurement interval? You know?

I do not know the requirements or test results today, but I DO know
back when we had mechanical meters our pumps (at the stations where I
worked)were never out by more than a couple oz in the 5 gallon
calibration, and the new electronic metering pumps (deployed when we
switched from imperial Gallons to Liters) were "significantly more
accurate" in metering. The accuracy changed a bit with delivery speed
on the mechanical pumps - can't remember if the change was that the
read higher or lower with reduced pump speed, but the variance was
quite low. The electronic meters were supposedly less susceptible to
volume arrors based on fuel velocity.
....

As noted, being they're dispensing quantities measured in units that are
roughly one-fourth the size one would suppose tolerances would be
adjusted similarly. The 0.5% error with the US compliance standard
would be almost 2% which seems as would be excessive in comparison to
what a governing consumer-protection function would deem adequate.

Nothing came up in the searches I did, but I didn't look for non-US
data, either, so being in US not surprising what I found was what I
found...

Anyways, I think we can put the subject to rest...

--