What is a realistic accuracy & precision of typical MPG measurements when
measured by the consumer using the typical method of dividing their
tripmeter miles by the gas-pump gallons during fillup?

On Thu, 20 Jul 2017 12:08:33 -0000 (UTC), Mad Roger wrote:

What is a realistic accuracy & precision of typical MPG measurements when
measured by the consumer using the typical method of dividing their
tripmeter miles by the gas-pump gallons during fillup?

Close to 100% accuracy if done right. I've done it on long trips. But MPG will vary
depending on terrain, weather, wind direction, stop-and-go traffic, etc.
So if you want "true" MPG for your car, you have to do it for the life of the car.
Once you do it initially, it's kind of pointless to do again except to satisfy your
curiosity.

On Thu, 20 Jul 2017 12:08:33 -0000 (UTC), Mad Roger
wrote:

What is a realistic accuracy & precision of typical MPG measurements when
measured by the consumer using the typical method of dividing their
tripmeter miles by the gas-pump gallons during fillup?
Repeatabilty is terrible. Accuracy can be pretty good over multiple
tanks. Can be pretty good even on single tanks IF there is a way to
ensure the tank is always filled to EXACTLY the same point (like a
level in the fill - tube, with the vehicle parked at EXACTLY the same
place for each fill-up). Relying on the auto-shutoff of the pump can
cause variance of several liters per fillup.

On 7/20/2017 9:04 AM, wrote:
On Thu, 20 Jul 2017 12:08:33 -0000 (UTC), Mad Roger
wrote:

What is a realistic accuracy & precision of typical MPG measurements when
measured by the consumer using the typical method of dividing their
tripmeter miles by the gas-pump gallons during fillup?
Repeatabilty is terrible. Accuracy can be pretty good over multiple
tanks. Can be pretty good even on single tanks IF there is a way to
ensure the tank is always filled to EXACTLY the same point (like a
level in the fill - tube, with the vehicle parked at EXACTLY the same
place for each fill-up). Relying on the auto-shutoff of the pump can
cause variance of several liters per fillup.


My year old car gives constant mileage and appears very accurate when I
calculate based on fill-up. It even has a moving mileage meter going up
to 80 mpg when just coasting down hill or maybe 10 mpg going up hill.
I recall many times on long trips mileage varying all over the map
probably because I was filling up at different stations.

Mad Roger wrote:
What is a realistic accuracy & precision of typical MPG measurements when
measured by the consumer using the typical method of dividing their
tripmeter miles by the gas-pump gallons during fillup?

IF the odometer is accurate and you do the math out to the 10ths of a
gallon the pump shows it can be VERY accurate.

--
Steve W.

wrote:
On Thu, 20 Jul 2017 12:08:33 -0000 (UTC), Mad Roger
wrote:

What is a realistic accuracy & precision of typical MPG measurements when
measured by the consumer using the typical method of dividing their
tripmeter miles by the gas-pump gallons during fillup?
Repeatabilty is terrible. Accuracy can be pretty good over multiple
tanks. Can be pretty good even on single tanks IF there is a way to
ensure the tank is always filled to EXACTLY the same point (like a
level in the fill - tube, with the vehicle parked at EXACTLY the same
place for each fill-up). Relying on the auto-shutoff of the pump can
cause variance of several liters per fillup.

Vehicle position or the auto shut off point won't make any difference.
You read the amount of fuel you pumped off the pump itself. The only
real issue is odometer accuracy. That can vary with tire size variations
and factory calibration.

--
Steve W.

In article , says...

wrote:
On Thu, 20 Jul 2017 12:08:33 -0000 (UTC), Mad Roger
wrote:

What is a realistic accuracy & precision of typical MPG measurements when
measured by the consumer using the typical method of dividing their
tripmeter miles by the gas-pump gallons during fillup?
Repeatabilty is terrible. Accuracy can be pretty good over multiple
tanks. Can be pretty good even on single tanks IF there is a way to
ensure the tank is always filled to EXACTLY the same point (like a
level in the fill - tube, with the vehicle parked at EXACTLY the same
place for each fill-up). Relying on the auto-shutoff of the pump can
cause variance of several liters per fillup.

Vehicle position or the auto shut off point won't make any difference.
You read the amount of fuel you pumped off the pump itself. The only
real issue is odometer accuracy. That can vary with tire size variations
and factory calibration.

If you just do it one time, you can not be sure you put in the same
ammount of fuel that was taken out.

If you keep a running total of the ammount of miles and fuel over
several tank fulls , the ammount of fuel will sort of average out.

Say you park so the back of the car is up hill and you fill the tank.
Go a number of miles and fill up again. This time the back of the car
is down hill. You may burn out 15 gallons, but only put back in 14
gallons. Ot it could be the other way around and you burn 14 gallons,
but only put back in 13 gallons. From tank to tank full there could be
a large variation. Over many tanks, the variation will average out to a
lessor error.
After say 10 tanks used you only have to contend with one or two errors
caused by the exact ammount of fuel put in the tank. Probably just the
last tank full would be where the error would come in. So instead of 1
gallon of error like the example above, you would have about .1 gallon
of error if the pumps are correct, which they should be.

On Thursday, July 20, 2017 at 9:42:15 AM UTC-4, Steve W. wrote:
wrote:
On Thu, 20 Jul 2017 12:08:33 -0000 (UTC), Mad Roger
wrote:

What is a realistic accuracy & precision of typical MPG measurements when
measured by the consumer using the typical method of dividing their
tripmeter miles by the gas-pump gallons during fillup?
Repeatabilty is terrible. Accuracy can be pretty good over multiple
tanks. Can be pretty good even on single tanks IF there is a way to
ensure the tank is always filled to EXACTLY the same point (like a
level in the fill - tube, with the vehicle parked at EXACTLY the same
place for each fill-up). Relying on the auto-shutoff of the pump can
cause variance of several liters per fillup.

Vehicle position or the auto shut off point won't make any difference.
You read the amount of fuel you pumped off the pump itself. The only
real issue is odometer accuracy. That can vary with tire size variations
and factory calibration.

--
Steve W.

Of course vehicle position and shut off point makes a difference.
If you fill it up in a situation that winds up with more gas
in it, then fill it up after your run in a situation that winds
up with less gas in the tank, you think you used less gas than
you actually did and you get slightly higher MPG.

On Thu, 20 Jul 2017 10:57:06 -0400, Ralph Mowery wrote:

In article , says...

wrote:
On Thu, 20 Jul 2017 12:08:33 -0000 (UTC), Mad Roger
wrote:

What is a realistic accuracy & precision of typical MPG measurements when
measured by the consumer using the typical method of dividing their
tripmeter miles by the gas-pump gallons during fillup?
Repeatabilty is terrible. Accuracy can be pretty good over multiple
tanks. Can be pretty good even on single tanks IF there is a way to
ensure the tank is always filled to EXACTLY the same point (like a
level in the fill - tube, with the vehicle parked at EXACTLY the same
place for each fill-up). Relying on the auto-shutoff of the pump can
cause variance of several liters per fillup.

Vehicle position or the auto shut off point won't make any difference.
You read the amount of fuel you pumped off the pump itself. The only
real issue is odometer accuracy. That can vary with tire size variations
and factory calibration.

If you just do it one time, you can not be sure you put in the same
ammount of fuel that was taken out.

If you keep a running total of the ammount of miles and fuel over
several tank fulls , the ammount of fuel will sort of average out.

Say you park so the back of the car is up hill and you fill the tank.
Go a number of miles and fill up again. This time the back of the car
is down hill. You may burn out 15 gallons, but only put back in 14
gallons. Ot it could be the other way around and you burn 14 gallons,
but only put back in 13 gallons. From tank to tank full there could be
a large variation. Over many tanks, the variation will average out to a
lessor error.
After say 10 tanks used you only have to contend with one or two errors
caused by the exact ammount of fuel put in the tank. Probably just the
last tank full would be where the error would come in. So instead of 1
gallon of error like the example above, you would have about .1 gallon
of error if the pumps are correct, which they should be.


If you're getting MPG on a long trip, you only fill the tank all the way when you start and
when you finish the trip. All the other gas stops you go by the meter reading.
Very little room for error if you write down the meter reading.

On Thu, 20 Jul 2017 12:08:33 -0000 (UTC), Mad Roger
wrote:

What is a realistic accuracy & precision of typical MPG measurements when
measured by the consumer using the typical method of dividing their
tripmeter miles by the gas-pump gallons during fillup?


This thread reminded me to run-the-numbers from my car's log book.
I record each fill-up and a few times a year I'll plug the numbers
into a home-made spreadsheet.
I'm not sure if my car even has a built-in calculator - I'll have to
check .. my brother's Buick will always show about 1 mpg
better than reality.

in liters per 100 km my last 10 fill-ups are :
6.5 8.0 7.3 7.9 7.0 7.4 7.4 7.4 7.3 7.4 avg = 7.36

The adjacent high/low values indicate that the fill-up was
filled-to-the-brim or not-quite-full due to the pump sensivity.

Interesting to compare winter driving with winter tires :
8.3 8.7 8.2 7.9 7.7 8.0 8.5 8.3 8.0 8.0 avg = 8.16

2015 Kia Rondo 6 speed auto ; 2 liter gas ;
rated 10.6 / 7.6 city/highway at Natural Resources Canada
http://oee.nrcan.gc.ca/fcr-rcf/public/index-e.cfm

John T.

On Thursday, July 20, 2017 at 11:21:33 AM UTC-4, Vic Smith wrote:
On Thu, 20 Jul 2017 10:57:06 -0400, Ralph Mowery wrote:

In article , says...

wrote:
On Thu, 20 Jul 2017 12:08:33 -0000 (UTC), Mad Roger
wrote:

What is a realistic accuracy & precision of typical MPG measurements when
measured by the consumer using the typical method of dividing their
tripmeter miles by the gas-pump gallons during fillup?
Repeatabilty is terrible. Accuracy can be pretty good over multiple
tanks. Can be pretty good even on single tanks IF there is a way to
ensure the tank is always filled to EXACTLY the same point (like a
level in the fill - tube, with the vehicle parked at EXACTLY the same
place for each fill-up). Relying on the auto-shutoff of the pump can
cause variance of several liters per fillup.

Vehicle position or the auto shut off point won't make any difference.
You read the amount of fuel you pumped off the pump itself. The only
real issue is odometer accuracy. That can vary with tire size variations
and factory calibration.

If you just do it one time, you can not be sure you put in the same
ammount of fuel that was taken out.

If you keep a running total of the ammount of miles and fuel over
several tank fulls , the ammount of fuel will sort of average out.

Say you park so the back of the car is up hill and you fill the tank.
Go a number of miles and fill up again. This time the back of the car
is down hill. You may burn out 15 gallons, but only put back in 14
gallons. Ot it could be the other way around and you burn 14 gallons,
but only put back in 13 gallons. From tank to tank full there could be
a large variation. Over many tanks, the variation will average out to a
lessor error.
After say 10 tanks used you only have to contend with one or two errors
caused by the exact ammount of fuel put in the tank. Probably just the
last tank full would be where the error would come in. So instead of 1
gallon of error like the example above, you would have about .1 gallon
of error if the pumps are correct, which they should be.


If you're getting MPG on a long trip, you only fill the tank all the way when you start and
when you finish the trip. All the other gas stops you go by the meter reading.
Very little room for error if you write down the meter reading.

Agree. I was thinking of filling it full once.

Here's related question. Many new cars measure and show MPG.
How do they measure how much fuel is being used? It's not
via the tank gauge, because it can change almost instantly.
Do they have something that measure fuel flow directly? Or do they
do it by knowing how long the injectors are open and using that?
Anyone know?

In article , says...

What is a realistic accuracy & precision of typical MPG measurements when
measured by the consumer using the typical method of dividing their
tripmeter miles by the gas-pump gallons during fillup?

Trip meter miles depends on circumference of driving wheels. I know my
speedo closely matches readings of roadside radar displays or my GPS, so
I guess trip meter miles will be accurate too.

On 7/20/2017 12:30 PM, trader_4 wrote:



If you're getting MPG on a long trip, you only fill the tank all the way when you start and
when you finish the trip. All the other gas stops you go by the meter reading.
Very little room for error if you write down the meter reading.

Agree. I was thinking of filling it full once.

Here's related question. Many new cars measure and show MPG.
How do they measure how much fuel is being used? It's not
via the tank gauge, because it can change almost instantly.
Do they have something that measure fuel flow directly? Or do they
do it by knowing how long the injectors are open and using that?
Anyone know?


Based on my checking over a tankful and what the meter says, they just
take an optimistic guess. It is always higher, but there seems to be an
explanation

https://www.edmunds.com/fuel-economy...s-fibbing.html
Roger Clark, senior manager of GM's energy center, explains that the
fuel economy gauge makes a calculation by counting the number and
duration of pulses made by the fuel injectors as they squirt gasoline
into the combustion chambers of the engine. The onboard computer system
divides the distance the car travels by this estimated fuel consumption.

Clark says the gauge is "dead nuts accurate" €” if you consider all the
variables at work during driving, including temperature, driving
conditions and driving style. The biggest fluctuation occurs because
ethanol, which is blended with gasoline in varying amounts, contains
less energy.

"When you fill up, you are paying for a gallon of gas, but the energy in
that gas varies significantly," Clark says. This means that while the
car's computer assumes the gasoline is providing energy to drive a
certain distance, the fuel might have less energy and not propel the car
as far.

On 07/20/2017 11:59 AM, Ed Pawlowski wrote:
....

Based on my checking over a tankful and what the meter says, they just
take an optimistic guess. It is always higher, but there seems to be an
explanation

https://www.edmunds.com/fuel-economy...s-fibbing.html

Roger Clark, senior manager of GM's energy center, explains that the
fuel economy gauge makes a calculation by counting the number and
duration of pulses made by the fuel injectors as they squirt gasoline
into the combustion chambers of the engine. The onboard computer system
divides the distance the car travels by this estimated fuel consumption.

Clark says the gauge is "dead nuts accurate" €” if you consider all the
variables at work during driving, including temperature, driving
conditions and driving style. The biggest fluctuation occurs because
ethanol, which is blended with gasoline in varying amounts, contains
less energy.

"When you fill up, you are paying for a gallon of gas, but the energy in
that gas varies significantly," Clark says. This means that while the
car's computer assumes the gasoline is providing energy to drive a
certain distance, the fuel might have less energy and not propel the car
as far.

I think that's hogwash, too. The mpg is volume/distance; the
computation is independent of the energy content; simply a higher
consumption rate/lower mpg value will be computed.

_Unless_ they don't use actual odometer reading and distance is imputed
from other measurements at least one of which is based on engine
efficiency under given conditions.

I don't have inside info on the precise algorithm, I'd always figured it
was, in fact, simply the instantaneous volume*t integrated and the
accumulated is simply the ratio of the two sums (integrals) of fuel
volume over total distance since totals were last reset.

The Buick longer-term values have always worked out to be pretty-much
identical to logbook records on long trips for several different models
from the old LeSabres to the (relatively) new Enclave and Lucerne even
though they're now getting long in the tooth by car sales standards...

--

On Thursday, July 20, 2017 at 2:29:01 PM UTC-4, dpb wrote:
On 07/20/2017 11:59 AM, Ed Pawlowski wrote:
...

Based on my checking over a tankful and what the meter says, they just
take an optimistic guess. It is always higher, but there seems to be an
explanation

https://www.edmunds.com/fuel-economy...-fibbing..html

Roger Clark, senior manager of GM's energy center, explains that the
fuel economy gauge makes a calculation by counting the number and
duration of pulses made by the fuel injectors as they squirt gasoline
into the combustion chambers of the engine. The onboard computer system
divides the distance the car travels by this estimated fuel consumption..

Clark says the gauge is "dead nuts accurate" €” if you consider all the
variables at work during driving, including temperature, driving
conditions and driving style. The biggest fluctuation occurs because
ethanol, which is blended with gasoline in varying amounts, contains
less energy.

"When you fill up, you are paying for a gallon of gas, but the energy in
that gas varies significantly," Clark says. This means that while the
car's computer assumes the gasoline is providing energy to drive a
certain distance, the fuel might have less energy and not propel the car
as far.

I think that's hogwash, too. The mpg is volume/distance; the
computation is independent of the energy content; simply a higher
consumption rate/lower mpg value will be computed.


Why is it hogwash? They seem to be saying the same thing you are,
that the energy content means that what you see as MPG will
vary because of it. If you have gas with higher energy content,
then you'll see 25 mpg. If it's lower energy content, you'll
see 23 MPG. How much the energy content varies, when it's
labeled as the same product, eg reg unleaded, IDK. It would
obviously vary depending on the amount of ethanol added.

On 7/20/2017 2:28 PM, dpb wrote:
On 07/20/2017 11:59 AM, Ed Pawlowski wrote:
...

Based on my checking over a tankful and what the meter says, they just
take an optimistic guess. It is always higher, but there seems to be an
explanation

https://www.edmunds.com/fuel-economy...s-fibbing.html


Roger Clark, senior manager of GM's energy center, explains that the
fuel economy gauge makes a calculation by counting the number and
duration of pulses made by the fuel injectors as they squirt gasoline
into the combustion chambers of the engine. The onboard computer system
divides the distance the car travels by this estimated fuel consumption.

Clark says the gauge is "dead nuts accurate" €” if you consider all the
variables at work during driving, including temperature, driving
conditions and driving style. The biggest fluctuation occurs because
ethanol, which is blended with gasoline in varying amounts, contains
less energy.

"When you fill up, you are paying for a gallon of gas, but the energy in
that gas varies significantly," Clark says. This means that while the
car's computer assumes the gasoline is providing energy to drive a
certain distance, the fuel might have less energy and not propel the car
as far.

I think that's hogwash, too. The mpg is volume/distance; the
computation is independent of the energy content; simply a higher
consumption rate/lower mpg value will be computed.

_Unless_ they don't use actual odometer reading and distance is imputed
from other measurements at least one of which is based on engine
efficiency under given conditions.

I don't have inside info on the precise algorithm, I'd always figured it
was, in fact, simply the instantaneous volume*t integrated and the
accumulated is simply the ratio of the two sums (integrals) of fuel
volume over total distance since totals were last reset.

Seems like a computation based on a given fuel energy value rather than
just volume. An accurate measure of volume consumed versus miles driven
would be a consistent and accurate readout. Evidently they all use the
same method as I always got high readings.



The Buick longer-term values have always worked out to be pretty-much
identical to logbook records on long trips for several different models
from the old LeSabres to the (relatively) new Enclave and Lucerne even
though they're now getting long in the tooth by car sales standards...


My Buicks were always 2 or 3 mpg high. I've not driven a Buick since
that last one fell apart. I used to be a fan of GM until they screwed
me. I should have wised up sooner.

On 7/20/2017 3:06 PM, trader_4 wrote:

Why is it hogwash? They seem to be saying the same thing you are,
that the energy content means that what you see as MPG will
vary because of it. If you have gas with higher energy content,
then you'll see 25 mpg. If it's lower energy content, you'll
see 23 MPG. How much the energy content varies, when it's
labeled as the same product, eg reg unleaded, IDK. It would
obviously vary depending on the amount of ethanol added.



Seems to be some confusion. Sure, the pure gas will give you more
distance per gallon. The computer readout though will say the same for
pure gas, gas with ethanol, or a bottle of bourbon.

The readout on the computer still says 25 even though you only get 23.
Sort of like saying if you had pure gas this is what you;d have gotten
but you got something else.

On Thursday, July 20, 2017 at 3:47:40 PM UTC-4, Ed Pawlowski wrote:
On 7/20/2017 3:06 PM, trader_4 wrote:

Why is it hogwash? They seem to be saying the same thing you are,
that the energy content means that what you see as MPG will
vary because of it. If you have gas with higher energy content,
then you'll see 25 mpg. If it's lower energy content, you'll
see 23 MPG. How much the energy content varies, when it's
labeled as the same product, eg reg unleaded, IDK. It would
obviously vary depending on the amount of ethanol added.



Seems to be some confusion. Sure, the pure gas will give you more
distance per gallon. The computer readout though will say the same for
pure gas, gas with ethanol, or a bottle of bourbon.

The readout on the computer still says 25 even though you only get 23.
Sort of like saying if you had pure gas this is what you;d have gotten
but you got something else.

I don't see how the computer readout will still say 25 even though
you get 23. Like DPB said, the car computer is simply computing
miles traveled by volume of fuel injected. With the lower energy
content gas it will have to use more gas to go the same distance
and the number calculated will show it. Instead of 25 mpg, you'll
get 23. Take it to the limits. Suppose you had gas that had very
little energy content and a gallon could only propel the car 1 mile.
What would it show? 1 MPG or 25 mpg?

On Thursday, July 20, 2017 at 3:47:40 PM UTC-4, Ed Pawlowski wrote:
On 7/20/2017 3:06 PM, trader_4 wrote:

Why is it hogwash? They seem to be saying the same thing you are,
that the energy content means that what you see as MPG will
vary because of it. If you have gas with higher energy content,
then you'll see 25 mpg. If it's lower energy content, you'll
see 23 MPG. How much the energy content varies, when it's
labeled as the same product, eg reg unleaded, IDK. It would
obviously vary depending on the amount of ethanol added.



Seems to be some confusion. Sure, the pure gas will give you more
distance per gallon. The computer readout though will say the same for
pure gas, gas with ethanol, or a bottle of bourbon.

No it won't. Put very low energy fuel in the car, one that can only
propel the car a mile with a gallon of it. What will the car MPG
computer show? In my world, it's computing how much volume of fuel
it injected and how many miles it went. 1 gal, 1 mile, it's 1 MPG.



The readout on the computer still says 25 even though you only get 23.
Sort of like saying if you had pure gas this is what you;d have gotten
but you got something else.

Not if it's doing MPG the rational way, which, like DPB said is
simply distance travelled and gallons injected. It's not
calculating BTUs injected, at least not from the description
given, nor is there any reason to factor in energy content,
since the car doesn't know what it is.

On 7/20/2017 3:47 PM, Ed Pawlowski wrote:
On 7/20/2017 3:06 PM, trader_4 wrote:

Why is it hogwash? They seem to be saying the same thing you are,
that the energy content means that what you see as MPG will
vary because of it. If you have gas with higher energy content,
then you'll see 25 mpg. If it's lower energy content, you'll
see 23 MPG. How much the energy content varies, when it's
labeled as the same product, eg reg unleaded, IDK. It would
obviously vary depending on the amount of ethanol added.



Seems to be some confusion. Sure, the pure gas will give you more
distance per gallon. The computer readout though will say the same for
pure gas, gas with ethanol, or a bottle of bourbon.

The readout on the computer still says 25 even though you only get 23.
Sort of like saying if you had pure gas this is what you;d have gotten
but you got something else.

This is interesting:

http://newsroom.aaa.com/2013/06/what...lend-gasoline/

It seems to me that I am getting slightly lower mileage this summer but
maybe because it has been very hot and AC is on all the time.

I prefer to store summer gas for my generator because winter gas puffs
up the sealed plastic cans in summer heat.

On Thursday, July 20, 2017 at 3:34:32 PM UTC-4, Ed Pawlowski wrote:
On 7/20/2017 2:28 PM, dpb wrote:
On 07/20/2017 11:59 AM, Ed Pawlowski wrote:
...

Based on my checking over a tankful and what the meter says, they just
take an optimistic guess. It is always higher, but there seems to be an
explanation

https://www.edmunds.com/fuel-economy...s-fibbing.html


Roger Clark, senior manager of GM's energy center, explains that the
fuel economy gauge makes a calculation by counting the number and
duration of pulses made by the fuel injectors as they squirt gasoline
into the combustion chambers of the engine. The onboard computer system
divides the distance the car travels by this estimated fuel consumption.

On 07/20/2017 2:06 PM, trader_4 wrote:
....

I think that's hogwash, too. The mpg is volume/distance; the
computation is independent of the energy content; simply a higher
consumption rate/lower mpg value will be computed.


Why is it hogwash? They seem to be saying the same thing you are,
that the energy content means that what you see as MPG will
vary because of it. If you have gas with higher energy content,
then you'll see 25 mpg. If it's lower energy content, you'll
see 23 MPG. How much the energy content varies, when it's
labeled as the same product, eg reg unleaded, IDK. It would
obviously vary depending on the amount of ethanol added.

Maybe he meant that; wasn't exactly what I thought was trying to say...

I thought he was saying the algorithm was _imputing_ miles travel based
on an assumed reference energy content; if he was instead just trying to
explain that the energy content of the fuel on a volumetric basis has an
effect on the total energy in a gallon of that same and hence how much
one gets out of a gallon, then we are in agreement.

And, of course, the mpg is the inverse of the ratio as I wrote it
earlier... (I was going to talk about distance per unit input but
decided it would get too longwinded and didn't fix all the initial to
match the final.)

--

On Thursday, July 20, 2017 at 4:31:25 PM UTC-4, dpb wrote:
On 07/20/2017 2:06 PM, trader_4 wrote:
...

I think that's hogwash, too. The mpg is volume/distance; the
computation is independent of the energy content; simply a higher
consumption rate/lower mpg value will be computed.


Why is it hogwash? They seem to be saying the same thing you are,
that the energy content means that what you see as MPG will
vary because of it. If you have gas with higher energy content,
then you'll see 25 mpg. If it's lower energy content, you'll
see 23 MPG. How much the energy content varies, when it's
labeled as the same product, eg reg unleaded, IDK. It would
obviously vary depending on the amount of ethanol added.

Maybe he meant that; wasn't exactly what I thought was trying to say...

I thought he was saying the algorithm was _imputing_ miles travel based
on an assumed reference energy content; if he was instead just trying to
explain that the energy content of the fuel on a volumetric basis has an
effect on the total energy in a gallon of that same and hence how much
one gets out of a gallon, then we are in agreement.

And, of course, the mpg is the inverse of the ratio as I wrote it
earlier... (I was going to talk about distance per unit input but
decided it would get too longwinded and didn't fix all the initial to
match the final.)

--

Until proven otherwise, I'd say what he's saying there is what you
say and what I would expect, that it just uses volume of fuel
injected and miles traveled.

On Thu, 20 Jul 2017 09:42:10 -0400, "Steve W."
wrote:

wrote:
On Thu, 20 Jul 2017 12:08:33 -0000 (UTC), Mad Roger
wrote:

What is a realistic accuracy & precision of typical MPG measurements when
measured by the consumer using the typical method of dividing their
tripmeter miles by the gas-pump gallons during fillup?
Repeatabilty is terrible. Accuracy can be pretty good over multiple
tanks. Can be pretty good even on single tanks IF there is a way to
ensure the tank is always filled to EXACTLY the same point (like a
level in the fill - tube, with the vehicle parked at EXACTLY the same
place for each fill-up). Relying on the auto-shutoff of the pump can
cause variance of several liters per fillup.

Vehicle position or the auto shut off point won't make any difference.
You read the amount of fuel you pumped off the pump itself. The only
real issue is odometer accuracy. That can vary with tire size variations
and factory calibration.
No, steve. You are wrong. The amount of fuel you put in is the amount
you can squeeze into the empty portion of the tank. The amount you
used is the amount that used to be in the tank. You need to fill it to
the exact same point each time to get an accurate reading. You may
have filled your 72 liter tank to only 71 liters the last time you put
in 50 liters to fill the tank. Now, at a different station, with
different levels, you may squeeze in 73, or only 68. COSISTANCY is the
key - and where most will fall down, because, like you, they just
don't REALLY understand.

On Thu, 20 Jul 2017 10:57:06 -0400, Ralph Mowery
wrote:

In article , says...

wrote:
On Thu, 20 Jul 2017 12:08:33 -0000 (UTC), Mad Roger
wrote:

What is a realistic accuracy & precision of typical MPG measurements when
measured by the consumer using the typical method of dividing their
tripmeter miles by the gas-pump gallons during fillup?
Repeatabilty is terrible. Accuracy can be pretty good over multiple
tanks. Can be pretty good even on single tanks IF there is a way to
ensure the tank is always filled to EXACTLY the same point (like a
level in the fill - tube, with the vehicle parked at EXACTLY the same
place for each fill-up). Relying on the auto-shutoff of the pump can
cause variance of several liters per fillup.

Vehicle position or the auto shut off point won't make any difference.
You read the amount of fuel you pumped off the pump itself. The only
real issue is odometer accuracy. That can vary with tire size variations
and factory calibration.

If you just do it one time, you can not be sure you put in the same
ammount of fuel that was taken out.

If you keep a running total of the ammount of miles and fuel over
several tank fulls , the ammount of fuel will sort of average out.

Say you park so the back of the car is up hill and you fill the tank.
Go a number of miles and fill up again. This time the back of the car
is down hill. You may burn out 15 gallons, but only put back in 14
gallons. Ot it could be the other way around and you burn 14 gallons,
but only put back in 13 gallons. From tank to tank full there could be
a large variation. Over many tanks, the variation will average out to a
lessor error.
After say 10 tanks used you only have to contend with one or two errors
caused by the exact ammount of fuel put in the tank. Probably just the
last tank full would be where the error would come in. So instead of 1
gallon of error like the example above, you would have about .1 gallon
of error if the pumps are correct, which they should be.

If the first and last are identical, none of the others matter. The
difference of 1, or 10 liters spread across many tanks becomes , more
or less, just noise. On the short term, like 1 tank, it can be a
pretty large percentage of error.

On 07/20/2017 3:32 PM, trader_4 wrote:
....

Until proven otherwise, I'd say what he's saying there is what you
say and what I would expect, that it just uses volume of fuel
injected and miles traveled.

That's certainly what the Buick displays indicate--if you reset then the
initial instantaneous and average are identical and you can watch the
average settle down as each instantaneous reading is added...

In a uni graduate seminar/project class we had access to firmware from
GM in order to build competition vehicles. But, being of the age I am,
this was long before the "veritable plethora" explosion of
microprocessors and the whole system was in one little M68HC11 and none
of the display functions currently were yet extant...

--

On Thu, 20 Jul 2017 11:58:39 -0400, wrote:

On Thu, 20 Jul 2017 12:08:33 -0000 (UTC), Mad Roger
wrote:

What is a realistic accuracy & precision of typical MPG measurements when
measured by the consumer using the typical method of dividing their
tripmeter miles by the gas-pump gallons during fillup?


This thread reminded me to run-the-numbers from my car's log book.
I record each fill-up and a few times a year I'll plug the numbers
into a home-made spreadsheet.
I'm not sure if my car even has a built-in calculator - I'll have to
check .. my brother's Buick will always show about 1 mpg
better than reality.

in liters per 100 km my last 10 fill-ups are :
6.5 8.0 7.3 7.9 7.0 7.4 7.4 7.4 7.3 7.4 avg = 7.36

The adjacent high/low values indicate that the fill-up was
filled-to-the-brim or not-quite-full due to the pump sensivity.

Interesting to compare winter driving with winter tires :
8.3 8.7 8.2 7.9 7.7 8.0 8.5 8.3 8.0 8.0 avg = 8.16

2015 Kia Rondo 6 speed auto ; 2 liter gas ;
rated 10.6 / 7.6 city/highway at Natural Resources Canada
http://oee.nrcan.gc.ca/fcr-rcf/public/index-e.cfm

John T.
That is a combination of snow tires, sloppy roads and cold weather -
not just tires

On Thu, 20 Jul 2017 12:59:17 -0400, Ed Pawlowski wrote:

On 7/20/2017 12:30 PM, trader_4 wrote:



If you're getting MPG on a long trip, you only fill the tank all the way when you start and
when you finish the trip. All the other gas stops you go by the meter reading.
Very little room for error if you write down the meter reading.

Agree. I was thinking of filling it full once.

Here's related question. Many new cars measure and show MPG.
How do they measure how much fuel is being used? It's not
via the tank gauge, because it can change almost instantly.
Do they have something that measure fuel flow directly? Or do they
do it by knowing how long the injectors are open and using that?
Anyone know?


Based on my checking over a tankful and what the meter says, they just
take an optimistic guess. It is always higher, but there seems to be an
explanation

https://www.edmunds.com/fuel-economy...s-fibbing.html
Roger Clark, senior manager of GM's energy center, explains that the
fuel economy gauge makes a calculation by counting the number and
duration of pulses made by the fuel injectors as they squirt gasoline
into the combustion chambers of the engine. The onboard computer system
divides the distance the car travels by this estimated fuel consumption.

Clark says the gauge is "dead nuts accurate" — if you consider all the
variables at work during driving, including temperature, driving
conditions and driving style. The biggest fluctuation occurs because
ethanol, which is blended with gasoline in varying amounts, contains
less energy.

"When you fill up, you are paying for a gallon of gas, but the energy in
that gas varies significantly," Clark says. This means that while the
car's computer assumes the gasoline is providing energy to drive a
certain distance, the fuel might have less energy and not propel the car
as far.
Roger clark forgets ONE major thing. The computer knows how much
FUEL is used - the energy density of the fuel doesn't enter into the
calculations at all - nor should it. You are measuring how much "fuel"
is burned - doesn't matter if it is gasoline or hooch. The computer
ASSUMES nothing. It calculates the volume by a combination of the
length of the injection pulse and the number of pulses, and the
distance travelled by the vss signal (which also provides data to the
speedo and odo)

On Thu, 20 Jul 2017 17:22:06 -0400, wrote:

On Thu, 20 Jul 2017 11:58:39 -0400, wrote:

On Thu, 20 Jul 2017 12:08:33 -0000 (UTC), Mad Roger
wrote:

What is a realistic accuracy & precision of typical MPG measurements when
measured by the consumer using the typical method of dividing their
tripmeter miles by the gas-pump gallons during fillup?


This thread reminded me to run-the-numbers from my car's log book.
I record each fill-up and a few times a year I'll plug the numbers
into a home-made spreadsheet.
I'm not sure if my car even has a built-in calculator - I'll have to
check .. my brother's Buick will always show about 1 mpg
better than reality.
in liters per 100 km my last 10 fill-ups are :
6.5 8.0 7.3 7.9 7.0 7.4 7.4 7.4 7.3 7.4 avg = 7.36
The adjacent high/low values indicate that the fill-up was
filled-to-the-brim or not-quite-full due to the pump sensivity.
Interesting to compare winter driving with winter tires :
8.3 8.7 8.2 7.9 7.7 8.0 8.5 8.3 8.0 8.0 avg = 8.16
2015 Kia Rondo 6 speed auto ; 2 liter gas ;
rated 10.6 / 7.6 city/highway at Natural Resources Canada
http://oee.nrcan.gc.ca/fcr-rcf/public/index-e.cfm
John T.


That is a combination of snow tires, sloppy roads and cold weather -
not just tires


Yep - warming up in the driveway while I scrape the windows ..
also I'll run the winter tires just a pound or two soft.
John T.

On 7/20/2017 5:15 PM, wrote:
On Thu, 20 Jul 2017 09:42:10 -0400, "Steve W."
wrote:

wrote:
On Thu, 20 Jul 2017 12:08:33 -0000 (UTC), Mad Roger
wrote:

What is a realistic accuracy & precision of typical MPG measurements when
measured by the consumer using the typical method of dividing their
tripmeter miles by the gas-pump gallons during fillup?
Repeatabilty is terrible. Accuracy can be pretty good over multiple
tanks. Can be pretty good even on single tanks IF there is a way to
ensure the tank is always filled to EXACTLY the same point (like a
level in the fill - tube, with the vehicle parked at EXACTLY the same
place for each fill-up). Relying on the auto-shutoff of the pump can
cause variance of several liters per fillup.

Vehicle position or the auto shut off point won't make any difference.
You read the amount of fuel you pumped off the pump itself. The only
real issue is odometer accuracy. That can vary with tire size variations
and factory calibration.
No, steve. You are wrong. The amount of fuel you put in is the amount
you can squeeze into the empty portion of the tank. The amount you
used is the amount that used to be in the tank. You need to fill it to
the exact same point each time to get an accurate reading. You may
have filled your 72 liter tank to only 71 liters the last time you put
in 50 liters to fill the tank. Now, at a different station, with
different levels, you may squeeze in 73, or only 68. COSISTANCY is the
key - and where most will fall down, because, like you, they just
don't REALLY understand.


You buy gas by the gallon and mileage is miles per gallon. Summer gas
has higher density so you get more gas by weight for your money but it
costs more per gallon in the summer. Sounds like you can't win.

On Thu, 20 Jul 2017 13:28:54 -0500, dpb wrote:

On 07/20/2017 11:59 AM, Ed Pawlowski wrote:
...

Based on my checking over a tankful and what the meter says, they just
take an optimistic guess. It is always higher, but there seems to be an
explanation

https://www.edmunds.com/fuel-economy...s-fibbing.html

Roger Clark, senior manager of GM's energy center, explains that the
fuel economy gauge makes a calculation by counting the number and
duration of pulses made by the fuel injectors as they squirt gasoline
into the combustion chambers of the engine. The onboard computer system
divides the distance the car travels by this estimated fuel consumption.

Clark says the gauge is "dead nuts accurate" — if you consider all the
variables at work during driving, including temperature, driving
conditions and driving style. The biggest fluctuation occurs because
ethanol, which is blended with gasoline in varying amounts, contains
less energy.

"When you fill up, you are paying for a gallon of gas, but the energy in
that gas varies significantly," Clark says. This means that while the
car's computer assumes the gasoline is providing energy to drive a
certain distance, the fuel might have less energy and not propel the car
as far.

I think that's hogwash, too. The mpg is volume/distance; the
computation is independent of the energy content; simply a higher
consumption rate/lower mpg value will be computed.

_Unless_ they don't use actual odometer reading and distance is imputed
from other measurements at least one of which is based on engine
efficiency under given conditions.

If they base their fuel consumption on exhaust gas analysis, rather
than actual fuel measurement he is correct. I believe that is how the
EPA rest works (and why it is NEVER accurate)
I don't have inside info on the precise algorithm, I'd always figured it
was, in fact, simply the instantaneous volume*t integrated and the
accumulated is simply the ratio of the two sums (integrals) of fuel
volume over total distance since totals were last reset.

The Buick longer-term values have always worked out to be pretty-much
identical to logbook records on long trips for several different models
from the old LeSabres to the (relatively) new Enclave and Lucerne even
though they're now getting long in the tooth by car sales standards...

On 7/20/2017 3:52 PM, trader_4 wrote:


The readout on the computer still says 25 even though you only get 23.
Sort of like saying if you had pure gas this is what you;d have gotten
but you got something else.

I don't see how the computer readout will still say 25 even though
you get 23. Like DPB said, the car computer is simply computing
miles traveled by volume of fuel injected. With the lower energy
content gas it will have to use more gas to go the same distance
and the number calculated will show it. Instead of 25 mpg, you'll
get 23. Take it to the limits. Suppose you had gas that had very
little energy content and a gallon could only propel the car 1 mile.
What would it show? 1 MPG or 25 mpg?


The translation I get from GM is that it will read 25. Makes no sense
to consider energy per gallon but it seems they do. I look at it as a
rough guide, nothing more. When I turn my car off I get a reading from
the last trip. Highway will be in the 25 range, city will be in the 18
range.

On 7/20/2017 3:56 PM, trader_4 wrote:

The readout on the computer still says 25 even though you only get 23.
Sort of like saying if you had pure gas this is what you;d have gotten
but you got something else.

Not if it's doing MPG the rational way, which, like DPB said is
simply distance travelled and gallons injected. It's not
calculating BTUs injected, at least not from the description
given, nor is there any reason to factor in energy content,
since the car doesn't know what it is.


Seems GM is not rational. In any case, it is always optimistic in my
experience.

On Thu, 20 Jul 2017 18:08:32 -0400, wrote:

On Thu, 20 Jul 2017 17:22:06 -0400, wrote:

On Thu, 20 Jul 2017 11:58:39 -0400, wrote:

On Thu, 20 Jul 2017 12:08:33 -0000 (UTC), Mad Roger
wrote:

What is a realistic accuracy & precision of typical MPG measurements when
measured by the consumer using the typical method of dividing their
tripmeter miles by the gas-pump gallons during fillup?


This thread reminded me to run-the-numbers from my car's log book.
I record each fill-up and a few times a year I'll plug the numbers
into a home-made spreadsheet.
I'm not sure if my car even has a built-in calculator - I'll have to
check .. my brother's Buick will always show about 1 mpg
better than reality.
in liters per 100 km my last 10 fill-ups are :
6.5 8.0 7.3 7.9 7.0 7.4 7.4 7.4 7.3 7.4 avg = 7.36
The adjacent high/low values indicate that the fill-up was
filled-to-the-brim or not-quite-full due to the pump sensivity.
Interesting to compare winter driving with winter tires :
8.3 8.7 8.2 7.9 7.7 8.0 8.5 8.3 8.0 8.0 avg = 8.16
2015 Kia Rondo 6 speed auto ; 2 liter gas ;
rated 10.6 / 7.6 city/highway at Natural Resources Canada
http://oee.nrcan.gc.ca/fcr-rcf/public/index-e.cfm
John T.


That is a combination of snow tires, sloppy roads and cold weather -
not just tires


Yep - warming up in the driveway while I scrape the windows ..
also I'll run the winter tires just a pound or two soft.
John T.
and the engine runs richer with low intake air temps too. Amd the
bearings and gears are stiffer in the cold, and you are pushing slush
aside or crawling over snow - all of which takes more power and fuel.

On Thursday, July 20, 2017 at 7:02:18 PM UTC-4, Ed Pawlowski wrote:
On 7/20/2017 3:56 PM, trader_4 wrote:

The readout on the computer still says 25 even though you only get 23.
Sort of like saying if you had pure gas this is what you;d have gotten
but you got something else.

Not if it's doing MPG the rational way, which, like DPB said is
simply distance travelled and gallons injected. It's not
calculating BTUs injected, at least not from the description
given, nor is there any reason to factor in energy content,
since the car doesn't know what it is.


Seems GM is not rational. In any case, it is always optimistic in my
experience.

GM says this:

Roger Clark, senior manager of GM's energy center, explains that the
fuel economy gauge makes a calculation by counting the number and
duration of pulses made by the fuel injectors as they squirt gasoline
into the combustion chambers of the engine. The onboard computer system
divides the distance the car travels by this estimated fuel consumption.


I don't see how it could be any clearer that they are simply
computing based on volume of gas used and distance traveled.

The latter part, where they talk about the fact that the
energy content of the gas makes a difference in MPG doesn't
change the fact that it's calculated the way you'd expect.

On 07/20/2017 10:59 AM, Ed Pawlowski wrote:
Clark says the gauge is "dead nuts accurate" €” if you consider all the
variables at work during driving, including temperature, driving
conditions and driving style. The biggest fluctuation occurs because
ethanol, which is blended with gasoline in varying amounts, contains
less energy.

We have mandated ethanol blended gasoline in the winter months. It's
good for a couple of mpgs. That's in early November before warming the
car up in below zero weather and running the heater full tilt happens.
That causes another drop.

On 07/20/2017 01:47 PM, Ed Pawlowski wrote:
Seems to be some confusion. Sure, the pure gas will give you more
distance per gallon. The computer readout though will say the same for
pure gas, gas with ethanol, or a bottle of bourbon.

Then they need to hire a decent programmer. Integrating the volume
delivered per injector pulse should give you the total volume. There
might be a slight correction for the temperature of the fluid.

On 07/20/2017 03:22 PM, wrote:
That is a combination of snow tires, sloppy roads and cold weather -
not just tires

And if you're like me, the car idling in the driveway until the ice
melts off the windows and the interior temperature is somewhere north of
zero (F).

On Thu, 20 Jul 2017 17:32:39 +0100,
Mike Coon wrote:

What is a realistic accuracy & precision of typical MPG measurements when
measured by the consumer using the typical method of dividing their
tripmeter miles by the gas-pump gallons during fillup?

Trip meter miles depends on circumference of driving wheels. I know my
speedo closely matches readings of roadside radar displays or my GPS, so
I guess trip meter miles will be accurate too.

Every reading a mom and pop does has inaccuracies that, I posit, are
tremendously higher than most people seem to think they are (at least most
people who quote mpg figures with decimal places in them).

Most people have a tripmeter reading and a gas pumpmeter reading.
Where they fill the tank and reset the tripmeter before driving away.

I can't find any reliable source that says what the accuracy or
repeatability of that mom-and-pop tripmeter/pumpmeter calculation, but
basic logic dictates that the errors compound such that there is likely
(IMHO) no way to get anywhere near decimal-point accuracy, and worse,
probably plus or minus 1 mpg is the closest anyone can get in terms of
repeatability and precision.

Even the EPA's $360,000 machine only claims plus or minus 2% of the
indicared reading. I can't find where I got the notion that a mom and pop
can't possibly get closer than about 4% with a tripmeter/pumpmeter mpg
calculation - but I'm still seeking those numbers as we speak.

On 07/21/2017 11:51 AM, Mad Roger wrote:
....

Most people have a tripmeter reading and a gas pumpmeter reading.
Where they fill the tank and reset the tripmeter before driving away.

I can't find any reliable source that says what the accuracy or
repeatability of that mom-and-pop tripmeter/pumpmeter calculation, but
basic logic dictates that the errors compound such that there is likely
(IMHO) no way to get anywhere near decimal-point accuracy, and worse,
probably plus or minus 1 mpg is the closest anyone can get in terms of
repeatability and precision.
....

Why do errors compound in your view?

And, it depends on what you mean in terms of accuracy -- in terms of
absolute one needs to know the calibration error of the odometer; most
folks are satisfied to just assume it's close enough for the purpose.

If you look at simply a single fillup, it's not unreasonable to expect a
few tenths of a gallon difference between the first fillup level and the
subsequent; if you try it on shorter distances than a full tank then the
fractional error goes up.

OTOH, if one keeps track over longer periods of multiple fillups and
take some care to use the same filling pattern and only fills up after
using near the full tank capacity, then over time plus/minus targets
_will_ tend to cancel out and I have no qualms in believing a relative
performance number in the 0.1 mpg can be determined.

As noted, I've done this on long trips a number of times (generally on
first trip or so with a new vehicle, either actually new (rare) or (most
often) new to me) just to see how it compared with previous and have had
quite good comparisons on recent ones with the computer-computed
results. These would be over total distances of 1500 to 2000 miles, not
just 20 miles test runs.

--