Gas mileage

[mjhpadi]

if that is the case. Are these pumps clicking off at 15 gals? I fill mine til it clicks off the 1st time, read that over filling would damage evap system. But I have notice even though my gauge says full it moves off very quickly.

I agree be careful with overfilling...I had that happen on my F-150. Dealer and Ford believes that overfilling caused problems with the evap canister, although changing it out and clearing codes would not eliminate warning light, would always come back on within 100 miles, no one could figure out how to eliminate problem...truck ran fine, just couldn't get "Check Engine" light to stay out. So with the SHO I've only put about .5-.75 gallon in after pump shuts off. Then only if I'm going to drive a bit.

 

[dramsey]

I'm surprised that people think that trying to develop a consistent and repeatable technique to fill the tank as far as possible will somehow enable them to derive a more accurate mileage figure than the SHO's computer. The mileage figure shown isn't an "estimate" (distance to empty, now that's an estimate), but is computed directly from the amount of fuel used to go the distance travelled since the last reset.

I don't know how accurate the fuel-measuring system is, but given that the direct injection system requires ultra-precise fuel delivery, I'd imagine it's pretty accurate. In any case I'd reckon it's way more accurate than any figure you'd derive from trying to fill the tank to the exact same level every time, using different fuel pumps with different shutoff mechanisms...

 

[Showgun]

I agree be careful with overfilling...I had that happen on my F-150. Dealer and Ford believes that overfilling caused problems with the evap canister, although changing it out and clearing codes would not eliminate warning light, would always come back on within 100 miles, no one could figure out how to eliminate problem...truck ran fine, just couldn't get "Check Engine" light to stay out. So with the SHO I've only put about .5-.75 gallon in after pump shuts off. Then only if I'm going to drive a bit.

That's a good point, Mark, and one which I had forgotten: Overfilling can cause issues which will become evident with drivability and certain warning lights.
Yes the excess gas can come back up the evap. piping to the canister and into the engine causing surges and rough idle.
I've had this happen, tho not in recent cars thankfully. Anyone filling their cars to the very top should take note of the possibility of this happening.
I believe it warns against 'overfilling' in the owner's manual for those who feel like reading it.

As far as the 'clicking"/shutoff for filling station nozzles -- I don't pay too much attention to those either since I've found that some of them click off way too early leaving the tank partially empty.
Careful and slow filling tho would be prudent to avoid spillage, tho after say the first click-off.
I can usually tell how much gas my car will take by distance traveled and make note of how much it takes on a fill to be fairly sure it's full (or almost full).
It's really not that hard to get a feel for how much it should take and go from there. For those who might think it's a chore -- it really isn't -- I do it now without much thinking ....

C.

 

[Showgun]

The mileage figure shown isn't an "estimate" (distance to empty, now that's an estimate), but is computed directly from the amount of fuel used to go the distance travelled since the last reset.

Then how do you account for the consistent differences between what the computer tells you and the old style, manual, way of doing the calculation?

For me it's easy: the computer just isn't that accurate.
For sure there's no disputing the distance traveled and the gallons taken on to refill. Provided you're careful to refill properly.
It's simple math ---

I don't know how accurate the fuel-measuring system is, but given that the direct injection system requires ultra-precise fuel delivery, I'd imagine it's pretty accurate. In any case I'd reckon it's way more accurate than any figure you'd derive from trying to fill the tank to the exact same level every time, using different fuel pumps with different shutoff mechanisms...

After doing this for thousands of miles with at least a dozen cars with mileage computers, I can say positively that the old way of measuring the distance and refill is more accurate.

I'm positive enough to bet on it if anyone's interested.


C.

 

[mjhpadi]

Two points that Shogun has made that I am in totally agree. First, I have had the station pumps click off very early, it happens consistently with our 99 Sable, went the pump first shuts off, the fuel gauge will not go all the way to full mark...it took several fills to figure it could take another 2-3 gallons after first shut off and not be overfilled.
Second, I have been keeping fuel records on vehicles for at least 15 years to monitor changes in fuel, vehicle, driving style, etc. and I have found that Colin is correct, most of the time the vehicle computer is inaccurate compared to using old fashioned math to figure mileage.

 

[dramsey]

Then how do you account for the consistent differences between what the computer tells you and the old style, manual, way of doing the calculation?

Easy: when you fill the tank, you have no way of knowing how full it really is. Different pumps shut off at different times; how much if any gas is in the filler neck, and so forth. Since you don't know-- you can't know-- that you're filling the tank to the same point each time, your mileage calculations have a built-in error.

What the magnitude of this error is, I don't know. But if I had to bet, I'd say it's way more than any errors in the car's odometer or fuel measuring systems.

After doing this for thousands of miles with at least a dozen cars with mileage computers, I can say positively that the old way of measuring the distance and refill is more accurate.

Unless you have some demonstrably repeatable way of refilling your tank with a high degree of accuracy, I don't see how you can say this.

 

[dramsey]

Two points that Shogun has made that I am in totally agree. First, I have had the station pumps click off very early, it happens consistently with our 99 Sable, went the pump first shuts off, the fuel gauge will not go all the way to full mark...it took several fills to figure it could take another 2-3 gallons after first shut off and not be overfilled.
Second, I have been keeping fuel records on vehicles for at least 15 years to monitor changes in fuel, vehicle, driving style, etc. and I have found that Colin is correct, most of the time the vehicle computer is inaccurate compared to using old fashioned math to figure mileage.

Given that you noted that you could put "another 2-3 gallons" in your Sable's tank after the pump clicked off, how accurately would you say that you could refill the car each time? To within a gallon? A half-gallon?

A small amount can make a big difference. Let's say you fill your SHO's tank, drive it 300 miles, and then fill it again. Here's are some calculated mileages based on how much gas you put in:

12 gallons: 25mpg
11.5 gallons: 26mpg
11 gallons: 27.3mpg
etc.

If your fillups vary by one gallon, your calculated mileage varies by 10%. That's a lot. Of course, gasoline volume (density) also varies somewhat with temperature, so you'd need to take that into account as well.

 

[Showgun]

Easy: when you fill the tank, you have no way of knowing how full it really is. Different pumps shut off at different times; how much if any gas is in the filler neck, and so forth. Since you don't know-- you can't know-- that you're filling the tank to the same point each time, your mileage calculations have a built-in error.

wodown a little there grasshopper: I already said I know when I have filled the tank -- and I know that because I can see it. There's only one way to know and see it and that is to fill it to the top. --- It can be done believe me.
As I've intimated b4 -- it's not for everyone and it takes a few extra minutes, but it can be done. I've done it for more years than I care to say.
So now once you know the tank is full it's easy to make a more accurate calculation -- correct.
These numbers don't lie.
The computer? I don't know about that. I'm thinking it's a rough guestimate or an educated guess at best.

;What the magnitude of this error is, I don't know. But if I had to bet, I'd say it's way more than any errors in the car's odometer or fuel measuring systems.

That's not what we're talking about here is it......Doesn't matter what the miles or kilometers say -- both the computer and I must have something to use for the calculation -- right.
If you're going to argue that the distance calculation indicated is wrong then I say that my calculation mpg's using the same distance is less wrong than the computer.
That has to be fair right.

BTW -- You can judge fairly accurately by mileage markers on the road and measured distances on major highways how accurate your odometer is (within a small error by judgment) -- and believe it or not I do this with my cars so I can tell you how accurate (within a small percentage) my odometers are too.
So you can see it's not just a guessing game here I'm talking about: It's something a little more accurate than that so please give me some credit will you.
As I've said b4....I've done this for years -- I know what I've learned and of what I speak.

Similarly you can check your speedo with the same markers. I doubt I need to go into how many seconds it takes to go a mile or how many to go a kilometer etc etc.
By doing so you can get a pretty dang close idea of how accurate you speedo is......

I believe you can use a navi to get a good idea of speed and distance too.
I know my navi shows speed and I have to believe that it's quite accurate too. Distance I haven't done on it (them) but I wouldn't doubt that you could too.

Unless you have some demonstrably repeatable way of refilling your tank with a high degree of accuracy, I don't see how you can say this.

Covered above.

sincerely and without prejudice,
C.

 

[shaker281]

I calculate my mileage and have done so for years. I consistently get .2-.5 better MPG from the computer (Ford Explorer, GT500 and SHO), than from basic math. It matters not how I fill the tank, it all averages out after a few tank fulls.

But, they are both perfectly consistent!

If my MPG improved from 23 to 28 (almost 22%) with a tune, I would be very interested in my A/F. Especially if the computer showed 21. The computer calculates the amount of time the injector is open based on pulse width and uses fuel pressure to determine total MPG. Then it looks at the O2 readings to make adjustments accordingly. If it is reporting 21, then it thinks the engine is running much richer than it actually is. Meaning it could go dangerously lean without ever compensating.

Has anyone ever posted a stage IV dyno with A/f ?

 

[dramsey]

wodown a little there grasshopper: I already said I know when I have filled the tank -- and I know that because I can see it. There's only one way to know and see it and that is to fill it to the top. --- It can be done believe me.

You mean you're filling it up to the point where gasoline is visible in the fuel pipe? I admit that would seem to be repeatable...unless of course the fuel pipe's attachment to the gas tank isn't the highest point in the system. And then there's the temperature thang...

The computer? I don't know about that. I'm thinking it's a rough guestimate or an educated guess at best.

Computers don't "guess", of course. I did a little research and it appears as though the gas consumption is derived from the mass airflow sensor readings. Since these are the readings the engine control computer uses to control the injectors (and they compensate for temperature variations, too), they should be pretty accurate.

At the end of the day we're both still guessing. I'm assuming the computer's figures are derived from accurate data, and you're assuming you've accounted for all the variables in filling the tank and that temperature differences aren't significant. But we don't really know.

It would be nice for a Ford engineer to comment here...

 

[Showgun]

You mean you're filling it up to the point where gasoline is visible in the fuel pipe? I admit that would seem to be repeatable...unless of course the fuel pipe's attachment to the gas tank isn't the highest point in the system. And then there's the temperature thang...



Computers don't "guess", of course. I did a little research and it appears as though the gas consumption is derived from the mass airflow sensor readings. Since these are the readings the engine control computer uses to control the injectors (and they compensate for temperature variations, too), they should be pretty accurate.

At the end of the day we're both still guessing. I'm assuming the computer's figures are derived from accurate data, and you're assuming you've accounted for all the variables in filling the tank and that temperature differences aren't significant. But we don't really know.

I'm still confident that I compute more accurately than any car's computer and would be willing to make a small wager to back it up.
After your feedback and counters on the subject I will concede that the computer's data will be fairly accurate but to my mind and with what I do (or can do) to fill a tank accurately I would still say I have the edge on accuracy.
-- and that's not bravado talking, it's merely experience . I simply do not find those computers accurate enough for me, variables notwithstanding.
There has always been a difference (high quote from the 'puter) between the 2 calculations for me.

It would be nice for a Ford engineer to comment here...

On that I have to agree. I too would be most interested in a Ford engineer's assessment.

I'll concede a draw for the time being, but make the stakes a beer for the winner and you're on. (you'll love southern ONtario in the spring)

Cheers,
C.