Did you miss the earlier conversation about load on the engine etc? I can put a 20,000 lb trailer behind my truck and drive 60 mph down the freeway and get much worse mileage than driving down the freeway empty at the same rpm.

Your statement also assumes the odometer is correct and the computer is accurate.

I wasn't referring to listening to some click. I was referring to the pump automatically clicking (shutting) itself off, and like I said, I've got 328,000 miles of actual logged data to back up my point. Since so many here refer to it as the lie-o-meter, I'm guessing I'm not the only one who finds fault in the computer generated mpg estimate.

Also, I don't know if the odo input or the PCM (ECU) programming is the issue. Since Ford offers (or at least used to) the same motor and tranny with different packages that include different tire sizes and gear ratios, it makes sense a standardized or average value would be used.

 

No, I did not miss anything.

When the load increases, the ECU either puts more or bigger shots of fuel or both into the engine.

Sorry, my math was off. Miles/gallons = mpg. I had it upside down.

As # and size of the shots goes up (the demonimator), if the # miles (the nominator) stays the same, then mpgs go down.

I really, really wish people would actually read what I post before commenting.

Didn't I say that if the mile-o-meter is off, it is most likely because of an odo error? So I guess, you are just agreeing with me?

As far as the computer being accurate, it almost has to be. If the ECU is over fueling, elevated EGTs result (exactly what happens in a regen). There are sensors all over the engine. For a given load and fuel rate, the ECU know the range of EGT temps to expect.

Further, the programming is what it is. It cannot be wrong. Now if the injectors are faulty, the more fuel could be going into the cylinder than the ECU thinks is going it. The result would be an over estimation of MPGs.

I suppose that the ECU programming could be overly optimistic from the factory. Why Ford would want to do this, I do not know.

People post all kinds of comments on this board. And the posters on this board represent only a small fraction of SD owners.

The ECU is programmed with axle ratio and tire size. The same info is used for the speedo. While the speedo is likely only accurate to 2-3 mph, it is unlikey that Ford uses an "average" reading from accross all of the different engine, tranny, axle and tire combos to drive the speedo.

 

Being the wide varity of tire and wheel sizes, if the truck didnt know exactly the tire diameter and gears in the rear end the speedo would be so far off you would end up getting tickets alot more often. Im sure the in dash computer uses the reading coming out the ecu to the cluster for its reading and not the spinning of the actual sensor.

 

If the speedo is off, for whatever reason, then you will have an inaccurate mileage reading and an inaccurate MPG reading, irregardless of whether MPGs are calculated by hand or electronically.

And yes, the raw data coming out of the rear pumpkin sensor is just one input that the ECU uses to make the calculations that it makes.

To finish my thought, there are too many inherent inaccurancies in the hand calculated methods discussed above to place a great deal of reliance on them. I am not saying that hand calculations are not more accurate, but unless you isolate and remove the inherent inaccuracies, you really will never know.

The inaccuracies can be averaged out over a long period of time. But this requires keeping track of many, many tankfuls of fuel over 10,000s miles. Tank to tank, listening for a "click" just looks like fuzzy math to me.

 

The click im refering too is the pump shutting off when it detects full. Not the click click click the pump itself makes.

 

I understand. But when the click occurs will vary from pump to pump and even with the same pump.

 

Sorry to have offended you.

I believe the computer generated estimate is off for more reasons than just the odometer. Just because it controls timing, fuel injection, etc. well, does not mean it can calculate MPG's in my opinion, but I see you believe I am wrong on that.

Sorry, I meant just tire size differences. Don't know why I typed in axle ratio.

 

ok so everyone is assuming a lot of things and everyone knows what assuming does.

First the click method works if you use the same pump everytime with the same grade fuel at the same temp and a lot of other variables.

The electronic lie-o-meter is taking a lot of measurements at fractions of a second. This assumes the readings are accurate and that every reading is being calculated with each reading at that exact moment. I dont think any electronics are quite there yet to make that kind of assumption. I mean I work with a lot of electronics in the chem lab that are very precise instruments. And you pay a lot of money to make them that precise. So I dont think a 60k truck is going to have the actual precision it needs to give an accurate reading. Now a close estimate I could believe.

but this is just my .02

 

Interesting.

Statistically taking a running average over multiple tanks is the most accurate way of doing it. Where the pump clicks off is meaningless using this method, because as long as you start with a full tank and end with a full tank, you can put in as much or as little fuel as you want each fill up, and as long as you end with that same full tank it'll be accurate.

Also when figuring over multiple tanks the ~0.5 gallon variance in when the pump clicks off becomes very meaningless, as you'd be averaging hundreds of gallons instead of 20-30.

I believe the computer and injectors are as precise as they can be given the technology they have to use. BUT if they were perfectly accurate we'd have no reason for an oxygen sensor in a gasoline powered vehicle!

Vehicles with the OBD-II system have what's considered "open loop" where input from sensors is disregarded and the injectors are controlled exactly as they should in a perfect world. This is done for lots of reasons, among them is the fact that oxygen sensors for some reason need to be hot to read the way they should. So once temps come up and things look like they should, the computer switches to "closed loop" mode, where it alters the way it runs the car based on what the sensors tell it.

As such the PCM constantly makes changes in injector pulse width to provide exactly as much fuel is required for the current driving conditions. There is a range of necessary corrections that can be made without setting a CEL. A common code on gasoline vehicles is P0171 and P0174, which is defined as bank 1 or bank 2 too lean. This means that the PCM is trying to make a large enough correction to it's fueling that it's out of range from what the manufacturer specified the acceptable corrections to be. This indicates a problem.

But there IS a range of acceptable values. And the only reason this could be is because fuel injection and air intake systems, like the rest of the car, aren't perfect. Therefore it seems entirely possible that you'd have some degree of variance from what the computer says it should have burned to what it actually burned.

Also, as I stated before my "lie-o-meter" is normally pretty good. I can't remember the last time it was more than 1.5 MPG off of my calculated value. It's possible that it's simply because of the variance between pumps and conditions which cause the pump to click off. Either way, I think they're both reasonably close.

 

I really do not know which method is more accurate, and both can be nearly exact or way off. All depend on an accurate ODO.

But, given the choice between the "filling until it clicks" method and the ECU calculation method, the ECU is to be more accurate given the bigger possible deltas in the hand methods.

 

Not a question of offense, but one of wasting time and energy arguing about something we agree on because people cannot be bothered to actually read a post thoroughly.

 

Just going to throw this out there to u guys. Last nite i did about a 45mile run. This is all based on the onboard. I hit maybe 15 redlights, did about 3 miles of 55mph highway. The rest was 35-40mph in light traffic.

It averaged 19.3mpg

From the onboard computer it seems That you could get better milage in the city than on the highway. Depending on your city layout and traffic of corse.

 

They are Lie-o-meters for a reason. Looking back in my Fuel log, the first actual 101,125.8 miles on my 01 was showing 18.2 mpg on the Lie-o-meter. Actual Fuel mileage was 16.817.

101,125.8/6,013.307= 16.817002 MPG

The odo showed 99,142.9 That's because the tire size (it's 2% off from the factory)makes the odo run slower then it would if it were perfect.

I'm just a little **** about the fuel mileage and the Lie-o-meter myself.

 

I noticed that too when I test drove the truck in UT. It got good highway mileage due to the gearing, but I was most impressed with how well it lugged around town and how the mileage stayed consistently high while doing so. It's amazing what having a motor that is producing a lot more torque down low will do for city mileage. This is where the Cummins have seemed to shine in recent years vs the Powerstrokes.

 

>They are Lie-o-meters for a reason.

I can see some Ford marketing department discussing how much they can make the lie-o-meter read high before people start complaining. Mine is pretty consistently 0.75 miles per gallon high (I average 9-11 mpg hauling my camper). And the pump clicks off pretty consistently 3.7 gallons before completely full (as checked by dribbling in the diesel until I see the level in the fill pipe, which I do when I hit a place with cheap diesel).