why estimate mpgs?

It's the *car* doing the estimating, relatively speaking. The car knows very well how far it's travelled, but it doesn't know very well how much gas it's used. The gas pump is more accurate than is the car's fuel-flow meter, and much more accurate than the car's gas gauge (witness all the grumping about the effect of the gas tank bladder).

 

Has it been proven that the car's fuel flow meter isn't accurate? Also, when you calculate your mileage, do you figure in wind resistance (which can sometimes make a major improvement/dent), or temperature, or your height above sea level, or anything else? Does the Prius, and where's the proof of what exactly the Prius uses for estimating MPG on the fly / over the tank?

I know the pumps are supposedly regulated by the government, but that doesn't give me much hope for those pumps either - who's to say they aren't miscalculating - before you fill up, do you check with the gas station to make sure your specific pump was recently checked by the government and is within standards?

Food for thought.

-m.

 

Ahh, and how do we know it is really gasoline in those pumps?

My Prius consistently provides a more optomistic estimate of my fuel efficiency than does my calculation based on distance travelled and fuel pump reports. This is regardless of which gas station I happen to go to for my fuel.

I understand that some other people have pretty consistently pessimistic estimates from their Prius.

 

If anything, I would expect the Prius calculated MPG to be more accurate since it's measuring the actual flow of fuel.

Because of the bladder and inconsistency of when the pump shuts off, wouldn't the manual calculation of MPG be more error prone?

 

Is that true? Because I too have gotten a number of 'pessimistic' estimates from my Prius. I just tossed it up to my poor math skills (even though I use a calculator!!)

I now just do a quick average of the two and figure that's gotta be pretty close. Either way, I'm getting at least double the MPG of anyone else I know!!

PeAcE

 

The Prius does not measure the amount of fuel delivered by the #1 injector. It measures the number of times the injector fires and multiplies the firing count by a fixed estimate of the amount of fuel injected in each firing. The actual amount of fuel injected in each firing can vary based on temperature, engine load, etc.

 

As mentioned previously, technically the Computer is the one 'estimating' the mileage.

I am 'calculating' the mileage when I divide miles travelled by gallons used.

The bladder and gas pump cutoff variance can make my tank to tank calculations less accurate, but averaged over time my mileage is most certainly most accurately determined by acutal miles and actual gallons.

 

wow... my head is spinning....

if i note how much gas went through my car and how many miles it carried me, where are the errors? 1) the gas pump is not measuring right. yeah, could be, but there are specified limits for its accuracy. if you don't trust the accuracy, complain to the folks responsible for the calibration. 2) the car's odometer is off. tire inflation, tire changes? wait a minute! does the car interact with the GPS for odometer calibration? did i read that here, somewhere?

hmmmm.. odometer can't be that far off.

so, TOTAL MILES on the odometer divided by TOTAL GALLONS that went into the tank SHOULD be pretty darned accurate, and more so the longer you do the computation: total miles and total gallons, versus tank-by-tank! heck, i can squeeze about 1.2 gallons into my Prius' tank after the nozzle first shuts off. if you don't fill up with the same pump at the same station, and only to the first nozzle cutoff, you're adding many more variables to the fuel mileage "estimate".

and hold on a second....

the car is calculating fuel consumption by the NUMBER of injections?! help me if i'm way off base here, but don't most fuel injected engines vary the DURATION of the injection?!!? how else can the computer maintain an air-fuel ratio control when the amount of air going through the engine varies with rpm, load, temperature, etc., etc., etc?! well, that IS how it does it, so counting injections or multiplying by four (or fourteen) makes no sense, since it's frequency (or number of injections) TIMES duration that creates the "quantity of fuel burned."

 

It doesn't matter how many gallons the car thinks it is - you're not paying the car for the fuel

 

Hypothetical bladder example:
The tank is completely empty. You fill it with 10 gallons.
You drive off 5 gallons.
The bladder shrinks to half its size (I told you this was hypothetical)
You attempt to fill the tank and can't because it is full.
Therefore you have used 0 gallons and have infinite milage.

But I still use this method, because I had no idea how the car estimates milage (or more precicely, how it measures fule flow)

If the odometer is off, it is going to be off the same amount no matter what the method used. (right?)
If you got tires with twice the circumfrence, both the computer nad hand calculated milage is going to be off by a factor of 2...

Brian
p.s. I don't see how wind resistance will affect one method more than another

 

Sorry, but this is incorrect. The injectors are rated in pounds of fuel per hour at a specified pressure. There are curves to arrive at the amount one wants for various pressures. The engine control computer, based on fuel and spark advance maps, "looks up" the amount of fuel needed for each cylinder firing based on engine speed and load then modifies this "base" amount as a result of fuel trim software, which is determined from the O2 sensor readings (rich or lean). A final calculated injector on-time is arrived at and the computer turns on a transistor or transistor array to turn on the injector for this this period of time. This calculation is done to regulate fuel flow in millisecond durations, thus its accuracy is high if the quality of the fuel injector is high and its flow is accurate (most are).

The fuel used is simply determined by keeping a running calculation of a single injector on-time * injector rated pounds per hour fuel flow and further multiplying this result by the number of cylinders. The distance traveled is then factored into the equation. The final result is then converted as needed to the measure for your area (MPG, litres/KM).

As a result, you have a vehicle fuel measuring system that is accurate to hundreds of a gallon. However, this is only conveyed to the driver if the software for displaying fuel economy is done correctly. Many algorithms average the fuel used over time and thus lose accuracy. Many algorithms are off 3 - 5 %. The service station fuel pump is also supposed to be pretty accurate. Most errors in manual calculations result from either an inaccurate station fuel pump, evaporation losses, fuel fill variability (air pockets in the fuel tank or pump click off sensitivity), or fuel siphoned back into the service station system by vapor recovery systems.

For whatever reason, my manual calculations are within 1% of the Prius displayed MPG, and that is a cumulative result. This is the most accurate fuel usage display I have experienced in the last 15 years.

 

I was noticing a similar behavior in my new Toyota Camry (I.E. that the car-calculated MPG was significantly higher than the simple miles/gallons calculation in my spreadsheet.) I've seen a consistent over-estimation by the car through 5 fill-ups now at different gas stations in different areas of the north east.

So, this isn't just a Prius thing...and you can't pin it on the bladder tank.

I'm inclined to believe the simple miles-travelled/gallons-purchased in the long run.

(P.S. There's no camrychat.com and I remembered that Prius owners were MPG fanatics so I figured there would be a discussion of MPG calculation accuracy.)

 

"I know the pumps are supposedly regulated by the government, but that doesn't give me much hope for those pumps either - who's to say they aren't miscalculating - before you fill up, do you check with the gas station to make sure your specific pump was recently checked by the government and is within standards"

Some states do better than others at checking. I Washington State a pump may not be checked for years. The pumps I get my gas from have not been checked in 3 years and I live in the state capitol! A recent new story showed many pumps that were off by a large margin. There are not enough inspectors and no money in the budget. Heck we can't even fund schools. I would trust the car more than the pumps unless I was willing to measure the gas myself.

 

a national study says that americans pay for up to 11 million gallons of gas that they DO NOT receive.

an estimated 11% of gas pumps nationwide are inaccurate. this is due to a value that controls fuel flow wearing out as a normal course of its life.

although the valve doesnt always favor the gas station, a random check in WA St found the pumps favoring the gas station by a margin of 3 to 1. now the pumps have safeguards built in to where its possible to tamper with the pumps but nearly impossible to not be detected as the pumps use a timestamp in the software controlling the pumps that tells when the software controlling the pump was last edited. if the last edited date doesnt jive with the inspection sticker on the pump, then the owner is probably tampered with the pump.

i will say that the difference i have in the estimated mpg from the computer and my manual figurings have been pretty consistant overall. 5 times my computer reading was slightly higher and twice the actual measure was higher. but the overall rating is only .2 mpg difference (53.32 to 53.14)

my last tank was 53.8 mpg computer to 55.68 mpg measured on a 439.7 mile tank. i attribute the difference in not so much the inaccuracy of the pump but other factors including the temperature and the "fullness" of each tank.

the problem with using the amount pumped in at the gas station is there is no way to verify the amount pumped in. the inaccuracies come in since we arent measuring what gas was used like the computer does, we are measuring the amount of gas we replace. the bladder makes it impossible to know the exact volume each time we fill up making a tank here and there inaccurate.

now considering all that, Consumers Reports uses meters that measure the gas being injected into each cylinder and they consider the measurement to be highly accurate and the "topping off the tank" method to be just as inaccurate.

but we must realize that it boils down to how much it costs to drive the car. so a better indicator might be not mpg but cpm (cents per mile) so with that, lets start a penny thread.

ill start... my last tank was 3.41 cpm
my lifetime average is 4.04 cpm

 

"we must realize that it boils down to how much it costs to drive the car. so a better indicator might be not mpg but cpm (cents per mile) so with that, lets start a penny thread.

ill start... my last tank was 3.41 cpm
my lifetime average is 4.04 cpm"

I had not considered that. Good thought. It clearly speaks to one of my motivations to for buying a Prius. The M.P.G. still is one easy measure of the environmental impact of the car, speaking to my other motivation. I will trust to the cars calculation as I don't any longer trust the pumps in Washington state given recent reports in the news. I do have to say that my present car does milage calculations and when I have checked it in the past it was within a few tenths of a mpg of the gallons pumped and miles on the odometer.

 

sorry I still run a spread sheet, maybe I'm a bit anal but I wanna know for sure and exactly. Plus it also logs the temp at the time of the fill up ( more important in the winter with a tight bladder) where and price and do a actual and a mfd reading. Also log the total miles and the tank miles.

 

oh i track all that too.

but i am a computer guy and i track nearly every trackable thing in my life on a spreadsheet.

i also track movies i have seen through netflix, my bank accounts, expenses, customer tracking from work, my stats from Unreal Tournament... u know, important stuff.

 

hhhhummmm anal springs to mind