Hand Calculation Accuracy?

I've always had a niggling thought about this as well...whose to stay any method is accurate, short of draining the tank and laser measuring your routes

At least the onboard computer gives some reasonable comparitive consistancy.

 

Hand calculation is a prime reason to emphasize long term cumulative fuel efficiency, not a single tank. You quickly begin to see credible results after 10, 20 and 30 tanks.

 

scruse nailed it. i was filling last week and around 9 gallons when i thought it should be done, the handle started buzzing. just as i was thinking, "that's odd", gas started shooting all over the place. it pumped 10 gallons all over me, the car and the ground. gas kept coming out of the filler neck for about 10 minutes. all i can figure is the bladder expanded and then contracted. first time in 8 years this has happened. anywho, i've put 200 miles on her with no ill effects that i can detect.

 

This happened to me once as well. It was terrible!

 

it took three days of showers to get the smell out!

 

At least is wasn't diesel. That smell lasts nearly forever.

Tom

 

i think i still have it from my 300sd in 1998.

 

Speaking of fueling up, Although I haven't had back splash, (knock on wood) I am wondering if this would account for my 5mpg difference between MFD and Fuelly on my second tank?

It was bitterly cold (8 above and I am used to 35-40degree weather for winters) Had to prepay cash inside and when pump kicked off I still had $1.49 left and didn't want to go back inside the store and wait for my change, (store was actually a truck stop and was super busy) so I slowly pumped the rest in which was roughly 1/2 a gallon more of fuel.

HA!! Guess I just answered my own question. I just did a calculation of 1/2 a gallon less and came up with 41.9 which would be spot on of fluctuating around 1.5-2 mpg difference between MFD and Fuelly...

 

...happened to me just once too (overflow) bad nozzle?
I thought gaso would smell better when we got to <10 ppm sulfur but still seems to.
Better I think though.

I go for long term average of real fill-up data

 

Yes, of course variations in the exact level to which you refill will result in variations (or anomalies) in your hand calculations. Any calculation is only as good as the data you feed it, and your measured fuel usage is only accurate if you re-fill the tank to precisely the same capacity to which it was filled prior to where you started logging the miles.

This of course is virtually impossible if yours is a model with a tank bladder. And even without the bladder you are still subject to all manner of variations.

- Variation of pumps with different cut out pressures.

- Variation in how far you've inserted the nozzle.

- Variations in the pumping speed (even at the same pump but on a different day, due to differences in the stations tank levels and filter condition).

- Even variations in slope (left-right ot forward-back) of the gas station driveway.

When I fill the Prius it's usually the same local petrol station and I'm fairly consistent in how I do it. So while my hand calculations show some variation it's usually only 1 or 2 MPG. When my wife fills the car however the calculations are all over the place, 40MPG on one tank and 70 MPG on the next. The explanation is pretty obvious. Actually I don't even bother doing hand calculations anymore unless I was the one who did the filling, both prior to and after logging the miles.

 

Get ye a spreadsheet and calculate rolling averages 5 tanks at a time. The standard deviation for the 4 results that 8 tanks allows is a good measure of inaccuracy.

Too lazy? keep a running total of lifetime MPG. Very quickly the result stabilizes to well within a few tenths from tank to tank -- at least until the weather changes.

 

With the mileage I'm getting, won't I be trading in before 30 tanks?

I know there are accuracy problems with both the computer report and hand calculation, but I'm so smitten with the reality of not gassing up every other block (I traded down, no that's not right, traded up to a 2009 Prius), I don't really care if it's 45 or 39, either way it is 3-4x the old gas guzzling suburban.

 

exactly. the best you can hope for is a running average. when i do the same type of driving week in and week out, in similar weather, my single tank figures are within 1-2 mpg of my running average for that time of year. air temp is my biggest change factor.

 

Been reading this thread with interest. Just had my 3rd fill-up today.

Is that change factor based on the (apparent) smaller bladder size in the cold or is it the cold air = dense air therefore burn more fuel to maintain 14.7:1 ratio?

As for pump blow-back, I'm using the Scangauge "expected liters burned" this tank-full to know when the pump should click off. Its been within 1% each time......

 

Cold temperatures affect everything from bladder size, air/fuel ratio compensation, tire rolling resistance, aerodynamics, battery efficiency etc.. Below about 60 degrees efficiency degrades.

 

Perhaps people are not understanding why fuel economy accuracy is improved over longer distances in a hand calculation ?

Start by filling the fuel tank to 'full' -- the nozzle shuts off.

1. Drive 400-500 miles or so, put in 8 or so gallons. Lets use 400 miles and 8 gallons for this example. Experience tells us that 'full' can vary by about +/- 0.5 gallons, or one gallon total error from full to full. So after one tank of gas the error is up to 1:8

2. Drive say 4000 miles, and by the last fill-up (this time to 'full') about 80 gallons have been put in the tank. Now the error is 1:80 -- accuracy has improved 10 fold. Note that there is no reason one has to fill to 'full' every time, just the last fillup for the calculation. All one needs is:

A. The odometer reading at the start of the test
B. The odometer reading at the end of the test
C. The total gallons put in the tank during the test (not including the fill-up at the start of the test.)

Easy as pie. I throw fuel receipts into my glove compartment for a couple of months, and then calculate fuel economy in about two minutes.

 

Yep. Anything to do with technical/performance standards in my old job (RCAF) was based on "an ICAO standard day" which includes a temp of 15C (59F).

My OCD :dizzy: demands I enter my numbers into an Excel spreadsheet (for all my vehicles) that includes a "total km's driven" and "total liters burned" based average L/100 km's......part of a sheet that totals all costs per km.

My 2007 HCH worked out to 36.88 cents per km (of which fuel was only 4.73 cents per km, closely followed by insurance at 4.58 cents per km).