Here is why calculated MPG is unreliable.

Status
Not open for further replies.
F

friendly_jacek

Joined
May 4, 2003
Messages
6,619
Location
southeast US
I filled up today at Shell and the gas pump clicked as expected. Normally, I stop on a first click. However, the volume was lower than normally. I restarted the pump and it took 2 more gallons.

If I had some gas additive in and stopped fueling at the first click, I would be thinking I hit the jackpot!
 
I think this is more of a statement about making a claim for a single tank. I've had similar experiences as yours, but it's the long-term data over many tankfuls that are more reliable imho.
 
Originally Posted By: TMoto
I think this is more of a statement about making a claim for a single tank. I've had similar experiences as yours, but it's the long-term data over many tankfuls that are more reliable imho.


Bingo!

It explains why one data point does not a trend make. Get a few dozen and you have a better idea of typical mileage.
 
Like this?
smile.gif


http://www.fuelly.com/driver/dalybrandon/spectra

Agreed, it is best to take several tanks, because the pump inconsistencies wash out over time and fills. Or, one could always top off and erase almost all of that doubt.




(Cue in debate on topping off.... now.)
 
Better yet, put your data into an Excel spreadsheet with the correct formulas and you easily track mpg. You are able to spot trends such as lower mpg due to running A/C or, if the vehicle is 4WD, extended time in 4WD operation. You just cannot get a reliable mpg using one (1) tankful.
 
I don't really understand the problem. The pump may have shut off sooner than it normally does (i.e. at "first click"), but the important thing is whether or not the reading for gallons pumped was accurate at that point. If it was, it would hardly matter if you topped off after the first click or not.
 
Use the same pump and same method every time you fill up to track gas mileage. Any other way is a roll of the dice.
 
I'd be more concerned with the accuracy of the pump ready versus the actual amount of fuel dispensed. I've had more than one occasion that I felt the volume read on the pump was not what was dispensed...if you have a 22-gallon cell that is for all intents and purposes dry, it should take 22-gallons (plus/minus) to fill it, not 25-gallons.

As noted, one tank tells you next to nothing, and trending does (if you're consistent in your fill methods). I have a book that has every drop of fuel used in my truck for the last 256k miles...the individual tankfuls vary widely, but the average is pretty consistent with what I would expect.
 
Originally Posted By: glum
I don't really understand the problem. The pump may have shut off sooner than it normally does (i.e. at "first click"), but the important thing is whether or not the reading for gallons pumped was accurate at that point. If it was, it would hardly matter if you topped off after the first click or not.


It's important because getting an accurate measure relies on filling the tank each fill up. If I note that I drove 400 miles, and then fill up with 14 gallons but there was room for 16 gallons, I'm going to get a much different number than if I filled the tank up completely.
 
The most accurate way Ive found to do it by fill gallons is run it on full until it cuts off then run it on low until it cuts off again. This seems to be effected less by pump volume.

TMC (Truck Maintenance Council) recommends using fill records and not electronic readouts when doing fleet test as the onboard read outs are even more inaccurate than the calculator method.

When doing actual Type II fuel efficiency test you always use fuel weight which has the added benefit of correcting for specific gravity variations.

I have pumped over 200,000 gallons into one vehicle so I have a fair amount of experience with how to get a consistent (+/- 2%) fill. Although Im at something less than 100,000 in the current one.

I am able to check how consistent my fill methods are because I can actually measure inlet to fuel level unlike most cars.

PS A 5% error in initial fill can invalidate up to a years worth of test data.
 
Last edited:
My Grand Prix typicall fills up very consistently regardless of the pump. My Scangage is typically within .1-.2 gallons of the individual fill up.

My wagon is very inconsistent. I can make the Scangage accurate over 4+ tank fulls but any individual tank will be off by up to 2 gallons on a regular basis. I can tell by looking at the gas gage if I got a "full" fillup or a "not quite full" fillup as I pull away from the pump. It is very sensitive to pump speed and being level.
 
I only go to the first click on my Vic. Panthers have a problem with topping up; they have a tendency to fill the charcoal canister when that is done.

I did have the auto cutoff fail on me once about a week ago. I was filling up my Mom's Fusion and Im waiting. Then I hear the sound of the fuel coming up the fill neck. Im going, "Waitaminute is that..?" *bloosh* It throws up fuel out of the filler. Her car got super filled that time.
 
Originally Posted By: hate2work
Originally Posted By: glum
I don't really understand the problem. The pump may have shut off sooner than it normally does (i.e. at "first click"), but the important thing is whether or not the reading for gallons pumped was accurate at that point. If it was, it would hardly matter if you topped off after the first click or not.

It's important because getting an accurate measure relies on filling the tank each fill up. If I note that I drove 400 miles, and then fill up with 14 gallons but there was room for 16 gallons, I'm going to get a much different number than if I filled the tank up completely.

Of course. Sorry, that was yet another of my frequent and famous brain pharts. I should probably see a neurologist.
21.gif
 
The nozzles turn off on back pressure and the flow rate affects that too. When the station is crowded, there's only 1 turbine sump pressuring all the fuel lines from each underground tank (grade) and flow rate can decrease.

But it's best to run a 1,000 mile mpg test, and you only have to max fuel it on the first fill up. just keep filling it until it won't put any more, record odometer reading. After that it doesn't matter how full you fill, just record every quantity put in. After 1,000 miles fill it up maxed out again and there you have the total fuel used for that many miles.
 
My fill procedure...

Slowest flowing automatic notch until cutoff.

Slowly squeeze the handle until cutoff 3 more times.

Always at the same station, the same pump, and at approximately the same time of day as well.

Jim

PS Most, If not all gasoline dispensing equipment is verified by your local weights and measures folks, for accuracy.
 
Only if you squirt raw fuel on the ground!

The vapor recovery system in most any newer car or truck will hold any excess and burn it in the engine during the driving cycle.

No harm, no foul.
 
Status
Not open for further replies.

Similar threads

UpNorthSooner
Traveling and can't always fill up with Top Tier Fuel
2 3 4
Replies
75
Views
6K
Gipple
69GTX
New fuel level sensor in 2001 Lincoln only updates from a refill.....AFTER the next restart. Normal?
Replies
12
Views
643
69GTX
Pablo
My first odd credit card delay....and
Replies
17
Views
775
Pablo
DriveHard
Hard starting Honda push fixed with pre-mixed Sunoco 2 stroke?
Replies
9
Views
392
Rmay635703
T
First bad tank of gas (I think)
Replies
9
Views
2K
SwampSurvivor
Back
Top