I could use some help with an experiment...

[camrycev6]

Good afternoon everyone,

I have been thinking about this one for a little while, and I realized I can't do it without some help from at least a few of you. It has been my experience, that the trip computer MPG is never accurate. So, either something wrong with both of the MS3s I owned, or it is just a fluke.

So here is what I would like to do...

I would like to have as many of you that are interested, run a full tank of gas (93 please... Jville I know you will appreciate that!) measuring the actual gas mileage (Miles driven / gallons filled) and then comparing it against what the trip computer says.

I am halfway through my tank, so I expect to have my new results very soon. If you would like to help out, please reset everything on your next fill up and let me know your results!

Thanks.

 

[imac]

i dont have exact numbers for you because i havent archived my results, but i've guaged just about every tank i've gone through and the reading on the computer is usually 2-3 mpg over what i calculate.

if i remember correctly, last tank was 273 miles on 10.89 gallons of 93 octane, which equates to slightly over 25 mpg. i believe my computer said 27.9 mpg (and yes, i reset one of the trip odometers and the avg mpg function each time i fill up)

 

[daonly1around]

i just took a highway trip, with cobb SF intake installed, and i got 29.999 miles per gallon

 

[MikeHTally]

I suspect they're all optimistic, due to them being an average. My Ram is the same way. I'll see what I can contribute, but I have no road trips planned for the Speed. It'll be just the daily runnin' around and ejoying the loud pedal. The next road trip is in the Ram.

 

[camrycev6]

i just took a highway trip, with cobb SF intake installed, and i got 29.999 miles per gallon

Can you tell me what your trip computer said though? I need your actual versus computer data...

 

[camrycev6]

i dont have exact numbers for you because i havent archived my results, but i've guaged just about every tank i've gone through and the reading on the computer is usually 2-3 mpg over what i calculate.

if i remember correctly, last tank was 273 miles on 10.89 gallons of 93 octane, which equates to slightly over 25 mpg. i believe my computer said 27.9 mpg (and yes, i reset one of the trip odometers and the avg mpg function each time i fill up)

This has been my expereince as well. The computer has always been a little too high. (2-3, just like you.) However, I expected it to be closer considering we are talking about 24 to 38 miles per tank. We are talking about over a gallon of gas! Anyway, I will reserve judgment until I see more than just my data.

 

[robin2660]

I usually record both numbers, actual and computer. They never agree but, if I recall correctly, there really isn't a consistency. I have a theory that the computer calibrates after each fill-up based on how you drive the first few miles.

I'll post some numbers when I can get to my car.

What is your goal (a little more explicitly, please).

 

[pidass15]

The computer computes an average not actual. That's why is says 'avg' in tiny little letters. So it will never be exact. Mine is usually off by ~2mpg though.

 

[///M Compact]

I've calculated the difference between my MS3's trip computer mpg and the "actual" mpg as a percentage for several months. Actual mpg is usually between 93% and 98% of the trip computer mileage. FWIW, every BMW w/OBC I've owned has also reported optimistic mpg figures. At least the BMW OBC can be adjusted by entering new calculation parameters while the OBC is in diagnostic/test mode.

 

[camrycev6]

I usually record both numbers, actual and computer. They never agree but, if I recall correctly, there really isn't a consistency. I have a theory that the computer calibrates after each fill-up based on how you drive the first few miles.

I'll post some numbers when I can get to my car.

What is your goal (a little more explicitly, please).

I just wanted to see what the average difference between the computer and the actual miles is. I also want to see if it is higher, lower, and if some correlation can be drawn from the data.

Also, that is an interesting hypothesis (not quite a theory unless you have some data to support it). I would be curious to know why you think that. Are you saying that the after you reset the computer when you fill up and drive 10, 15, 20, etc. miles, it uses some calculated variable for determining the rest of the data it reports?

 

[camrycev6]

The computer computes an average not actual. That's why is says 'avg' in tiny little letters. So it will never be exact. Mine is usually off by ~2mpg though.

Yes... I am aware it is an average. You are also calculating an average (it just happens to be the most accurate -- hence actual) when you divide your miles driven, by the number of gallons you put in. That is why it is imperative that you reset your computer when you fill up. It will then give you the average for the tank when you are ready to fill up again.

We would be comparing apples versus apples here.

 

[pidass15]

Yes... I am aware it is an average. You are also calculating an average (it just happens to be the most accurate -- hence actual) when you divide your miles driven, by the number of gallons you put in. That is why it is imperative that you reset your computer when you fill up. It will then give you the average for the tank when you are ready to fill up again.

We would be comparing apples versus apples here.

Yes, when you do the actual calculation when you fill up you are calculating the average rate. What I'm saying is the trip computer computes a running average to try to approximate the actual.

So yes, the way you drive the first xx miles after resetting the computer will make a significant change the final value at your next fillup.

To graphically represent it look at the picture I've attached, the wave is what the trip computer is trying to do. The longer you drive the better it's able to converge on the final average (the axis in this case).

 

[camrycev6]

Yes, when you do the actual calculation when you fill up you are calculating the average rate. What I'm saying is the trip computer computes a running average to try to approximate the actual.

So yes, the way you drive the first xx miles after resetting the computer will make a significant change the final value at your next fillup.

To graphically represent it look at the picture I've attached, the wave is what the trip computer is trying to do. The longer you drive the better it's able to converge on the final average (the axis in this case).

I totally agree with what you are saying here. I was aware the computer did this. That is why it is so inaccurate when you first reset it. I guess to say it another way, I was hoping that after 300 miles (a tank roughly), the computer would be a little closer to actual.

I suppose another way to do this would be to NOT reset the computer each time you fill up, and just keep track of your miles / gallons and see how long it takes to get a decently accurate average. The problem with this method is that it is meaningless for me. I want to know what I am averaging during a long trip without having to do the calculation myself. If the computer requires 1000 miles of driving before it gets close, it won't be of very much use.

 

[phillyb]

how does i 93 octane?

 

[camrycev6]

how does i 93 octane?

What?

(confused)

 

[phillyb]

they sell 93 octane by you in md?
idk where i can find 93

 

[mordant80]

they sell 93 octane by you in md?
idk where i can find 93

Here in WV it's a mix. Some stations only offer 91 and others have 93. You can tell around here by the price. If there is a 10 cent difference between grades they have 87 thru 91 gas. If there's more than a 10 cent difference they have 87 thru 93 gas. That's what I've notice locally anyhow.

 

[robin2660]

I just wanted to see what the average difference between the computer and the actual miles is. I also want to see if it is higher, lower, and if some correlation can be drawn from the data.

Also, that is an interesting hypothesis (not quite a theory unless you have some data to support it). I would be curious to know why you think that. Are you saying that the after you reset the computer when you fill up and drive 10, 15, 20, etc. miles, it uses some calculated variable for determining the rest of the data it reports?

Uh, hypothesis, yes. Sorry about that.

I base this thought on my recollection that the computer shows inconsistent differences from true mileage. I might be wrong on this - won't know for sure until I look at my records.

 

[camrycev6]

they sell 93 octane by you in md?
idk where i can find 93

Yes... everywhere actually. We have 87, 89, and 93. You can even find a little higher (95 / 96) at regular stations if you know where to look.

 

[camrycev6]

Uh, hypothesis, yes. Sorry about that.

I base this thought on my recollection that the computer shows inconsistent differences from true mileage. I might be wrong on this - won't know for sure until I look at my records.

Cool. Based on my previous posts, I was just hoping that the trip computer would be a little more acurate after less than 300 miles. Sure... drive 3000 miles on it and I bet it is very close to your average for 3000 miles. I was looking for better accuracy over shorter distances. Oh-well, let see what we find....