Theoretical Max

[trebuchet03]

Originally Posted by MetroMPG

Treb: sounds like you may be after a metric that I wish government agencies would provide: just show me the fuel consumption across a range of constant speeds in top gear. Forget trying to simulate an entire trip to grandma's house for freaking sake!

I got my frescale 33290 just now... I'm going to the surplus store shortly to get some necessary componentry. I'm going to use an RJ45 jack so I can use my Scangauge cable. I think I'm going to start with the OBDuino code, strip the LCD stuff out and have a laptop log data (arduino doesn't have much space to store data).

What I'd ideally like to do is generate a real time 4D map. Basically, a scrolling 3D map that can be played back later. Freezing the map over a range and having live updates would be cool too

Realistically, I'll log a bunch of data then go back and generate maps manually But, the nice thing is - I can theoretically plug the cable in to any car and run these same metrics again

[metroschultz]

Originally Posted by bbjsw10

Wow, Frank didn't know you had it in ya.

trebuchet03, I may have to try this. I finally got my MpGuino going. I am sure you are probably not in the market for an Xfi though. I am afraid to bog it too low though. 3cyl=no torque.

I've had my Metro (Xfi cam & 353 trans) down to 19 MPH.
Idling along in 5th gear coming down my street. (I've cut the idle down to bare minimum)
You will be surprised at the bottom end that cam gives you.
Many times I am in high gear before 30 MPH and just keep pulling to 50.
Since I don't have a 'Guino you will have to be the guinea pig. (I'll have one when Joe the Plumber gives me his money)
Schultz

[roflwaffle]

Originally Posted by trebuchet03

I need to put more thought into what it is exactly I'm trying to find....Or maybe... I don't know exactly what to call throttle v mpg v velocity, but I still think it will be somehow useful. Especially if I can make a way to determine it quickly (ie, car shopping )

So... A MPGvT map? Just pick a gear and a range of power outputs needed. For instance I used a 5% grade on either end, and then just convert (V*(Fuel weight in grams/(P*Fc))) the fuel consumption figures to mileage in a grid. Something like the attached graph, except, you know, good.

W/o a BSFC map, as was mentioned earlier I imagine we could make the same thing if we could log mileage, speed, and throttle in order to calculate a MPGvT graph for a specific gear. Or better yet...To make things easier, we could just record the values and pick the lows/highs in order to get a feel for the extremes.

[Andyman]

Try this formula:

MPG=1204*MPH/(L*RPM*VE)

L: Liters of engine displacement
VE: volumetric efficiency of the engine

Usually the volumetric efficiency of the engine at low constant vehicle speeds is about 0.30 but when I read about the Oldsmobile Turnpike Cruiser I discovered that the VE was only 0.26 when cruising slowly. High compression ratios help to reduce the fuel consumption of engines running at light loads. When the throttle is completely open, the VE of an engine without a turbocharger or supercharger should be close to 1.0 when the engine speed is near the torque peak. Using a VE of 1 in the formula will let you estimate the fuel economy when using full throttle.

Lets try an example: A 1999 Saturn SW2. I'll use these numbers:

MPH=25
RPM=850
L=1.9 liters
VE=0.30

Now here is the calculation:

MPG=1204*25/(1.9*850*0.30)
MPG=30100/484.5
MPG=62.126

That seems to match the measured result closely. There are a lot of factors that could make the VE estimate of 0.30 inaccurate. For example: unusually high gears, unusually small engine, unusual compression ratio, and installation of a high performance camshaft.

[Blue Bomber Man]

My roommates Yaris gets around 80mpg at like 38mph according to his scangauge.

[Funny]

Subscribed. I have to see if we can make this progam of a 4D, laptop based BSFC program, which can plug into OBD2... Man that is a tall order...