|
Hello to all.
As many "posters", I have a non-turbo non-electronic aspirated diesel engine in my car, a Mercedes 300E originally.
As many have posted, it is not an easy task, but I am confident I will succeed.
My method is triple:
Method 1: Using 2 fuel flow meters.
I am developing a system to monitor fuel consumption in real time, using two fuel flow sensors, a donation from a Swiss company.
One is attached to the tank and the other to the return pipe, as many has pointed is necessary in diesel engines.
Analyzing pulse signals in the scope, they are very square, and sensible to acceleration.
This is under development and I am dealing with programming now, donīt have definitive figures. Results will be displayed as every second marginal consumption and mean for the trip.
Method 2: . Using rpm and throttle position sensor.
My engine has a potentiometer for sensing throttle position. The engine is a 6inline RD28 from Nissan. 2.8 liters.
With these only two variables, I will get a second estimation of fuel consupmtion. I have a first model that estimates consupmtion. Data considered so far is bench marks of Bosch injector at different loads and rpms.
Adding speed sensing, I can have a second fuel performance measure. This will be displayed once a second and compute the mean for the trip also.
Method 3: Tank & odometer.
The first thing to have an accurate estimate is to calibrate the tank resistor sensor vs real liters in the tank. I made this by emptying the tank and filling it at slow rate, writing down results every 3 liters approx.
The reserve light turned off when I filled 10.5 liters app. and went on and on until full at a 55 liters. This is different for every car, I would not use it for another Mercedes 300E tank, because my tank was collapsed by a mechanic mistake and inflated later by me. So my tank has not the original shape. However, model is a 3 grade polynomial, with 98% of variance explicated. Available upon request.
Results displayed will be marginal consumption for the last minute and mean consupmtion for the trip. It will use odometer figures also. The mean for the trip of this method will be a very good validation figure.
Summurizing: I will have three (3) independent estimates for performance, which, I hope, will allow me to validate any measures with the others.
My car will serve as laboratory and I hope I will be able to have models for the three methods.
I donīt know how to insert pictures here. But may share some of the engine, of the sensors and of the car upon request.
Good experimenting, OldBeaver.
Viņa del Mar, Chile
|