fuel sensor ~ obd2

[user removed]

I did the voltmeter to sending unit setup with my brothers 90 Civic hatch when I was doing some testing and wanted more accuracy. No other MPG instrumentation. One of the issues that became apparent to me after repeatedly driving a 9 mile loop was the huge variation in tank readings.

It finally dawned on me that after a 9 mile run and a tank reading that was actually HIGHER than when I began that loop, that the return fuel system combined with 90+ degree temperatures resulted in the fuel in the tank heating up. Adding the hot asphalt pavement to raise the temperature even higher made short distance calculations an exercise in futility requiring the car to cool down overnight for decently repeatable readings.

I did see 90 MPG loops without cold starts, significant traffic, no traffic lights, and P&G engine off from 25-40 MPH, but even the heating and cooling as the daytime temps increased the heat content and expansion of the fuel in the tank made true precision an exercise in frustration. Even parking in the sun affected the readings versus parking in the shade, when you consider the asphalt was probably at a surface temperature of over 150 degrees in direct sunlight.

regards
mech

[oldtamiyaphile]

Giving it some further thought, what we need is to weigh our tanks.

Create a hinge support on one side and a load cell on the other (like from a luggage scale). The tank would sit on a stop when full and this would prevent damage to the load cell on a big bump.

Would only be accurate when on smooth, level ground, but it should work pretty well otherwise.

[Vekke]

Actually measuring the fuel level sender resistance is used in skoda economy run to measure actual fuel consumptions and results of the fuel economy race.

What they did was:
- Start with empty tank and measure the resistance value with VAG com etc
- add 1 liter of petrol or diesel and check value. You can use which liter value but less volume more accuracy. First check which is minimum which changes the value...
- Repeat the procedure until the value does not change and you have created a accurate way to measure the used fuel in a fuel economy race or just for your own fun.
- In their races all cars are the same model and when you come to race tank cannot be full. They know it from the resistance value...
- When you finish the race you just need to read the resistance value and fill the tank with that amount of fuel the resistance value suggests from the chart you generated. If the value becomes the same you have your exact fuel value burned. Ofcourse you need to be very precise with positioning of the car when you do your fillups in a race and for that there are ways to do it accurately.

This way you dont have to build a super long race of 400+ km to see which is best and fillup data is usually more accurate than with regular fillup methods as you can check the results from that resistance value.

Now you just have to measure that resistance value in same place like some level part of your regular route if you use this method. More precise data means less km or miles drived in testing purposes. you just need to test which is the minimum amount of added fuel the resistance value changes.

From that you can calculate the distance you have to drive to see results that are pretty reliable.

Have fun with this .