Trip computer for vehicle (looks complex but comes down to very simple problem).

So over the holidays, I'm going to try to install a Ford Explorer Message Center into my Ford Ranger. Shown below.
**broken link removed**

This computer takes signals from your Engines computer (fuel flow rate and speed) and from your fuel pump (fuel level), and it can output your MPG and Range.

Now here is the problem I face. The gas tank that comes on the Explore is 18.4 gallons and my tank is 16.5 gallons, which doesn't mess with the MPG reading but it can mess with the amount of miles left till empty (range). Here is some Matlab graphs showing what the real range is vs the output shown on the computer (assuming a MPG of 23):
**broken link removed**
**broken link removed**.

As you can see the difference is pretty significant at a full tank (around 44 miles), but the differences becomes smaller with the decrease of the fuel of the tank.

Both vehicles are set up to show the resistance range 22 ohms-145 ohms. 145 being full, 22 being an empty tank.

Is there a way to have the output resistance of the fuel pump be modified before it gets to the computer? In other words instead of 145 ohms when it is full, the computer reads 132.3 (the resistance the computer would see as having 16.5 gallons), but still show 22 ohms when empty.

Pretty much create a mathmatical function with the input variable being the resistance from the fuel pump and the output being the modified resistance for the computer to calculate the range correctly. Then take this function and make a physical application.

Hi,

I'm pretty sure you'll have to set your tank amount of fuel by hitting any button. The trip computer will then calculate on Ford's tank capacity information being fed into the computer.

As the capacity of your gas tank is higher than the liquid fuel tank you'll run into problems anyway.

The flow sensors sense fuel flow forward (through the injection system) and back into the tank (return line for excessive fuel) and use the difference to calculate for consumed fuel.

Switching from gasoline to gas you might reset the tank amount, but still have the disadvantage of less fuel indicated.

Flow sensors work with a rotating wheel supplying pulses (or capacitive) and are designed to indicate real fuel flow on the trip computer.

You should obtain a diagram of the trip computer and then set the fuel amount (gas) to the real value when switching from gasoline to gas.

Boncuk

There is a wire that selects the tank capacity, either you ground it or apply 12v. The only options are 18.4 and 20.8 gals. I just chose 18.4 because it is the closest.

After doing some calculations I have came up with the equation y=0.89674x+2.2717 where x is the resistance being read from the fuel pump sender and y is the resistance being fed into the computer. Now how would you apply this?

How about extending the bit of wire that has the float on it? That would make the shaft turn less, so the range of the gauge would be less. You would also need to bend the wire to get the zero in the right place.

The gauge would not quite read full with the tank full, but the remaining mileage would be about right.

Boncuk said:

Switching from gasoline to gas you might reset the tank amount, but still have the disadvantage of less fuel indicated.
...
You should obtain a diagram of the trip computer and then set the fuel amount (gas) to the real value when switching from gasoline to gas.

Click to expand...

I think he's using the word "gas" to mean "gasoline (petrol)", I don't think he's converting the fuel system to LNG or LP.

You don't need any complex circuits or modifications. Just connect 1500 ohms from the fuel tank sensor output to ground. That would give 132.2 ohms when full (145 ohms from sensor) and 21.7 ohms when empty (22 ohms from sensor), close enough, I would think.

Edit: To get closer, you could also add a 0.3 ohm resistor in series with the sensor output along with a 1470 ohm resistor from the sensor out to ground. That would give 132.3 ohms when full and 22 ohms when empty, as desired.

crutschow said:

You don't need any complex circuits or modifications. Just connect 1500 ohms from the fuel tank sensor output to ground. That would give 132.2 ohms when full (145 ohms from sensor) and 21.7 ohms when empty (22 ohms from sensor), close enough, I would think.

Edit: To get closer, you could also add a 0.3 ohm resistor in series with the sensor output along with a 1470 ohm resistor from the sensor out to ground. That would give 132.3 ohms when full and 22 ohms when empty, as desired.

Click to expand...

This sounds like a good answer. Wouldn't this create a problem with the actual fuel gage in my instrument cluster? Wouldn't the gauge be seeing the same resistance as the trip computer?

Diver300 said:

How about extending the bit of wire that has the float on it? That would make the shaft turn less, so the range of the gauge would be less. You would also need to bend the wire to get the zero in the right place.

The gauge would not quite read full with the tank full, but the remaining mileage would be about right.

Click to expand...

Seems like a pretty good "thinking outside the box" idea, but I'd rather not take my fuel tank apart.

sheltonfilms said:

This sounds like a good answer. Wouldn't this create a problem with the actual fuel gage in my instrument cluster? Wouldn't the gauge be seeing the same resistance as the trip computer?

Click to expand...

Yes it would. So the choice is my simple fix to have the new gauge read correctly and your old gauge read somewhat incorrectly, or use as is, with the old gauge reading correctly and the new gauge incorrectly.

But if both gauges are actually measuring resistance then they will likely interfere with each other and neither will read correctly. I think you may need to use either one gauge or the other, not both.

About the the two interfering: The way it would be wired in would be the exact same way as the it was suppose to be wired. So somehow the computer senses the resistance without compromising the resistance seen by the gauge. BTW the gauges are the exact same in both vehicles.

Yeah i double checked on the wiring. On the Explorer the fuel pump is grounded and the other end coming from it is spliced into three wires, one going to the gauge, one to the trip computer, and one to the PCM (powertrain control module). On my Ranger (from factory), it splices into the gauge and the PCM. So I would just be adding into this splice.

Looking at a simulation using both ideas (1500 in parallel and 1470 in parallel with 0.3 in series), it appears that both give very good results. The 1500 resistor suggestion has the lowest difference in range compared to the actual range at 50 percent tank capacity yet will show a range of 0 miles before the actual tank is emptied (around 0.275% of tank being full, which is about 1/20th of a gallon). You could almost say this is will keep you from driving all the way till empty. The parallel and series combination shows a higher difference at 50 percent capacity (only 1 more mile than the 1500 resistor setup), but is more accurate when the tank is almost empty. These are very small differences between the two but I always over think/engineer ( I guess that can be a good thing).
Take a look at the graph:
**broken link removed**

Ok, here's a different tack. The Distance-To-Empty computer is likely just reading the voltage generated by the resistance and the gas gauge voltage. Thus you probably could just use a voltage divider to reduce the voltage to the DTE sensor by the 16.5/18.4 = 0.897 difference of the two tank sizes. That way the gauge reading wouldn't be affected but the DTE sensor would see the correct full tank value.

At the low end, the sensor will think you are out of gas ahead of the actual. If desired you could compensate for that by adding in a small dc voltage from the battery through a resistor to the DTE sensor.

You want the resistive divider to be much larger than the maximum tank sensor resistance of 145Ω, say 1000 times or 145kΩ to avoid affecting the gauge reading. You might be able to use the DTE input resistance as the bottom half of the divider. Just measure it with an ohmmeter to see what it is. Then calculate the input resistor you need in series with that to give the desired 0.897 voltage reduction.

Would a Op Amp Buffer isolate the resistance?