Quote Originally Posted by avidflyer View Post
Am I wrong?

A bit, yes. When there's no fuel detected, both outputs are high impedance, or floating (OFF). When fuel is detected, both outputs are active (ON). The active high output (yellow wire) provides 12V (or whatever voltage the sensor is powered with), and the active low output (white wire) pulls to ground. That's the important point: with fuel present, both outputs are ON; with fuel absent, both outputs are OFF.

If you want to light an LED when fuel is present, simply connect it to the active-high output with an appropriate current-limiting resistor (R):

R = [supply voltage - LED forward voltage] / desired current

For example, a typical green LED will have a forward voltage of 2V at 20mA. So, for a bus voltage of 14.4V...

R = (14.4 - 2) / 0.02 = 620Ω (a standard value!)

LED.jpg

This will indeed confirm that the sensor is working. However, when fuel is absent, both outputs are OFF, so it won't send current to your red LED.

If you want to light the red LED when fuel is absent, you need to invert the active-high output to obtain 12V when it's OFF. The simplest way to do this is with two components: a resistor and a P-channel MOSFET. That circuit looks like this:

Inverter.jpg

When the active-high output is ON (fuel present) the MOSFET's gate (labeled "G") is driven to 12V by the sensor, and when the gate is at the same voltage as the source (labeled "S"), a P-channel MOSFET is off. When the active-high output is OFF, the 10kΩ resistor pulls the MOSFET's gate down to ground, which turns the MOSFET on, allowing current to flow, lighting your LED (you still need the current-limiting resistor above). The 10kΩ gate resistor value isn't critical. It just needs to be high enough to limit current through the sensor output when it's ON; anything from 1kΩ to 100kΩ will work fine.

So, if you want two LEDs, green turned on when fuel is present and red turned on when fuel is absent, the circuit looks like this (again, LED current-limiting resistors not shown):

Inverter_2_LEDs.jpg

Now, with all of that said, I should point out that virtually all aircraft manufacturers these days adhere to the "dark cockpit" concept, which means that no indicators should be lit when systems are operating normally. So, your green LED would not be installed. In any case, the fact that the engine is running makes it kind of redundant.


As I mentioned in an earlier post, I'm designing a small interface board to connect between the sensor and a warning indicator. It will actively hold the LED off when fuel is present, and flash it when fuel is absent. It may also have a press-to-test function to confirm that the flasher circuit is working. If you press the test button and the LED flashes, then goes off when you release it, both the sensor and the flasher are working. Using an illuminated pushbutton switch would incorporate both the warning indicator and the press-to-test function in one panel hole.

It may turn out that the sensor outputs are OFF when it's first powered up, and it takes a couple of seconds to self-calibrate and detect fuel in the tank. In that case, the press-to-test feature would not be necessary because the warning indicator would flash briefly each time the master switch is turned on. This would confirm that the flasher and LED are working, then when it stops, you would know that the sensor works. The switch could instead be used to de-power the circuit, eliminating distraction during a genuine fuel emergency.

Anyway, that's enough middle-of-the-night thinking out loud. More testing to be done. Stay tuned...