non invasive car fuel level detector

[pakitos]

fuel detector in cars

Hi there I want design a precise fuel level detector.
what do you suggest to use as sensors/
mechanical like the ones used in cars?
ultrasonic?
light?

pls need your suggestions if someone has worked with it will be easier to eliminate options

 

[betwixt]

how a car measures fuel in the tank

I have never done this but it may be worth following up, I 'm assuming the fuel is in the tank:

Place a small audio transducer (a small loudspeaker) at the top of the tank. Make the back of the transducer airtight and the front open to the tank through a small hole.

Now send a pulse of about 200mS into the transducer so it makes a pressure wave through the hole into the tank. About half way through the pulse, measure the current into the transducer. If my idea works, an empty tank will have more air in it so the compression from the pulse will be less and the current low. If it is full, there will be less volume of air to compress so there will be more resistance offered to the transducer and the current will be different. If the principle works, it may be possible to position the transducer further away and use a rigid pipe to conduct the pressure wave.

Just an idea...

Brian.

 

[FvM]

car fuel level detection

Any sensor placed inside the tank or to it's atmosphere should be resistant to saturated fuel (hydrocarbon) vapour respectively liquid phase. I fear, this criterion is already excluding most acoustic transducers, except for hermetic ultrasonic and similar transducers.

Any measurement, that takes place in the tank's gas space has to consider it's varying composition and respective acoustical properties. A time-of-flight measurement (tank top to liquid surface) e.g. won't give sufficient accuracy without a reference measurement sensing the actual sound speed.

The suggested pressure pulse measurement additionally suffers from the effect, that a gas space filled with a saturated vapour has a virtually increased volume due to phase changes induced by pressure variations.

These restrictions make classical mechanical techniques favourable. Among the strictly non invasive methods are ultrasonic "through-wall" measurement from tank bottom, weighting the tank or hydrostatic level measurement (if the required sockets already exists).

 

[wh00per]

fuel tank accuracy car

For CAR fuel tanks they're mechanical as you already know, for various reasons ..

The ultrasonic sensor will suffer from a lot of interference (noise) when the car is moving or engine is running.

The differential pressure or weighting the tank won't work either, because the car fuel in the tank is subjected to various acceleration/deceleration forces killing any pressure sensor over the time. Also the level trend will be unreadable.

The existing mechanical ones are good as long as they have a big inertia on the readings and the level reading in the tank does not vary when you press the brakes. Also they're pretty cheap .. to be manufactured on large scale.

Another option you should look into is "guided wave radar"or "magnetic floats". Consider it as a tank "sight-glass" with a device inthere to send out some output. The sightglass being vertical, the sensor won't be affected by the acceleration or deceleration of the car, but the fuel level might be.

 

[pakitos]

how car fuel level sensor works

I know the mechanical is the best choice but if you want to add a transducer it is easier not to use invasive methods.
there is a noninvasive sensor at:
**broken link removed**
that I want to know how it works

and another ultrasonic at:
**broken link removed**

what do you yhink the prob of using the ultrasonic ( send a short pulse and measure time taken to be reflected back)
thanks to all