Level meters for Liquid Petrol Gas (butane, propane) tank.

[SpaleKG]

Hi to all,

I have (maybe as a lot of other ppl) car with gasoline/gas reservoars. Now anyone know about issues for gas tanks (butane-propane). If you want to see gas level you need to take a look on mechanic meter placed on reservoar.

So I would like to make electronic MCU controlled or whatever  gas level tank meter which output (LCD, LED display) will be mounted easily on driver dash board.

Anyone can some ideas what to use as sensor for measuring level of gas. Maybe complete project already done ?

Regards.

[Ichan]

By measuring the tank pressure, perhaps?

-ichan

[Magnox]

I don't think you can use the tank pressure with lpg because it is mostly liquid. As gas is taken, more liquid boils off to maintain a fairly constant equilibrium pressure inside the tank. It's not like compressed air SCUBA tanks (which I use for my PCP air rifles) where the air is stored as compressed gas, not liquid air.

The usual methods are by weight or by using a float inside the tank.

Maybe it would be possible to interface something to the existing mechanical gauge...

[FTL]

The pressure does not really change with liquid level. The pressure is mostly Dependant on the vapor pressure of the liquid at the current temperature. It will also increase slightly with agitation.

One method that is sometimes used it to measure the temperature of the side of the tank. If the fuel is being used, it will be evaporating off of the surface of the liquid, so it is cooler at the level of the liquid. I'm not sure how useful that would be for a tank that has movement (like in a vehicle), as the liquid will move and mix a lot.

Maybe an ultrasonic sensor can be used inside the tank to detect the liquid level?

[Langley]

Well, for large tanks we used Cs-137 (or even Co-60) gamma sources on one side with detector on opposite to gauge specific gravity. I suppose you could put a source at the bottom and detector at the top to measure volume. Of course, depending on where you live in the world, acquiring a suitable source and license might be a big problem… 

Actually, though, while my suggestion was mostly in jest, for a small tank such as in a car the source might not need to be very large at all given that averaging of counts over many minutes should be fine.
If the tank is a low density material like aluminum you might even be able to use something that’s actually feasible for private use like an old Radium clock face as a source.

Of course, an ultrasonic or optical sensor is far more practical, but not nearly so much fun!

[PaulC]

Would the temperature of the bottle be slightly different on empty level than on fill level no matter what the ambient temperature of storage area is, this could be an option maybe ?

[solutions]

The problem, as had been originally defined (and most of you missed, because he wandered off into sensing), is remote reading, not sensing the tank level.

Mount a small magnet on the needle of the tank level indicator that's there, then attach an Allegro angle sensing chip to it.

Don't hurt/kill yourself - what you are playing with is a bomb.

[gabriel]

Milone has made a very simple fluid level meter (www.milonetech.com), and promised to launch one for petroleum.
HTH

[PaulC]

Hi SpaleKG
I & Probably others would be interested to see which route you took to your problem.
this is an interesting project to follow.
as i use gas bottles for gas fires ect (I live out in the sticks).

[jsds]

Hi,

I have had to build a level sensor in the past for measuring fuel levels, its basically 2 tubes, one inside the other with holes drilled to alow fluid to pass freely without air locks, you then measure the capacitance between the 2 tubes (you can do this easily using a 555 timer circuit). It is a very acurate method of sensing fluid levels

[solutions]

http://www.enertrac.com/support/open/EnerTracPropaneTransmitternew.pdf

[eSilviu]

how about measuring the resonant frequency of the tank?
a small piezo glued on the tank make a short bust of sound and a microphone captures the echo.  

[sarah90]

What does the current mechanical indicator look like? Could it be read with a camera?

[dipchip]

I know very little about this subject... that said, I might go about it this way.
You know the size of the bottle.
You will be driving and the liquid inside the bottle will be moving around... a lot.
You (as do I) would never want to put an electrical charge/signal inside that bottle!

Of the propane tanks I've seen, most have separate fill and outtake valves.
I would look for something like a flow meter with a pulse output on the outtake.
It might take a few trial runs before you got a count per gallon, the rest would be
straight forward.  Write the running count to eeprom continuously (or better yet,
a battery backed i2c ram) recall it when you power up.  Reset flow button on the display.

--Chip

[PaulC]

Could it not be as simple as weighing the bottle empty then full then dividing the weight by as good as you want it to scale, electronic scales then wireless transmitter this way there are no electrical interference with the gases involved,
i think i will give a go in this direction ??
will let you know how i get along..

[Magnox]

There are some fair ideas here for measuring an LPG bottle in one's shed, which is where I keep mine, but...

The thing is, we are talking about an LPG tank fitted into a vehicle. I doubt that modifying the tank, its mounting or the plumbing is really feasible, possibly not legal and almost certainly not very safe unless one is well versed in the mechanics of handling LPG. Also, any methods that might suffer from the vibration of the engine or vehicle movement could be tricky. A lot of mechanical damping or electronic integration might help, but even so...

As Solutions pointed out, the issue is more one of remote reading the existing meter, not implementing a new sensor.

Sarah90 has an idea. A simple CCD or CMOS sensor had occurred to me also, but it seems like using dynamite to kill an ant.

Solutions' magnet idea is another possibility.

The OP has not been here since the question, but it would help to know how the existing gauge works. I've not seen one. Is it rotary or linear, is it physically accessible or is it encased within the tank? Is it a moving needle, scale, or float bubble?

Without that information it's difficult to know what to suggest, but there are a range of possibilities floating about in my head.

The dynamite idea is probably the simplest... it might even be the only really feasible one!

[sarah90]

The idea is not that original. Cameras are used throughout the petrochemical industry to monitor installations where it is too dangerous to have electronics nearby.

[solutions]

On large tanks, they are floats that are magnetically coupled to the gauge across a gas/liquid barrier.

[SpaleKG]

Thanks to all on ideas. Specially to 'solution' and 'sarah90'. Primary my thoughts are about no major changes in current tank like to mount something inside tank.
The tank have already mechanical level meter (based on some magnets) and idea to use hes needle is pretty acceptable.
That idea is started from 'solution' and next thoughts can be directed to something like OCR recognition of position of scale or like attaching something to needle. 1st I need to try to open current level meter to see is it safe to break or re-mount glass on it to be able to access needle.

And yes, someone will say that original (current) mechanical level meter are not precise or can jam on some position (specially when the tank is almost empty) but I am looking for some cheap solution and no need to know how much exactly LPG is left. It is good to know when it will be near empty.

Simply remote reading. Maybe just to put simple camera (thanks to 'sarah90') That can be enough to see needle everytime. Camera + IC light and small LCD monitor will handle all what I need until I find some advanced (MCU or whatever based) solution for precise measurement too. That can include mounting sensors inside tank or whatever is needed but safe in first place.

This is image of my tank:

[David_1]

would ultrasonics work ?

[insomartin]

ultrasonics should work. if the gas is in liquid form. that should give a rough estimate of the liquid inside.
ultrasonics is also used for hazardous environment where sensors are irradiated with radiation and cause errors.
(measuring temperature using the speed of sound)
 

[user77]

I'd seen level indicator that use ultrasonic sensor, it works, you can buy one and unmount to get sensor.

[pickit2]

I'd seen level indicator that use ultrasonic sensor, it works, you can buy one and unmount to get sensor.

repeat of post 21 no value added to post.

topic locked