Need a proximitiy sensor only detect liquid not foam! Any suggestions? Recommendations?

If all you need to know is when the gallon is 100% full, you could use a "see-saw" and a micro switch, with a full jug as the counter balance.

| |

Or a pressure transducer in (or coupled to) the base of the liquid. That would be independent of the tank size and shape.

I believe he meant that the transducer would be immersed in the fluid and located at its bottom.

I wonder if the OP has considered using a vacuum hose with its orifice located at the desired level of the fluid? It would have the benefit of clearing away at least some of the foam (which could be returned to the source tank to settle). As soon as the liquid level reaches the hose there should be a marked decrease in pressure in it which should be detectable by flexure or pressure or load current change in the pump, or optical detection of the fluid in the (transparent) hose.

Ultrasonic transducer?

I meant it would be dependent only on the height of the liquid.

If you are looking for a "calibrated gallon" rather than "an inch from the top" then the weighing idea seems better.

Or look at the technology used on the delivery nozzles at gasoline pumps. Those have to be pretty idiot proof, survive rough handling, temperature extremes, etc. and do use a vacuum, albeit one produced by a venturi in the nozzle line. Strictly mechanical and insensitive to flow rate (to a point) but it does require that the nozzle end and sense port contact the fluid in the container as it rises which I guess could be a problem in some scenarios.

=A0

Yes that could work, but the jug has a very small opening, plus the fluid is going in as it is filled (only 1.5 in opening). I thought of the same thing but measuring the foam level, which will rise within a small tube close to its mouth, thus I will measure the foam level that is corresponded with a liquid level of "full gallon". but i need to check if the foam rises uniformly.

=A0

I don;t know what you mean by the apparatus, but is simple machine. Liquid "detergent" is located on a tank above. Tank has 8 valves. User puts gallon under valve, then a switch is activated, and liquid falls, filling the gallon. Sensor stops the gallon from overflowing. Works with water, but not with "detergent". A sensor that is less sensitive will work nicely, and will be straightforward, if not then I will try differentiatiing the different foam/ liquid capacitances...

at

sound like a good idea.

Typical jug & bottle filling machines extend a tube to the bottom of the jug which reduces foam formation (and associated product waste) on filling. Insert jug, tube drops, jug fills, tube retracts. Manual or automatic depending on scale and budget.

You appear to underestimate what you can cram in a 1.5 inch opening - a half-inch fill tube with a 1-inch float (annular ring) would suit the need very simply/inexpensively, though product weight or a metered pump are more typical approaches.

The machines I've seen doing this insert the fill tube almost to the bottom so that foaming is minimized, also they have a mechanical rotary stage "pipeline" (in the computer science sense) arrangement that allows a number of bottles to be filled in parallel, thus reducing the average fill rate for a given production rate.

I'd say, "thump", like you do to a melon. Then just measure the frequency of the main return signal. ;-)

It'd be cool to take one of his containers, start "thump"-ing it, and just listen as it reaches the 1 gal. mark.

Cheers!

Rich

Load cell ?

Yes, that'd work. In either case, some attention to shielding the plates (so a hand nearby won't trigger the sensor) is worth while, too.

The capacitive sensor will not always reject foam IF the liquid is electrically conductive, though.

Bill Sloman's milk-tanker approach detects not just the capacitance, but the series resistance in the capacitance, and will work best for electrically conductive liquids (the foam MIGHT exclude electric fields if it's conductive enough, but it WON'T be a good conductor). If I read his suggestion correctly, it used synchronous detection to reject the capacitance-change part of the signal, so it won't work well if the liquid is a nonconductor, unless you have a switch on the synchronous detector to select between the two cases.

There's nothing to stop you detecting the in-phase (resistive) and quadrature (reactive) components separately and independently; getting the in-phase and quadrature waveforms to drive separate demodulators requires some ingenuity, but it is perfectly practical.

-- Bill Sloman, Nijmegen

Float and a Hall effect pickup? Probably not moving fast enough.

Could you use a small laser? For clear liquids, have it redirect to a target or parabolic mirror, for opaque, it would cut the beam entirely. You'd have to tune the cutoff signature some.

FMC sells positive displacement meters. Those probably cost too much, but they're common in fuel dispenser applications. The ones I used to interface to were good to fairly small increments. Met W&M forever.

-- Les Cargill

But those are overfill detectors, not something that is accountable for volume and must comply with Weights and Measures. They are ingenious, though.

SFAIK, all fuel dispensers use positive displacement flow meters for measuring the volume sold.

-- Les Cargill

It'll be biased by the action of the liquid, unless you can arrange for flow to be create no pressure in the ... z axis.

Might be something you could cal. out...

-- Les Cargill