Circuit design - induction based liquid level meter

Hello circuit gurus.

I am trying to come up with a simple circuit design for the induction based meter of the liquid level in the tank.

The idea is simple:
- on the input side, a coil is placed in the tank,
- in the output side is a gauge, showing the liquid's level.
As the liquid level in the tank fluctuates, the submerged portion of the coil will change, affecting its impedance, which will be reflected by the gauge.

I am thinking of two possible principles:
1). Coil based oscillator - its frequency changes with the liquid level, and the gauge indicates the frequency
2). Fixed frequency alternating current passes through the coil, its impedance changes with the level change, and gauge indicates the current flowing through the coil

Other sensor types were considered: open contact, float, capacitive, optical/IR, sonar, etc., but they cannot be used because it is a moving dirty liquid (sewage) tank on the boat, and whatever sensors are placed there will become clogged/obstructed, as the dirty liquid splashes and dries over the sensors.
Obviously, the circuit will have to be calibrated and adding a loose ferrous floater inside the coil is an option.

Any help (ideas, circuit drawings, etc.) will be appreciated.
Thank you in advance.

Captain Alex

an waterproof coated iron bar sliding in the coil would impact the frequency response of the coil, not sure about water since its only slightly magnetic. you dont want the water touching the wires furshure.

i did similar with a capacitor circuit where i took a piece of pvc pipe, glued a plastic cap on it to waterproof it, then stuffed it full of tinfoil for the anode. i then used another piece of tinfoil as cathode to the water in the tank.

I used a microcontroller to apply a voltage to the anode then flip the pin to read the capacitor discharge time, then i would flip the voltage to the cathode charge the cap the other way, then flip that pin to read the discharge time.

the plus about using an "alternating current" is that it minimizes the effect of corrosion of the plates

Dr_Doggy,

Interesting - so u used tank as electrolytic capacitor, and amount of liquid/ "electrolyte" there was changing its capacitance. In this case, it could be part of the oscillator circuit as well. Sealed foil solution is appealing.
I have aluminum tank, hence it can be used as one of the contacts itself. Also, measurements will be taken occasionally and momentarily, so corrosion is not an issue. My concern is consistency of readings, as surface area will become covered inconsistently.

Thanks for the idea.

Hi Argalax

Welcome to ETO

We use a fabric conditioning liquid which has a varying consistency, stored in IBC's .. .. I tried various methods of keeping track of content and eventually settled on this .. ..

which bounces a sound wave off the surface of the contents and calculates the content from a fixed starting point. No contact, no contamination and reasonably accurate. Uses a simple Arduino sketch and a HC-SR04 module.

Just a thought ?

MM

MM,
Thanks for the idea
Will it work if both sonic emitter and sensor are covered in a fluid?

argalax said:

MM,
Thanks for the idea
Will it work if both sonic emitter and sensor are covered in a fluid?

Click to expand...

I would have thought not, fluid would ruin the transducers.

I think I'd look at placing one or more strain gauge(s) under the tank, between the tank and its supports. Effectively weighing the contents.

With a high enough sample rate and a moving average calculation it should be possible to eliminate the effects of the contents moving about.

The Australian Silicon Chip magazine has published several Water Tank meters in April 2002, July 2011, November 2007, September 2009 & February 2018. One of the later ones (I can find which one if you wish as I have every copy since 2001) employed an air pressure sensor. This seemed to me to be the best. Another had an ultra sonic detector.

argalax said:

Will it work if both sonic emitter and sensor are covered in a fluid?

Click to expand...

Nigel Goodwin said:

I would have thought not, fluid would ruin the transducers.

Click to expand...

Nigel is right, of course, but there isn't any need for such an event.

I used a piece of tube with the sensor mounted in the top and a wooden 'float' at the bottom. I presume you are mostly interested in when the tank is approaching capacity ?

MM

Thanks for replies.

Weighing the tank is a very good idea, I have considered it as well. Unfortunately my tank (as most of tanks on boats) are integrated with the hull,. and weighing it is not feasible.

Pressurising wouldn't work, as it is vented outside and would not hold pressure.

BTW, it is a challenging problem, and to my knowledge, nobody have solved it elegantly (at least for general boating market). There are plenty solutions for most fluids (water, petro fuel, etc) but not for a moving sewage holding tank with splashing, drying and splashing stuff all over inside - all moving parts soon become clogged and stuck, and other types of sensors become covered with inconsistent changing layer of wet crap.

Yes, I am mostly interested to know when tank is getting full, perhaps 3/4.

well actually the tank water is the cathode, regardless of volume. it is the amount of surface area of the sealed pipe filled with foil submerged that impacted the capacitance.

to compensate for splashing may i suggest averaging values sampled

argalax said:

BTW, it is a challenging problem, and to my knowledge, nobody have solved it elegantly (at least for general boating market).

Click to expand...

Why wouldn't you use one of these resistance sensors?:

If the tank is metal then you may be able to "hit" it with a transducer of some sort and measure the frequency it vibrates at. The air volume will cause a particular resonant frequency which will vary with content. The "contents" may have a dampening effect as well so measuring the length of tone could also be a useful indicator.

Mike.

argalax said:

BTW, it is a challenging problem, and to my knowledge, nobody have solved it elegantly (at least for general boating market).

Click to expand...

I don't believe that's true .. .. . I've seen a few versions of foul water storage with level indicators that work. They are common in the caravan & mobile home industry.

Personally, I'd think more about solutions and less about problems .. .. .. .and besides, you've got some quite mythical options to consider

MM

Here's another solution for you:

(http://imgur.com/99KPRDF)
Looking around, most marine waste tanks have a maintenance/inspection cover about Ø40mm held in by 5 bolts.

If the underside of that cover had a boss, and in that boss was a rod with a float attached; when the fluid level rises, float will apply a force to the rod proportional to the level of fluid in the tank.

The force on the rod will act to distort the surface of the cap. Add 3 strain gauges symmetrically around the cap and they will measure that force. If the fluid is static, the force from any one sensor will be proportional to the fluid level.

If the fluid is moving around the 3 strain gauges will allow a differential calculation to eliminate the effects of the movement and derive a single 'fluid level' proportional figure.

Dr_Doggy said:

an waterproof coated iron bar sliding in the coil would impact the frequency response of the coil, not sure about water since its only slightly magnetic. you dont want the water touching the wires furshure.

i did similar with a capacitor circuit where i took a piece of pvc pipe, glued a plastic cap on it to waterproof it, then stuffed it full of tinfoil for the anode. i then used another piece of tinfoil as cathode to the water in the tank.

I used a microcontroller to apply a voltage to the anode then flip the pin to read the capacitor discharge time, then i would flip the voltage to the cathode charge the cap the other way, then flip that pin to read the discharge time.

the plus about using an "alternating current" is that it minimizes the effect of corrosion of the plates

Click to expand...

Hi Alex
More than 20 years ago a friend of mine build some successful sensor based on what you mention.

Just in case it could inspire you, this is what I've built recently and described here (thanks again JimB)

What about just "thumping" the tank and measuring the resonant frequency of the void (air cavity) portion of the tank itself? The sensors could be on the outside of the tank. Just a microphone to listen and a speaker/piezo to make a thump. Look up helmholtz resonator.

Might work if the sea is dead calm, but harder when the contents are slopping around in even a mild swell.

Beau Schwabe said:

What about just "thumping" the tank and measuring the resonant frequency of the void (air cavity) portion of the tank itself? The sensors could be on the outside of the tank. Just a microphone to listen and a speaker/piezo to make a thump. Look up helmholtz resonator.

Click to expand...

Great minds think alike. (see post #13 above).

Mike.

Hm, did not consider "thumping" from the outside, thought measuring feedback would be complicated.
That may work - I actually knock the tank trying to guess how full it is .

Will appreciate if anyone suggest / point to a relevant schematic, with suitable parts (transducer, mike, etc.).

Thanks!