RoV buoyancy

[hargard]

Hi
This is the info bit:-
I am no electronics buff.
I am into building RoV’s (Remotely Operated Vehicles) the underwater types.
I am making modifications to my mk2 RoV.
I will be putting inside a chamber to control the depth. This will be flooded with water or air to achieve neutral buoyancy.
I need to know what level the water level is inside the chamber.
I am using a Avr mega8535 chip to control the RoV.

Now for the Question.
If I stick two stainless steel bars vertically in the chamber one at each side and measure the voltage ADC input will this give me a indication of water level inside the chamber ??.
The measurement need not be critical just a indication.
Stainless as it is in seawater.

I am also open to any other ideas that you can offer.
Regards Hargard

[Gary Warner]

I understand the probes would be stainless, but what is the chamber made of? If it's plastic you can use two foil tape strips on the side and detect the capacitance changes (water level) in the chamber. This would be done with simple RC (resistance/capacitance) time discharge on a digital port if you want to avoid a lot of analog work.

[hargard]

Hi Gary
Yes the chambers are made of plastic.
RC (resistance/capacitance) time discharge on a digital port ?? can you explain more please :confused:
As I said "I am no electronics buff."

The reason I opted for stainless is to avoid corrosion from sea water.
Regards Hargard

[GlowFly]

I think Gary was referring to using external to the tank foil strips possibly on opposite sides of the vessel, and measuring changes in vessel capacitance as the water filled/emptied. That way no salt water resistant materials are required.
--
Steve

[hargard]

Thanks Gary and Steve

That sounds just the job for this application. :D
Ok 1 strip on each side of the chamber. Connected to what ??.
And how would I read it on the AVR chip ??
Gary talked about a digital port input. Pulsing input ?? :confused:

Please lead me to the light at the end of the tunnel :)
Regards Hargard

[Dan Baldwin]

You could connect the strip on one side of the tank to ground (- voltage to AVR chip), and the other would connect to an I/O on the AVR chip, AND a large pullup resistor (a 100K or so resistor connected between this pin and +5volts). You would first configure the pin as an output, set it to low, and leave it there long enough to discharge your "capacitor". You then set the pin as an input and see how long it takes for the pin to go above the threshold. The longer it takes to go high, the larger the capacitance, and the higher the level in the tank. The strips on the sides of the tank should probably be as large as possible to get as large a change in capacitance as you can. I did a test with a pill bottle with a foil strip on each side. The capacitance was about 4 pf with no water in the vial, and about 25 pf full.

Dan

[Acetronics]

Hi, hargard

The "capacitance" measuring is probably the easy way ... as this principle is used for model helis tanks.

Buuuuuut, the question for me here is .... : what size for the tank ???

I send you a small example ... and will post the "Analogic" version ... as soon as I have found it ( on the HDD ... but Where ??? )


Regards

Alain

Edit : here also

http://www.parallax.com/dl/docs/cols/nv/vol1/code/nv27.zip

[Gary Warner]

sorry I wasn't around to answer the questions. I think they have been addressed now.

To add, this has been done on 10 oz fuel tanks - that's the smallest I've heard of. Two strips of foil are placed very close and parallel to each other. This is on the outside of the tank. The fluid acts as the dielectric of a capacitor. Like was said before, use a large resistor (~100k-1meg) to attach to the port to ground. Setting the port High will be the ide state. Reading the port for a Low is the reading state. Because of the resistor, the capacitor you just made will take time to discharge. Polling the port for a Low state in a tight loop will be longer for high capacitance (full tank) and shorter for a low capacitance (empty tank). Count the number of polls it takes to get a low state or start/stop a hardware timer to get the length of time. This is one of the ways It can be done. Like said before, you could use a pullup arangement too. Either way, you get the capacitor charge/discharge over a short period of time. Here's a drawing:

[Rufe0]

You could use a variable resistor moved by a level arm that had a small ball attached to the end.

[Gary Warner]

You could use a variable resistor moved by a level arm that had a small ball attached to the end.
Not sure the variable resistor would care for the salt water.

[hargard]

Hi
Thank you all for the replies and suggestions.
The drawing is worth a thousand words :).
BTW the chamber is 180mm diameter and 130mm long.
I will start playing tonight :D
Regards Hargard

[Brandano]

the way this sort of things is achieved in chemical plants with HIGHLY corrosive substances is by having a float magnetically coupled with the sealed sensor, say an array of hall switches, a series of reed switches or a linear hall sensor. You could encase an array of hall switches in a thin plastic tube sealing them with a potting resin, and make a donut shaped float with a magnet inside out of styrofoam, I bet that would survive the sea water for a few years, and I think it's unlikely to jam even for pretty extreme attitudes.
another way is to make a potted coil, and have the float move a ferromagnetic core. That way you can measure the changes in inductance the same way a metal detector works. Double the coil and you have a nice variable transformer, feed an alternate current and measure the output dc voltage and you have a sealed transducer

[hargard]

Thanks for all the inputs.
What I did in the end is to use the two strips of foil on a thin tube mounted on the side of the main cylinder.
I then input the capacitor into a CD4060 chip(CMOS 14 Stage Ripple-Carry
Binary Counter/Divider and Oscillator)
And measured the oscillations with the M8535 chip.
Regards Antony(UK)