Water level controller

Hi,
I have designed the water level controller for my tank. Prototype is working good. But the problem is, I have used conductivity method.
I have used simple telephone cables for sensing the water level. Due to corrosion, it wont last long.

I have read, that using stainless steel rod, the corrosion can be much reduced. But the problem will occurs slowly.
Also i have read in similar thread, using AC supply to the rod, corrosion can be reduced.

But i want some good solution to solve this.
Float sensors can be best choice??? or is there any other sensors??

I'm designing it for very low cost. How to make simple float sensor???

Can anyone help me to solve this issue.

Thanks in advance

I usually use chrome-coated probes as water sensors.
And to eliminate polarisation on the probes, I use an AC low voltage (I add a capacitor in series to insure that the average current is zero). The AC source should be relatively weak (its resistance is high). When the two probes are shorted by water, the weak AC voltage drops to a low level that signals the presense of water.

I have used thermistors that have a current set to cause self heating.
The operating temperature of the thermistor is set to be above ambient and the resistance is fairly stable for ambient temperature variations experienced indoors.
But as soon as the thermistor comes into contact with fluids it is cooled quickly due to the thermal conductivity of the fluid and its resistance changes substantially.

Key good points are:
it works with none conductive fluids
thermistors can be purchased encapuslated in glass hence chemical resistance.
Small size means easy to locate, works with very small volumes.

bad points are:
It needs to have known operating ambient temperature ranges for the fluid and ambient to be able to set the thermistor temperature reliably.
It can be affected by air currents which will cause cooling

This method has been used as liquid level sensors in laboratory vessels and leak detectors in bund trays and around instrument tubing connections which can be prone to leak when the injectors block.

This is a bit extreame for your question, but I thought it presented an alternave perspective on fluid level sensing, particularly for smaller volumes.

hemnath said:

Hi,
I have designed the water level controller for my tank. Prototype is working good. But the problem is, I have used conductivity method.
I have used simple telephone cables for sensing the water level. Due to corrosion, it wont last long.

I have read, that using stainless steel rod, the corrosion can be much reduced. But the problem will occurs slowly.
Also i have read in similar thread, using AC supply to the rod, corrosion can be reduced.

But i want some good solution to solve this.
Float sensors can be best choice??? or is there any other sensors??

I'm designing it for very low cost. How to make simple float sensor???

Can anyone help me to solve this issue.

Thanks in advance

Click to expand...

Hi,

If using stainless steel try and use 316 Grade, also known as food grade, it withstands corrosion better.

Have sometimes found a low voltage, low current AC circuit from a 741 can have difficulty sensing pure water in an enclosed tank situation.

Making your own float switch is possible, but decent all plastic ones are so cheap its hardly worth it.


If you can find one cheap enough the optical fluid sensors are another option.


Think Pinouts idea using a thermistor sounds interesting, will have to try that out.

Use like this: . Its a water level controller. Working as:



It works just like a relay. You can use this like a sensor or directly with the motor(if motor is low hp).

The software is slightly more complicated but you can do it by immersing a sealed plastic tube into the water from above, around it you place a floating torus shaped magnet ( a ring magnet in a plastic donut! ) and inside the tube a line of reed switches. You can multiplex the reeds to save on wiring. As the magnet floats up and down around the tube it operates the reeds in sequence. It is completely water tight and corrosion proof but might cost a little more. It also gives more accurate level reading rather than just a few fixed points. With care it might be possible to simplify it to pairs of reeds at fixed intervals, one reed being slightly above it's partner, it can then sense depth and whether the level is rising or falling. The output is digital which helps but you could in theory place a resistor across each reed switch and calculate the water level from the total remaining resistance, making it a two wire analog system.

Brian.

Good ideas being posted and just thought of another.

As used in washing machines as the fill sensor.
It is a low pressure switch with a tube immersed in the water as the water rises the head pressure increases the pressure in the tube operating the switch.

Thank you all for your valuable suggestions.