I built a simple indicator for Water Level. attached image. Its working fine. But when the Point A touches the Water its giving bubbles in the terminal B(may be corrosion). How to avoid this? any hep...

Supply voltage : 6VDC

 

The bubbles are from the current flowing, A BJT transistor is probably not the best choice, try this circuit. http://www.redcircuits.com/Page103.htm

 

I do have a separate circuit for this water level using IC CD 4066 (Bilateral Switch). But compare to the transistor switch it is somewhat complicated as it needs to put more components and etc.. So thought of doing this simple Indicator..

 

Any time you have current flowing across the electrodes, this will happen. (see electrolysis) A float switch is one way to eliminate the problem. I am building the circuit I posted and will use sacrificial electrodes. It is only a alarm that my main sump pump has failed and need to be changed out. It uses no current on standby, so the battery life is extended.

 

If I use AC supply instead of DC will it be avoided? in Net somewhere I read AC will be corrosion free...?

 

I am not the best person to answer that, perhaps someone else has more knowledge there. I would not be messing around with mains voltage(your wall plugs), not safe.

 

This is the best AC coupled detector concept I've found: http://www.simplecircuitdiagram.com/2010/10/19/cd4093-water-level-sensor-detector/ (I can't find the web page of the original designer)
Low current and no electrolysis. I used the AC-coupling and voltage doubler concepts with a PICAXE microcontroller for an automatic Christmas tree stand water level control system.

Another interesting circuit for line voltage switching:
http://www.sentex.ca/~mec1995/circ/sensor1.htm

Ken

 

Yes, it will electrolyze with AC.

Are you familiar with Darlington transistors? Try using one of those and a 100KΩ resistor, it will slow thing down a lot (but it will still eventually corrode). Copper is a pretty reactive metal, you probably need to find a different electrode material. Some stainless (but not all) will hold up pretty good.

 

Yes, it will electrolyze with AC.
Some stainless (but not all) will hold up pretty good.

Yes, but minimally. 316L stainless steel works great. You can buy TIG filler rods from a welding supplier. Soldering wires to stainless is a problem. I first tin the end with silver solder. Then you can use standard electrical solder to attach the wires.

Ken

 

Hi Ken, I was a welder in another life, I once repaired a winch that had the main contacts corroded by Tig welding the new copper contact onto the steel using everdure rod. Push the bronze onto the copper as it has a much lower melting point. The guy never returned, so I must assume it lasted. might work with stainless...

 

That's also another reason to tin the end in with silver "hard" solder (I should have said). Copper near water will corrode eventually.

Ken

 

Thanks for all your input.. But here I have an another doubt. while I am using this type of circuit in water tank, will it be a problem to use the water for drinking purpose? as i am using wires into tank ..

 

As Bill said: "...it will electrolyze". This could put very small amounts heavy metal ions in the water. For a drinking water supply, I would not use I would not use a circuit that depends on ionizing conduction through the water. There are many other sensing methods: electromechanical, capacitive, pressure, optical, ultrasonic...


Ken

 

Can you give some idea about the methods u mentioned ? or can I get any link to those methods?

 

I wouldn't worry about some of the metals, such as iron. We already get lots of that anyhow.

The main thing is the amount of current. A Darlington transistor will reduce that dramatically, by a factor of 100 or so. Think µa instead of ma.

As for other methods, look at McMaster Carr catalogs online. These are catalogs full of machine parts, and include such things as proximity detectors, optical wet/dry sensors, and much more.

 

Bill,
I wasn't concerned about ionizing Iron...it's the Manganese, Chromium, Nickel, and Molybdenum in 316-SS that would concern me. And yes, uA currents would minimize that, and the AC would minimize electrode errosion. But, in most parts of the world the drinking water contamination is bad enough that anything this sensor added would be insignificant. Yes, I was nit-picking. So, a few heavy metal ions added to a 500L tank would probably not make a significant difference in the consumer's health.

shivaa,
The requests for liquid level sensors are common these forums. You can do a search here for other methods...and
Google: liquid level switch sensor indicator alarm monitor
and...
Google: liquid level circuit

Ken

PS: You have to love McMaster-Carr...order by 3PM and it's here by 10AM the next day...at ground rates!

 

I remember talking to the old hands about TACAMO, a airborne ELF communication system for subs. The transmitter screwed with their instruments something fierce, especially the gas gauge.

They wound up running a bunch of fibers to the tanks, and used light to sense when the fiber was covered. It wound up being a very analog appearing system, even though it was fundamentally digital.

 

Have a look at the attached; it's a 4093 quad Schmitt-trigger NAND gate used as a variable frequency pulse generator followed by a missing pulse detector.

A capacitor (C1) is formed from two plates that are 4" square (or roughly 12 square inches, the shape can be just about anything you could imagine) and spaced 0.1" apart. The plates should be conformal coated so that there is no contact between the water and the metal plates. This completely avoids the problem of electrolysis and corrosion. It also avoids the variable of the water's conductivity, which will be very low if the water is pure (no dissolved minerals such as salt, etc.)

One possibility of failure is if fungus or other organisms grow and attach themselves to the plates, holding water between them at all times. Viscosity of the water/capillary action may also be problematic; some experimentation will be necessary to determine the optimal spacing of the plates.

If there is no water between the plates, the pulse generator will operate at around 50kHz, as the dielectric constant of air is ~1. As water comes between the plates, the capacitance increases due to the high dielectric constant (~78). The frequency of pulses at Out decreases, causing C2 to be recharged less frequently. When C1 increases to about 1,400pF, Out2 starts giving short positive pulses. When C1 is completely covered with water, Out2 exhibits approximately a square wave.

Note that the inputs to the unused 4093 gates (not shown) should be connected to either Vdd or GND. Note also that a 0.1uF bypass capacitor between the 4093's Vdd and GND terminals is required (not shown).