Arduino Input_pull up , Sensing water level with bare wires.

Other Hardware Home Automation
1

Hello All,

Mine is a home automation project. Trying to automate 1 water pump that turns ON and OFF based on the water levels in 2 tanks. 1 Tank is on Ground floor from where the motor sucks water and the other tank is on terrace where the water is being collected. So the pump should STOP when the top tank gets full AND also STOP when the bottom tank gets empty. I am not using any water level sensors but just using bare wires with arduino’s builtin input pullup functionality to achieve the intended. I have also done the programming part and have tested it and its working fine except one issue.

Details:-

There are 2 tanks, one is on my terrace (20 mtrs from arduino) and the other one is bottom tank (3 mtrs from arduino). I used the 2 input pins using builtin input_pull_up functionality and then connected 2 separate wires to both the pins , 1 wire goes to the bottom of bottom tank (To sense when its empty) and the other wire goes to the top of terrace tank (To sense when its full). Both the 2 wires going to the respective tanks are along with a ground wire from arduino. So now when the bottom tank is empty i.e. water NOT touching input + ground wire togethor , the respective pin goes ‘High’ and arduino turns off the motor (I programmed High = motor OFF). Same way when the terrace tank is full arduino knows that it is full and turns off the motor.

Problem:-

Everything is working perfectly for first 3-4 days and then the BOTTOM tank’s bare wires ‘SEEMS’ to get corroded. I did expect it to get corroded but 3-4 days was really fast. As I clean the wires it works for another 2-3 days and then again not working. I tried a fix by tying both wires i.e. ground and input to 2 steel nails and then used waterproofing putty to cover the wire joints on nails . Even with this fix it worked for just 3-4 days, however I did not see any corrosion trace on the steel nails this time. I am pretty sure that the issue is not with the coding part as whenever I face this issue I only have to shake the wire and then it picks the correct reading for next few minutes.

Do you think there is something else then corrosion that’s causing the problem as its faster then i expect ?
Do you have any suggestion for my project, since I am new to arduino ?
Do you think I should use hot glue gun instead of water proofing putty ?

Appreciate your help !

2

HI, your problem is electro-chemical and it's called electrolysis. The current through the wire has to pass through the water and in simple terms, electrons leave the wire, enter the water and then enter the second wire. At the entry and exit points, reduction/oxidation (redox) reactions occur which reduce the metal wire or nail or oxidises certain minerals in the water and deposits them on the surface of the wire or nail.

These reactions occur between water and nail or wire naturally. Passing a current through accelerates the reactions significantly, as you have noticed. The pace of the reactions depends on the chemical makeup of the water, the amount of current passing, the materials (wire or nail). Note that a similar degradation occurs when copper wire is present. It is marked by the presence of aqua or green coloured deposits and is commonly called verdigris. Verdigris, arises in the presence of oxygen, moisture and copper, and modifies the copper to produce CuSO(4) - copper sulphate (and other products). It also causes degradation of the conducting surfaces.

Simple probes will always suffer in this way. You can prolong the probes a little longer by:

  • Reducing the current
  • Only power the probe to test intermittently
  • use a reversing drive current (change the current direction from time to time)
  • use an ac supply
  • use float switches