There is a mistake (already fix it on post #38) It is not Y2/Y5/Y7/Y5 but Y2/Y5/Y7/Y4

I don't understand why I get 0.1v on lines Y2/Y5/Y7
I thought the reason of having 3.1v on line Y4, instead of 5v was because the potable water is not a perfect conductor….. But you might be right because although I don´t have metallic taps (everything in contact to water is plastic) I have just realized that I have a second circuit to heat the water through a heat interchange device which is metallic!!!!

I will look at the option you suggest (with pull-up resistors)

There is no reason to use a switch mode regulator. On a previous project I made, I used it because I was powering some 5v relays and some more sensors so I needed more power I found that this particular module was not heat up as much as others. Once I have them I am using them with no other reason.
As I explained on post #1, I don’t have any studies on electronics (I just like it). I will look at the linear power supply options to change as I suspect, by your comments is a better option. Any option/further suggestions will be very welcome!

However, the trial I made removing completely the DCDC module didn’t help (see post #26)

 

Could this replace my DCDC voltage regulator?
Would it be enough?

 

Yes that will work.

Les.

 

I my humble opinion, this has NO relationship with my problem..........

If the tank is metal then that is also a connection to the circuit providing an unknown input. and if switching on the lights changes things then it is clear that there are unknown connections having an effect. My question is why waste time and effort with an arduino when a fairly simple analog circuit can do a better job and be FAR more reliable?