Electrician needed please! The water gives me shock!

fr3n0z · Nov 29, 2023

Hi guys,

II have a problem of 1 item putting current in the water. I live in a old 70's french house (220v).

Responsible of the leak is a ATO system made by me. Is a 12V membrane pump connected to a float switch and a PSU. The PSU is a 12v 5A PC style psu (with 3 pole wall plug).

The only solution that i've found is to put a titanium probe and that solve the problem of current going thru me (isn't that hard but if i have a cut in the hand and i'm bare feet, i definetly feel it and also quite annoying (i would not say strong, but if you don't expect it you surely pull out the hand fast)

What can cause the problem? Any idea in how to solve it? I've tried 3 different PSU and 2 new pump and the result is always the same.

The pump and the switch are as in photo:

GK3 · Nov 29, 2023

A grounding probe is going to be there to keep you safe I.e current should go through the probe to a proper ground and not through you. But this should be there to ground the tank and alert you of an issue, not to be continually running voltage to ground.

Have you removed the pump and switch from the water and verified the voltage also goes away? First step is to make 100% sure you are diagnosing the right piece of equipment.

Silly question but you aren’t submerging that pump, correct? If not then really the only source from the ato system could be the float switch.

fr3n0z · Nov 29, 2023

GK3 said:
A grounding probe is going to be there to keep you safe I.e current should go through the probe to a proper ground and not through you. But this should be there to ground the tank and alert you of an issue, not to be continually running voltage to ground.

Have you removed the pump and switch from the water and verified the voltage also goes away? First step is to make 100% sure you are diagnosing the right piece of equipment.

Silly question but you aren’t submerging that pump, correct? If not then really the only source from the ato system could be the float switch.

I'm 100% positive is the ATO. I've tried with multimeter and when plugged in it goes crazy. And yes, i'm not submerging that pump, that pump is like 3 feet away from tank and you are right, only float switch is in contact with water

What i was wondering is how it can leak current. i've changed also float switch and problem stay there.

GK3 · Nov 29, 2023

fr3n0z said:
I'm 100% positive is the ATO. I've tried with multimeter and when plugged in it goes crazy. And yes, i'm not submerging that pump, that pump is like 3 feet away from tank and you are right, only float switch is in contact with water

What i was wondering is how it can leak current. i've changed also float switch and problem stay there.

It’s a reed switch. So inside there is a switch that closes when a magnet inside the floating part gets near. I don’t see how it can leak a small amount of voltage and not just completely short. My only thought is that it’s not properly rated to turn a motor on and off. Have you checked the voltage/current specs on the switch vs. the motor?

It would make it more complex but using the float switch to turn a relay on and off that then turns the motor on and off would isolate the motor power from the water.

Fat Ninja · Nov 29, 2023

Get an electric tester then start shutting off equipment one by one. When you shut off the right equipment that is leaking electric in your tank, the meter will stop reading the electric current. Then you will know what piece is faulty and you can get rid of it. You could also have a bad ground wire in your electric outlet or breaker box in your home. Lol if you don't have a tester and you can feel the shock and dont mind it, have a buddy pull plugs until it stops shocking you. A little unorthodox but that's how I figured out my heater took a **** and was leaking electric.

fr3n0z · Nov 29, 2023

GK3 said:
It’s a reed switch. So inside there is a switch that closes when a magnet inside the floating part gets near. I don’t see how it can leak a small amount of voltage and not just completely short. My only thought is that it’s not properly rated to turn a motor on and off. Have you checked the voltage/current specs on the switch vs. the motor?

It would make it more complex but using the float switch to turn a relay on and off that then turns the motor on and off would isolate the motor power from the water.

i've got a Ssr lying around i'll give it a try right away

Paul B · Nov 29, 2023

fr3n0z said:
I'm 100% positive is the ATO. I've tried with multimeter and when plugged in it goes crazy.

I am a Master electrician. Exactly how many volts are reading on your meter to ground?
That pump runs on DC current and can't "induce" any voltage in the water.

If it was leaking actual voltage it would not run because at 12 volts the thing would slow down and probably stop. Also the pump is rated at 10 volts not 12 so you should only use 10 volts
(but that wouldn't mean much)

I do assume you are feeding the thing with pure dc and not pulsating DC.

fr3n0z · Nov 29, 2023

Paul B said:
I am a Master electrician. Exactly how many volts are reading on your meter to ground?
That pump runs on DC current and can't "induce" any voltage in the water.

If it was leaking actual voltage it would not run because at 12 volts the thing would slow down and probably stop. Also the pump is rated at 10 volts not 12 so you should only use 10 volts
(but that wouldn't mean much)

I do assume you are feeding the thing with pure dc and not pulsating DC.

Ok,

No probe and no 12v ATO (the responsible) 38v
Probe and NO 12v ATO 0
No Probe and 12V ATO 133v
Probe and 12v ATO 0.41v

Photo is purely to give an idea, i have not that one but a similar one and mine is rated at 12v

I am feeding it with a normal 12v PSU, i think is pure DC.

I have lyiing around a SSD relay, i've just installed it and nothing changed.

Crabby48 · Nov 29, 2023

I would use a volt meter in the tank and start unplugging things until you see the voltage drop on the meter

Paul B · Nov 29, 2023

I don't understand your readings to well but are you saying you are getting 133 volts in the tank to ground or even 38 volts in the tank to ground?

fr3n0z · Nov 29, 2023

Paul B said:
I don't understand your readings to well but are you saying you are getting 133 volts in the tank to ground or even 38 volts in the tank to ground?

Yes, you are right. If I take out the grounding probe and leave just the 12v ATO I am reading 133Volt. If I take out the 12v ATO is 38V. If I put the titanium probe I'm at 0.048V

Paul B · Nov 30, 2023

I can't tell from here what is going on but it is physically impossible for your 12 volt pump to put 133 volts into the tank. I wish my 12 volt pump could do that as I would power my house with it.

There is something else going on there. If it were real voltage say from your 220 volt line the circuit breaker would trip and nothing would work. That has to be induced voltage from your 220 volt line but I don't know how it is getting into your water.

I know it goes off if you remove the 12 volt pump but I can't understand it from here.

Boehmtown · Nov 30, 2023

Paul B said:
I can't tell from here what is going on but it is physically impossible for your 12 volt pump to put 133 volts into the tank. I wish my 12 volt pump could do that as I would power my house with it.

There is something else going on there. If it were real voltage say from your 220 volt line the circuit breaker would trip and nothing would work. That has to be induced voltage from your 220 volt line but I don't know how it is getting into your water.

I know it goes off if you remove the 12 volt pump but I can't understand it from here.

His PC power supply could be malfunctioning and outputting too much voltage?

fr3n0z · Nov 30, 2023

here there i

Paul B said:
I can't tell from here what is going on but it is physically impossible for your 12 volt pump to put 133 volts into the tank. I wish my 12 volt pump could do that as I would power my house with it.

There is something else going on there. If it were real voltage say from your 220 volt line the circuit breaker would trip and nothing would work. That has to be induced voltage from your 220 volt line but I don't know how it is getting into your water.

I know it goes off if you remove the 12 volt pump but I can't understand it from here.

Here there is a video of the problem with the readings on the meter. I may be reading it wrong, not sure. If i read in with the meter set on DC the 138v become 10v DC roughly.

Could be the electrical system of the house/building that is causing this problem? (it's a '70 building and system is not really up to today code...like i've just have ground in some plugs in the house (electric wall heater, kitchen and toilet...the tank is connected to a grounded plug tough)

Boehmtown said:
His PC power supply could be malfunctioning and outputting too much voltage?

I have tried few different power supply, at least 3. Same problem with all of them.

Paul B · Nov 30, 2023

fr3n0z said:
If i read in with the meter set on DC the 138v become 10v DC roughly.

That makes more sense but DC doesn't get induced into the water so it has to be the actual voltage from the pump. It won't trip any breaker if it is the 12 volts DC.

That pump (or ATO whatever it is) seems defective as no current is supposed to go into the water from it. You can test it if there is any metal on it like the shaft may be stainless steel. If it is, use your test meter from one of the feed wires to the shaft or any other metal you find on it.

fr3n0z · Nov 30, 2023

Paul B said:
That makes more sense but DC doesn't get induced into the water so it has to be the actual voltage from the pump. It won't trip any breaker if it is the 12 volts DC.

That pump (or ATO whatever it is) seems defective as no current is supposed to go into the water from it. You can test it if there is any metal on it like the shaft may be stainless steel. If it is, use your test meter from one of the feed wires to the shaft or any other metal you find on it.

Ok bi have changed the pump with a different model. Now without grounding probe, it gives 80v when the multimeter is set to AC and 2.5v when is set to DC. still can't understand what can cause this leak and i wonder if with this values what could be the problems.