Had a call from a homeowner regarding blueish stains in the basins of a house I wired about 15 months ago.
She has a contractor friend that told her he had the same problem and it was "2 volts on the water causing electrolysis>
I have been burning up the internet looking for information about copper in the water supply and have learned a bit about cuprosolvency, galvanic reaction, and electrolysis.
What I find is that it can be very much a gray area with a lot of finger pointing.
I can't get out there until Saturday and I was wondering if anyone had any tips for what to look for so as to either prove or disprove that my installation is at fault.
I have read that having both ac voltage and current on the waterline with respect to the disconnected bonding conductor is normal but a dc offset could be a culprit.
I have also read that blueish water can be normal for the first year or so of a new copper piping system. I have read about low PH, high PH, oxygen content, and some other stuff that I can't remember now.

The electrical system is as follow:
A pole mounted 200A single phase overhead service feeding a 200A disconnect. 5/8" 8' ground rod at pole. 4 wire 3/0 copper feeders (w/ 1/0 EGC) ran underground 100' to a jbox that connects the copper to AL 4/0 SER (using NSI connectors) for about 30' to a 200A sub in laundry room. The jbox also holds the crimp on splice I had to make to extend the #4 bare copper GEC that clamps to the bottom bar of the footing rebar to the laundry room (no bond between neutral and ground buses). (I originally planned on the service being underground to the exterior wall and set up accordingly but that plan changed).
A #4 bare copper runs from the subs ground buss to the nearest accessible bonding point for both water and gas (they make us bond the gas here) which is about 20' away in the attic near the crawl hole.
The piping is all copper and run overhead (I seem to recall) except for about 15' ran under the slab to a kitchen island. I think it goes pvc before it enters the ground heading to the well.
There is a 20A solar system backfeeding a remote sub that was installed by a solar contractor. I also installed a tile floor heating system in the master bath.

What procedure would you use for determining if electrolysis could be the cause of the blue water?

City gas? If so, TEMPERARILY disconnect the bonding wire between the gas line and the rest of the house. Measure for a offset voltage between the gas line and the rest of the house. There might be an issue with the active galvanic protection system of the gas line.

Larry C

Measure for a offset voltage between the gas line and the rest of the house. There might be an issue with the active galvanic protection system of the gas line.


Larry, could you expand on this a little please.

If this is a new install (gas) PG&E uses Non metallic gas line and if a older one, there should be a non metallic bushing to isolate piping systems (installed by PG&E gas crew).

It is my understanding that utilities will sometimes inject a small DC voltage between the steel piping and a sacraficial anode to ensure that the anode will electrolyse away instead of the steel pipe. On ocasions someone will inadvertantly reverse the polarity.

Back around 1990, a high end housing developement where Jose Conseco, the baseball player, lived when he was playing for Oakland, was suffering from blue water which was traced back to a miswired galvanic corrosion system.

So, when the gas piping bonding is disconnected from the building's ground system, AND a DC voltage is present, then a galvanic protection system is present, and it MIGHT be the cause of the blue water.

Larry C

Thanks for the response.
It's a semi rural area and the gas is a propane tank but there is a steel line running underground to it. I'll check that too.

Is that gas line acting as your best ground electrode? I think I would start out by making sure you have a good ground electrode system and that all the piping is bonded. If you shunt out all the "batteries" you won't have any electrolysis.

I wouldn't think the gas is better than the ufer.
The gas is wrapped in 3M tape and the ufer ties to several hundred feet of rebar in the footing. But I could be wrong and will check. I'll check for dissimilar metal connections as well.
I suspect chemical erosion of the copper lines by the water softener. The water softener salesman wants to sell her a $700 filter to remove the copper from the water. That's why she talked to her contractor friend and on his advise called me.

I can't confirm 25 ohms at the ufer but what I've read leads me to beleive the resistance of a ufer system is going to be about as low as I'm likley to get. Worst case is to drive 1 additional rod and call it good. All bonding wires originate from the sub panel ground bus as does the buildings GEC. There are no other bonding jumpers other than mixing valves. To my knowledge the only point the copper goes underground is at the kitchen island.
There are also 2 on demand gas water heaters, one at each end of the house. One of them uses a 120V igniter and the other uses flow to generate its own ignition source. Maybe some bonding going on in there? I'll check everything I can think of in the morning.

The Ufer should be your best electrode if it is done right but you are having that kind of problem, it appears.
The other thing that may not be your problem at all is water quality. Do you have a pH kit? Even one like you use in a pool would give you an indication. Low pH water will eat copper too.

Well I went out this morning and here is what I found.
There is no measurable ac current flowing on any of the water, gas, GEC, or bonding wires. There is some current on the EGC going back to the main that I have isolated to 3 circuits. One is the office plugs circuit and is only a few miliamps. Another is the entertainment circuit and is also only a few miliamps. The third is an exterior 125A sub that I took an AC, a couple of general lighting circuits and the well off off.
Actually, I temporarily back fed the old existing house from this panel and it is providing power to the well for now. This circuit was good for about an amp flowing on the main EGC when energized. I expect that to go away when we install the permanent well feeders.

I measure about 5 milivolts DC between the disconnected water bonding wire and the clamp. It goes down to 2.5 when I plug in the water heater ignitor.

The water softener guys ran a low volt wire for their xformer and poked it through the stucco of the (BRAND NEW) house and plugged in their xformer. I forgot to unplug it when testing for voltage between the bond and the pipe but I wonder if it could be the source of the DC. Next time out I'll check. Why they did this when their softener is sitting right next to a surface mount sub panel with lots of room in it is beyond me. I'll fix that when I run the permanent well feeders.

The only dissimilar metals I noticed were at the softener and looked cast nickel. Bright and shiny. And at the water heaters that looked like stainless going to stainless hoses.
No isolating fittings on either.

She's going to get the water tested both pre and post softener.

I ask you: In the absence of any chemical explanation, is 5 milivolts enough to cause the blue water?

Oh yeah, the 5 mili was only measured with the main on.
Main off was 0.

Take care with your analysis. If you have chemically induced corrosion, this will generate a voltage, typically 1 to 2 volts dc. Chicken/Egg!

As Greg says ph is the key. All metal corrosion involves electrical phenomena, not all of which may be measurable, due to closed-loop circuits.

Water softening creates potential corrosion scenarios.
Hard salt ions like calcium replaced by sodium ions reduce limescale deposition, but this also prevents protective layers of lime being deposited on the copper.
The softener resins are occasionally backflushed with common-salt brine to remove the calcium from the system. Any brine getting back into the pipework would cause corrosion, although that's unlikely.

Another common problem is cement or concrete. A copper pipe in contact with, or cast into, concrete or cement mortar will probably perforate in less than 10 years. This corrosion will induce local voltage, which in turn can induce copper salts to form in the pipe bores, if the ph is not neutral.

An AMP is alot of RF filters! Either you found your source of current or something is not right in the old house. Second neutral-earth bond? I suspect that the well is acting as the earthing electrode.

If that is the case, I would recommend keeping a spare well pump on site. Lightning loves low resistance bonding electrodes AKA well pump motors.

The amp on the GEC is roughly equal to the neutral current so I suspect re grounding in the old house somewhere. That house should be getting knocked down soon so I haven't spent any of their money looking into it.

I'm still waiting to hear the results of the water tests.