Steve Maves and his wife haven't discovered the true source of electrocution hazard related in a recent Sherlock Ohms post, The Case of the Shocking Bathroom Countertop. That's the conclusion of this followup post, submitted by Dennis Darling, who believes they face a dangerous situation.
Many of the conditions and facts of Steve’s story have not been presented, probably because Steve believes he has found and corrected the problem. The first thing that comes to mind is "What is the material used for the countertop surface?" Many materials such as granite, quartz, concrete, tiles, and water-soaked wood are mildly but deadly conductive. What supports the countertop? Is it fastened to the wall? Older homes sometimes have metal lath in the plastered walls. If this is in contact with a metal electrical box, and the trim at some point is in contact with the lath in the walls, this could supply a conductive path.
Steve says he measured a voltage from the trim, but he doesn’t say where the other end of the voltmeter was connected. We are left to assume he measured to ground. That’s a problem. If the trim is a continuous loop (doubtful) around the counter, then the loop is shorted, and no voltage is generated at any point -- only a current that circulates in the loop. Assuming the loop is isolated from any other sources, no voltage can be measured to anything else, even ground.
If there is a gap in the loop, then it may be possible to measure an inducted voltage across the gap, but it would be very small, well under five volts. Measuring from the gapped loop (trim) to ground will still not provide a voltage reading, because there is no conductive path -- the other end of the loop goes nowhere. In short, whatever he measured, it is not inductive.
I have a similar toothbrush charger at home. I was concerned about shocks when I first purchased the unit, but it appears to be totally encapsulated and fully insulated. I'm sure Steve’s charger is built to the same standards. The only way to induce leakage between the charger and the trim would be to drive a large screw or spike through the charger into the trim.
The real clue is that having it plugged into an outlet near the countertop and its associated plumbing produces a voltage on the metal trim. The test would be to plug in a non-inductive device such as an incandescent night light into the same outlet and measure from the trim to same point used before. I've got $5 that says the voltage is again present at those points.
The current path is from the hot side of the outlet, through whatever load is plugged in, and to the neutral wiring. The wiring is not really neutral or at full ground potential, and it’s connected somehow through the countertop plumbing and then to the trim. This contact is to the plumbing at some point, maybe at or underneath the faucets. Contact could be intentional or from dirt and water or a corrosion "bloom." It is also possible that the outlet ground connection in question is connected to the water pipes intentionally.
Here is my best guess based on the limited information available. The electrical grounding system in that old house has been compromised. Sometimes in old homes the plumbing system is used as the ground connection, and the pipes are no longer at ground potential. Maybe a continuity strap has corroded or has been removed during repairs. Maybe a repair used plastic piping that has broken the ground connection. A new pump may have an insulated connection. Another possibility is that older homes sometimes use the flexible metal conduit protecting the hot and neutral wires as the ground conductor -- a really bad idea on so many levels. There is a reason modern codes require a separate ground wire.
The cure is to ensure that the electrical system ground and the water pipe grounds are independent (according to today's electrical code), proper, and that the grounding extends all the way to the outlets, and that the outlets are properly wired and the connections are solid.
By the way, a GFI (ground fault interrupter) in the circuit may not provide protection unless the ground connection is good. However, GFIs are an absolute necessity. It is my understanding that without a proper ground, a good modern GFI cannot be reset. I am not up to date on this, so consult a knowledgeable licensed electrician. Make sure all GFIs are tested regularly.
To illustrate and describe all the possibilities would take a small book. My bottom line advice is to find an electrician who can do a complete inspection and follow this problem to its source and correct it. The cure will likely take some time and will not be cheap. Old electrical systems = not cheap.
This entry was submitted by Dennis Darling and edited by Rob Spiegel.
Dennis Darling spent 20 years in the U.S. Navy as a cryptologic/communications technician in maintenance. He has over five years of training in electronics and ship/shore power systems. He installed (under supervision of a licensed electrician) residential electrical systems both before and after the Navy gig. He also has 15 years of experience in designing and installing electrical and communications systems in U.S. embassies and consulates around the world.
Tell us your experience in solving a knotty engineering problem. Send to Rob Spiegel for Sherlock Ohms.