Live Voltage at the Handrail on a 1998 Gunite Pool
A 1998 gunite pool (22,000 gallons) at a two-story stucco house. The owner reports a mild shock when he grips the stainless-steel handrail while standing barefoot on the wet deck at the shallow end; it happens only when the circulation pump is running. With the pump on, your meter reads 1.9 V AC between the handrail and a cast-iron deck drain cover, and 1.1 V AC between the pool water and the handrail. With the pump off, both readings fall to about 0.05 V. At the equipment pad, the pump motor's bonding lug is empty - no conductor attached - and the pump sits on a plastic isolation pad. A #8 solid copper conductor emerges from the ground near the heater and ends in soil about 3 ft away. There is no water-bond fitting in the skimmer throat or on any return line, and the handrail is set in a fiberglass anchor with no bonding conductor. The main panel is 40 ft away; a subpanel is mounted on the wall 6 ft from the pool edge.
1. What do the voltage readings establish, and what is the defect?
Read the worked answer
The readings are the case. A potential difference that appears only while the pump runs means two separate things: current is flowing through the pool's surroundings, and the metal and water that should sit at the same voltage are not bonded together. Equipotential bonding exists to keep water, deck, shell, handrail, and equipment at essentially one potential so a bather cannot bridge a gradient. A 1.9 V difference between rail and drain cover, and 1.1 V between water and rail, is a measurable gradient across touchable surfaces - the grid is broken or absent. Every gap in the setup is a required bond: the empty pump motor lug, the missing water-bond fitting (skimmer throat or return line), the unbonded handrail anchor, and the #8 conductor that terminates in soil instead of running continuously back to the pad. The 0.05 V with the pump off does not clear the pool; it only shows the source is intermittent. Bonding is also not grounding - an equipment ground conductor or a ground rod is not a substitute.
2. What do you recommend, and what should you not do?
Read the worked answer
Treat it as an immediate hazard: do not grip the rail, and advise no swimming until it is corrected. The repair has two independent parts, and doing only one leaves the pool unsafe. First, a licensed electrician must find and eliminate the source of current - a fault in the pump motor winding, a compromised equipment ground, or current arriving on the utility neutral or from a neighboring service. Bonding will equalize potential and hide the symptom, but it does not remove a fault. Second, the bonding grid must be completed per NEC 680.26 (as locally amended): #8 AWG solid copper, listed connectors, landed on the pump motor lug, a listed water-bond fitting, the handrail, the deck, and the shell, continuous back to the pad. Then re-measure water-to-deck and rail-to-deck with the pump running; the target is essentially zero. Test the GFCI protecting the pump circuit and report the result.
3. The seller asks whether driving a ground rod at the equipment pad would solve it. How do you answer?
Read the worked answer
No - and it can make the situation worse. Grounding provides a fault-current path back to the source so an overcurrent device opens; bonding equalizes potential among everything a bather can simultaneously touch. Driving a rod near the pad adds a parallel earth path and can create a new gradient between deck, water, and rail rather than removing the existing one. The measured problem is a potential difference between conductive parts, and the only fix for that is a continuous equipotential bond, coupled with removal of the stray current source. A rod also gives no assurance of a low-impedance path, since soil resistance varies widely. Recommend the electrician address both items and re-measure; if the gradient persists with the bond intact, the source has not been found.