After investing about 40 hours or more of my own time (or more, along with
@mas985) across 190+ posts in that other shock/tingle thread a few months ago I’m almost afraid to ‘wade into’ this one. Sorry for the pun! But this can be so frustrating for folks, what the heck….. I think
@ajw22 and
@JohnT, and others have given you some solid advice and starting points. Absolute NO to grounding rods. And a probable NO to ripping out the deck, at least not until it’s been methodically proven to be the culprit. And the deck seems an unlikely culprit anyway, at least to me, for several reasons, at least thus far. Hopefully that’s a bit reassuring at this point.
As others stated, the whole point of equipotential bonding is to make all points (water, rails, ladders, lights, fences, cages, and deck up to 3 feet out from the pool) at the same voltage level, pretty much regardless of what that level is. Then you can’t get shocked no matter what the neutral is carrying from the power company. All that described in gory detail in the video I’ll reference again below. To make that very point, in the video they actually raised the pool and all equipment to 120 volts or so (not to be tried at home) but nobody got shocked because it was all bonded to the same level. So, since the power folks were already out, you’re probably done chasing them too. Neutral voltages / NEV happens – we get around it with bonding.
I believe what
@ajw22 means by incremental testing of each structure is to test each of those pool system components to find out which ones are not bonded together and which ones are not bonded back to the pool equipment at the pump/pad. In other words, again, individually testing each rail, ladder, light, cage, fence, any other metallic surfaces, as well as the deck at several points, water at several points, each tested separately to determine if each is bonded together and all the way back to the equipment pad. Test the easy ones first, the water and deck will be more confusing.
There are several ways to conduct those incremental tests. But I would first suggest, if you have not already done so, watch that video
@ajw22 posted (linked again below), a few times until you grasp it well, then proceed to test. Consider also reading through the test plan referenced as a document in the “more” section of youtube info about the video. You can test with a long length of #8 copper wire tied solidly back to the pump/pad bonding and use a meter to measure resistance to various components and structures, and/or voltage to each, or simply use @ajw22’s suggestion to attach that temporary wire to each structure, one at a time, then to all, to see if shocks cease. I would test all those ways if I were doing it. Voltage tests, resistance tests, then bond each with the temporary wire to confirm (always #8 solid wire), also write down and log every step, etc. And remember there could be more than one culprit.
Referencing that former tingles thread again, we determined that at least one rail and ladder were not bonded. In theory, that would have allowed portions of the water to rise to different voltage levels compared to the bonded rail, thus offering tingles between the water and the bonded rail - and perhaps the deck too. We never heard back from the OP about final tests to the light.
Please also remember: When reporting voltage, resistance or current always specify the exact two points you’re measuring between – otherwise we’re lost, or worse we draw incorrect conclusions. Best of luck! Don’t worry, it's not easy but it's solvable!
The video is here:
YouTube