We have a new construction home that I am installing a whirlpool tub in. I have ran a 12/2 with ground dedicated circuit that will have a gfi on the outlet that the tub will plug into. My questions are:

1. The entire house is ran in pex what do I ground the motor ground lug to on the motor itself?

2. The directions say "The circuit must be a (3) wire circuit from the electrical supply panel. A grounded neutral wire and a third wire, earth ground, are essential."
Is this what I already have with the 12/2 with ground? The wording of that confuses me for some reason.

If it matters the home has no uffer ground, only 2 ground rods will be installed.

To be clear, we are talking about a hydro massage bath tub, not a spa?

If there is no metal around the tub anywhere that needs to be bonded, you may not be using that lug. This is not to connect the pump to the grounding system or any electrode. It is only for equipotential bonding of all conductive parts in the vicinity of the tub.

1) A properly functioning GFCI receptacle is going to be protection.

2) But GFCI's -- especially exposed to the elements -- fail all of the time. That's why one is expected to test them at least once a month.

3) If the GFCI has failed then there is a small, non-zero, risk that the insulation within the motor could breakdown.

4) Due to human sweat, every tub or spa gradually becomes salty -- and quite conductive.

5) The very act of pumping a conducting fluid through an electric field induces a current -- and voltage. This happens even if everything is totally squared away. Very few tub designs shield the circulating water from this effect. [Faraday discovered it, BTW.]

6) Consequently one is well advised to bond every hot tub/ spa to ground with any factory supplied lug. It almost certainly is doing double duty: bleeding off induced voltage (per 5) and providing a back-up short circuit path back to the panel, lest the GFCI fail in use.

The way this is done in elevator pits (notoriously wet) is that a bonding conductor (green) is wirenutted inside a nearby junction box (W/P, Bell box) and brought out to the equipment lug. Depending upon the circumstances, this whip can be naked (a forgiving situation) or enclosed in a raceway for protection from physical damage.

BTW the voltage effect can build and build. I've tested waters that had reached 45 VDC above the local ground voltage. Every time anyone exited the tub, they got a jolt to their groin. A trivial bonding run cured the problem in no time.

BTW, the national 'help' line for the manufacturer was totally unaware of the physics of their product: not a clue.

Their design had no mechanism for bleeding off the induced charge. (It was building up like a gigantic Leyden jar.)

http://www.instructables.com/id/Make-A-Water-Leyden-Jar/

The code specifically says the equipotential bonding is not connected to the electrodes but it always will be, simply by the construction of the equipment.
A plastic tub, connected with plastic pipe, no heater and no metal nearby leaves very little to bond.

Exactly.

My B-i-L had EXACTLY such a hot tub.

In pretty short order, he -- and family -- were being shocked something fierce. (45 Volts measured, Fluke meter.)

A draining ground eliminated all hazards.

Check out my link.

The NEC has missed this: human sweat = salty water.

As time goes by, the salt concentration only goes UP.

You end up with a salt-water Leyden jar.

The BETTER the insulation/ non-conduction -- the more extreme the voltage can build.

Yeap. It can get to alarming thresholds.

It's commented upon time and again. The link to the physics is lost.

Everyone thinks that the induced voltage is either a phantom or a loose wire.

Nope!

Hence, the requirement to 'bond the water'.

I am still not sure what kind of tub we are talking about here. If it is just a jetted bath tub, the bond the water requirement is not there, at least not in the 2011 (680 part VII). You are also not bonding the tile floor. You do bond all existing metal parts in contact with the water but you do not need to add a bonding fitting if it is not there. There is no part of the motor shaft exposed to the water in the pump. If the plumbing is plastic and no heater, I am not sure how you bond the water.
Obviously if this is a "spa", we get a lot more serious about it and the hazards are virtually identical.

In case I've not been clear: the water in the tub, itself, is a conductor.

No existing design totally protects the pumped water from the magnetic field generated by the pump motor.

It's counter-intuitive, but even an AC motor produces a DC bias in its field. (See Buss and the short circuit fault calculations in Ugly's.) This is due to the material physics of the steel. No motor/ transformer steel is ever cast/ rolled without a bias in its magnetic grain.

The energy bleed/ pump from this effect is trivial. However, as a Leyden jar, the energy simply builds and builds -- hour after hour.

It doesn't kill anyone: it just gives them a very painful jolt as they step out of the salt bath and onto the very grounded exterior world.

And, yes, the manufacturers (and NEC) have totally missed this effect.

I have tested its voltage in the real world. This is the source of COUNTLESS phantom voltage service calls that befuddle electricians across the land. They keep looking for METALLIC conductors as the source of the voltage.

They fail to recognize that the circulating salt water has become a Separately Derived Service -- a Leyden jar of a huge scale.

Visit the link up top to get an idea of the physics.

The hot tub/ spa may start out as fresh, potable water. It does not stay that way with sweaty human users. The salt can never boil out/ evaporate from the solution. It starts out trivially low, and proceeds to build and build.

Without a doubt, these tubs were tested with fresh, pure, water. It is not a conductor, and will not create the Leyden jar effect.

ALL of the current designs on the market have this conceptual flaw. They all need to have a metallic bond running through to the water -- taking the voltage back down to the local equipotential plane.

When this is done, the problem entirely goes away. The actual energy stored by these Leyden jars is trivial. It's just that no-one enjoys discharging it through their groin, which is where the sensation is the most intense.

It gets bad enough that owners stop using their hot tubs. This is why you see so many units scarcely being used.

It's a rare individual that admits they're getting so jolted. I had to experience it myself. Then, when queried, everyone admitted it had happened to them, too. It was simply too personal to bring up in conversation.

So don't be surprised if troubled customers don't broach the topic, either. It's taboo. It's also a real problem faced by millions.

It's ironic: the entire industry has designed spas to have zero conduction between the water in the tub and the outside world. It turns out that's THE problem.

A trivial #12 bare wire from water to the Service ground solves everything. Poof, the problem is gone.

You can stop chasing your tail hunting for the phantom voltage source.

The problem you are talking about will affect hot tubs and spas where the water is not usually changed very often but a jetted bathtub, usually gets fresh water every time. That is probably why they are exempt from the water bonding requirements.

Nice theory, Tesla, except for one thing ....

The proposals that led to the equipotential bonding plane requirements were based on some issues that arose in Las Vegas- in FRESH water pools. No such (voltage gradient) problems were found in the also popular salt water pools.

I suspect those Vegas pools were concrete so bonding the water was not an issue. The water was a conductor and bonded everything in contact with it, where fresh water might not. Tesla is talking about a plastic tub, where the water is at a different potential than anything around it. If everything remains non-conductive for 5 feet or so around the tub, there is no real problem. If you have a grounded tile floor, you are going to feel that shot, even if it is just static electricity.
I bet, if you did nothing but punch a stainless bolt through that tub wall, in contact with the water and bond it, you would be fine.
680.26(C) says you should have 9 sq/in. tho

No, Greg, quite the opposite.

Vegas started having problems with pools that were made entirely of plastic, or plastic-coated metal. Pools where the "tub" was trucked intact to the site, then set in the ground.

Add to these pools entirely plastic plumbing and a plastic pump, and there was no way - or seeming need- to bond the pool.

As mentioned before, it seems the very conductivity of salt water prevented voltage gradients from forming, while relatively non-conductive fresh water pools had issues.

It is academic in pools and spas now anyway because we bond the water.

"Academic" it may be, but there was some theory put forth here that was inconsistent with experience.

Does it matter? Yes, it does - as evidenced by that classic picture of someone 'grounding' a generator on a catwalk by sticking a rod into a bucket of dirt!

Reno...

The physics describe above also under pins MHD power generators.

http://en.wikipedia.org/wiki/Magnetohydrodynamic_generator

In a MHD design, the 'salt' used is typically potassium based.

The actual power density/ energy density is quite low. It's voltage potential is easily measured with a Fluke.

This Faraday effect does not occur in totally pure water. Btw, some areas of the country don't even have 'pure' water at the tap, potable, yes. So even the garden hose introduces more salts into the hot tub.

None of these units would be deemed to be 'salt water' tubs, per se. The salt concentrations are still so low as to merely taste bad.

The users get jolted because water is splashed out. Then when they exit -- obviously dripping -- they end up having one leg at zero potential/ the ground plane and their other still sticking straight into the hot tub.

If you're still unconvinced -- channel Dr. Faraday, Dr. Hall and the rest.

I've seen it, tested it, and cured it. BTW the hot tub in question was in Reno.

&&&

If a device has a very well shielded motor then the above effect figures to be totally suppressed.

I've lost track of how many tub owners have gone onto the Internet to complain about phantom voltages -- near their hot tubs.

They're on the Internet -- because the service electricians -- more than one -- can't cure their problem -- and don't believe them -- or don't know what could POSSIBLY be causing the problem.

The typical owner assumes that the equipment is shorting -- and that's that. Of course, a true short would trip the GFCI -- if not the circuit breaker.

BTW, the impressed voltage also does a cracker jack job corroding any metal it contacts. No-one connects this galvanic, stepped up, corrosion with the Faraday effect.

At all times, these are low energies... kind of like building up a static charge by shuffling across carpeting.

The ability of an AC current to build up and impress a DC bias in transformer steels is covered by Buss, and aluded to in Ugly's. (short circuit calculation section)

The AC current flip flops the induced field. But because of imperfections in the magnetic domains of the rolled steel, it actually develops a slight DC bias. It favors one orientation more than the other. This effect is also at a (very) low level. It's never brought up in ordinary calculations or class discusions. You could be in the trade for fifty-years and never have to deal with it. It's counter intuitive, too. We're ALL taught that AC is symmetrical. But nature is complicated -- and actually is always slightly away from idealized physics.

We can't rely on anything as science, can we? If we did, we'd be forced to repeal all those AFCI requirements.

After all, doesn't Paschen's Law bluntly assert that there can be no arcing faults at household voltages on copper wires:

http://en.wikipedia.org/wiki/Paschen%27s_law