It certainly wouldn’t hurt. IMO if there was a real danger then the NEC would already require the bonding of metallic water pipes in the vicinity of the tub.
Yes it needs bonding, in the event of a burned element in the water heater that would allow current to flow in the water while the tub is being filled or if someone does like the wife and add water after sitting in the tub for a while. This bonding will bring everything including the fill water to the same level.
I will fight hard through the comment stage to squash the proposal to leave the fill valve out to the requirement to be bonded.
There isn’t a requirement to actually bond the metallic water pipes that only fill the tub. If you feel that this is necessary then you should write a proposal to include such wording. The current wording is confusing so the change will only clarify what’s already part of the requirement and that is that only metallic piping that contains the circulating water must be bonded.
I have been out of state working this weekend and have not kept up with this thread.
I’m just jumping in, and need to back through this stuff.
Joseph, so am I to understand that a burnt element in the water heater is going to energize the water in the tub 80 feet away and not go through the water heater bonding?
Around here, water is not that good of a conductor.
In my example, I have a pump motor bonded to the fill valve.
CPVC piping to the valve.
If the pump fails to the case of the pump motor, I see that the bonding wire will become energized as well and because it is connected to the metal valve set, will energize it. If your wife is sitting in the tub, filling water and the pump fails, will she not be grabbing full impedance from the failed circuit as it has no place else to go?
Will she not be a better conductor to/through the water than the bonding on the CPVC pipe?
All this code stuff that I have talks about copper pipe, which this is not.
To expect water to be the conductor, what makes you, sitting in the water a better path to ground? Like a bird sitting on a wire. Till you connect the circuit, aren’t you safe?
It looks to me that the fill valve is just that path.
If the water heater energizes the water (and you) and you touch the bonded valve, which is bonded to the grounded pump motor, will you not complete the circuit?
Which ever direction the electricity flows (from the pump or from the water heater) this bonded valve completes the unwanted circuit (in my opinion).
I really don’t care what the code says…
But I am interested in the electrical theory. If there is a code that conflicts with this theory, I’ll call it it like everything else I do.
How or why would someone try to bond to a plastic water line? Would you use weed eater line?
David, I believe JW was talking about the heater in the tub, not the storage tank.
The thread has morphed a bit from the original post, so we’ve kinda moved on to a case where the water pipes of the residence are metallic - i.e. copper or galv.
I have always been under the impression that the distribution piping in such a case would require bonding to the pump-motor. Apparently, I may have been incorrect. So I’m very interested in the disposition of the proposal, but I am also interested in hearing opinions on both sides.
Once again gentlemen the bonding done at the tub is not to remove current but to bring everything that a person might touch to the same potential.
If any and all metal and water are bonded together and has 1 million volts on it there would be no danger to anyone touching it. Just as the bird sitting on the bare power line and then flies away without harm.
If everything is bonded together there is no path to complete by the human interaction as the bonding completes the path. The bonding is not done to remove anything nor is it done to clear a fault by opening the overcurrent device. It is done solely to make everything to the same potential.
Two parts hydrogen and one part oxygen (water) which neither are a good conductor of electricity but put in foreign material such as purifiers (utility water) or minerals (well water) and then we have a conductor.
This debate of what is to be bonded has been going on for several years with some looking at nothing more than the profit of the job while others look at the safety of the user. I promise that your opinion on just what is to be bonded will revolve around one or the other of these issues. Personally I look at user safety not the profit of the installation.
When it is my family that is involved I go far beyond the minimum safety standard of the NEC. A good example is bullet resistant glass. I can purchase glass that will stop one bullet fired from a 5.56 by 45 mm but will not slow down a 300. If it is my family being protected then I am going with the stronger glass.
The underside of the whirlpool tub in my master bath looks like a spider web with all the #8 bare copper used to bond everything within sight of the tub. I love my wife. Anything metal within reach of the tub is bonded including the supply line to the water closet which could be touched while standing in our tub.
When doing an inspection of a home the main concern that should be in the inspector’s mind is the safety of their client not code rules. This is evidenced throughout this site by the repeated pointing out the lack of GFCI protection. No one here has a problem of pointing out the lack of them no matter when the home was built. The same rule should be used in all aspects of home inspections should it not?
Well explained Thanks much appreciated .
No heater in this tub.
Thanks Mike, I’m starting to see the light through the clouds (to some extent)!
My problem is that I am not seeing connected circuits in this case. My bad.
All this plastic stuff is clouding the schematic! 
My first impression is that the pump motor is isolated from the water it start with. When you connect the motor case to the tub valve, the valve becomes fully energised. If you connect the path to earth better than the water can, your toast.
The harder I look at it, the more it looks like it’s OK.
Clear as mud!
Don’t think of current going to earth as this is not true, it must always return to its source. Yes the earth can be part of this path but not very much current will flow through earth at a level of 120 volts and yes enough to hurt or kill.
Also remember that the motor has a steel shaft that the pump connects to. It is the pump motot that is most likely to energize the tub water so if everything is bonded to the pump motor then it is all at the same potential
I call out pumps that are not bonded to metallic water pipe (when metal pipe is available).
I think that because the pump has a bonding lug, it should be used. Metal pipe should carry the impedance back to earth some way better than we can expect water to.
If you have plastic pipe for water supply, I think that the pump ground should relied upon (and a grounded outlet, or branch circuit should be available). But to bond a metal water valve that goes nowhere but through the water seems off.
A wet finger is a better conductor than a dry finger, but this does not mean water is a better conductor than copper…
But then I start thinking that the tub valve is really not bonded to the water as there is no metal to metal connection through all valves…
Don’t go gettin me confused now. I’m starting to follow where you are going with this! 
The confusing part is making someone bond the faucet water piping because of the way the first sentence in 680.74 is written. Although it won’t hurt it’s not required. 
I’ve been arguing this point for years with CEU instructors and inspectors. Hopefully this will be resolved in the 2014 NEC.
that would be nice!