hot tub trouble

Status
Not open for further replies.

WESL

Member
Location
Wewahitchka, fl
Got a call for a outdoor hot tub shocking some people yesterday. When I arrived at almost dark I got a voltage reading of around 20-30 volts from the water in hot tub to any ground source. I disconnected and returned today, reconnected and I cannot duplicate the reading. Matter of fact all seems fine. Tub fed with 2 pole 60 gfi breaker. There is 3 #6 conductors and a #8 ground to hot tub, however this hot tub did not have a neutral connection and diagram showed only a 3 wire connection. Thus the neutral was just capped in control box. I know a 2 pole GFI does not have to have a neutral to work, just trying to give details about installation. The breaker was installed correctly and test button trips it. I did find that the ground had a somewhat loose connection. I completely disconnected ground to see if I could duplicate and could not. Also 1 pole of the 2 pole GFI breaker hot wire was possibly loose, my helper checked that and said he got a good 2-3 turns on that. Any advice on how to proceed with trouble shooting. With out being able to duplicate the readings I got yesterday I am not sure what could have been causing this. Another note is this spa is on a dock over salt water. The conduit feeding and the spa is high above water line. We have troubleshot in the past transient voltage issues at close docks to this one and found corroded grounded conductors on both PC side and client side, but these issues were always with boat lifts actually in the salt water. Also have checked all ground connections and appear good. I have scheduled PC to put beast on meter next week as this always seems to highlight any problems! Thanks for any advice!
 

kwired

Electron manager
Location
NE Nebraska
What do you have for equipotential bonding between the equipment grounding conductor and anything grounded near this tub?

Any rise of voltage on the service grounded conductor which is natural because of voltage drop when there is load on the neutral can cause this kind of problem because our equipment grounding conductor is bonded to the service grounded conductor back at the service equipment and this is why we need to bond things in the vicinity as well as provide equipotential plane - especially if on grade or a slab on grade - to keep everything within a users reach at the same potential. A GFCI will not trip because of these "stray" voltages as they are not seen by the GFCI.

When you came back and problem seemed to be gone - the load on the grounded conductor (by other items connected to it) likely changed and was not creating the problem at the time.
 

GoldDigger

Moderator
Staff member
Location
Placerville, CA, USA
Occupation
Retired PV System Designer
Got a call for a outdoor hot tub shocking some people yesterday. When I arrived at almost dark I got a voltage reading of around 20-30 volts from the water in hot tub to any ground source. I disconnected and returned today, reconnected and I cannot duplicate the reading. Matter of fact all seems fine. Tub fed with 2 pole 60 gfi breaker. There is 3 #6 conductors and a #8 ground to hot tub, however this hot tub did not have a neutral connection and diagram showed only a 3 wire connection. Thus the neutral was just capped in control box. I know a 2 pole GFI does not have to have a neutral to work, just trying to give details about installation. The breaker was installed correctly and test button trips it. I did find that the ground had a somewhat loose connection. I completely disconnected ground to see if I could duplicate and could not. Also 1 pole of the 2 pole GFI breaker hot wire was possibly loose, my helper checked that and said he got a good 2-3 turns on that. Any advice on how to proceed with trouble shooting. With out being able to duplicate the readings I got yesterday I am not sure what could have been causing this. Another note is this spa is on a dock over salt water. The conduit feeding and the spa is high above water line. We have troubleshot in the past transient voltage issues at close docks to this one and found corroded grounded conductors on both PC side and client side, but these issues were always with boat lifts actually in the salt water. Also have checked all ground connections and appear good. I have scheduled PC to put beast on meter next week as this always seems to highlight any problems! Thanks for any advice!

I don't have any revelation for you, but some troubleshooting steps that may help. Not necessarily in the order I have listed them.

When you checked "to any ground source" was that just to things that you knew were part of an equipotential or ground system or did you include a screwdriver in the earth?
One possibility is that what was energized somehow was the ground electrode system. A fault current of only one amp from anywhere in the system (not on a GFI) to a 25 ohm ground resistance would give you 25 volts on the whole grounding system.

How is the water in the tub itself grounded? Is the pump chamber itself (not just the motor) or some part of the circulating piping metal and grounded? If not, is there a ground plate in the tub?
A second possibility is that the water is what was energized. It would not take a lot of current leakage from something not GFCI protected to cause a pretty high voltage in the water if there is no effective grounding of the water.
You may not be able to duplicate what energized the water, but you could improve or completely fix the situation by making sure the water is grounded.

Finally did the people actually in the tub complain of a shock or just people getting in and out or touching someone outside the tub?
If people in the tub noticed it, there would have to be a voltage gradient in the water. Current flowing in at one point and going to ground at another.
If it was only people getting in and out, what is the surface around the tub and is it capable of being grounded effectively to the main grounding system like the pump, etc.?

Figuring out where the potential difference was located may help you track down the source. It may be that it was just a minor current leakage somewhere and proper bonding is all that it will take to correct things. The loose ground is suspicious, but it may only have been the factor that allowed a voltage to develop, and by itself disconnecting it would not replicate the situation.

Where was water splashed or dripping around the tub when you arrived? Could any of the water have gotten to an unsealed electrical source somewhere in the area? Under the tub deck? Try hosing the whole area down and waiting 30 minutes or more when trying to replicate.
And finally, how long had they been using the tub before they first noticed the shocks? Was there any electrical equipment like dock lighting that was on at night but not when you tried to replicate?

Good luck!
 

WESL

Member
Location
Wewahitchka, fl
The hot tub is located on a wood deck with no metal around except a light about 6' away mounted on a wood post 8' away. The voltage reading I was getting was between this metal on this light and the water in the hot tub. Now caint duplicate. Also the light is still wired but hasn't worked in years and off at a switch. I went back several times thru the day hoping to catch a water heater being on or some load different to see if I could duplicate. No luck. Also there is no bond between that light and the spa. All the motors seem to be grounded with grounding electrode and water is grounded, but only by equipment grounding conductor. I think that's ok!
 

GoldDigger

Moderator
Staff member
Location
Placerville, CA, USA
Occupation
Retired PV System Designer
The hot tub is located on a wood deck with no metal around except a light about 6' away mounted on a wood post 8' away. The voltage reading I was getting was between this metal on this light and the water in the hot tub. Now caint duplicate. Also the light is still wired but hasn't worked in years and off at a switch. I went back several times thru the day hoping to catch a water heater being on or some load different to see if I could duplicate. No luck. Also there is no bond between that light and the spa. All the motors seem to be grounded with grounding electrode and water is grounded, but only by equipment grounding conductor. I think that's ok!

Motors and water must be connected to EGC and the ground rods at tub should be bonded to rest of house/dock/whatever grounding system. Nothing should rely only on the isolated ground rods for protection.
 

WESL

Member
Location
Wewahitchka, fl
I said the water was grounded. May have made a mistake there. The pumps were grounded so I assumed the water was grounded. Maybe a dumb question but how do you test if water is grounded? I assume you would test resistance to grounding conductor? But we all know what assume stands for! Also with no metal or ground around it is not necessary for any bonding grid on a wood dock is there? I know a slab would need it but with a package spa on a wood dock? By the way thanks for the help!
 

GoldDigger

Moderator
Staff member
Location
Placerville, CA, USA
Occupation
Retired PV System Designer
I said the water was grounded. May have made a mistake there. The pumps were grounded so I assumed the water was grounded. Maybe a dumb question but how do you test if water is grounded? I assume you would test resistance to grounding conductor? But we all know what assume stands for! Also with no metal or ground around it is not necessary for any bonding grid on a wood dock is there? I know a slab would need it but with a package spa on a wood dock? By the way thanks for the help!

If the pump housing is metal and is not lined with plastic, then grounding the pump will ground the water. But if the pump housing is plastic, grounding the pump motor will not ground the water.
You need to put a several foot (or more) length of metal pipe in the circulation and ground it or else put a metal plate in the water and ground it. To the EGC, not just to a ground rod!

There are criteria about when an equipotential grid is or is not needed for a package spa. They have to do mostly on the distance from the spa to metal or a conductive surface. If you are required to do a grid, there are those who argue that you have to ground the wood deck too, but it is not clear how to do it.
If the dock is over salt water and the wood gets wet, even from above, it might actually end up as a pretty good conductor while wet.
 

WESL

Member
Location
Wewahitchka, fl
Getting late I said the motors were grounded by GEC. I meant Equipment ground. The spa is about 100' away from house power panel and there is no ground rods near the hot tub at end of dock. 2 pole gfi in house power panel. The only ground at spa is the equipment ground. I have been going thru 680 and I thought this was ok. What am I missing?
 

WESL

Member
Location
Wewahitchka, fl
Sorry to go on and on but been a while since I've been hitting the books like this. I have got in the grind of scheduling work ,invoicing, etc I forgot how fun good old fashion trouble shooting was! I was thinking the loose Equipment Ground was probably the problem.y It was barely making connection. At Bare minimum I know it needed repair. But I am correct in thinking if the EG was completely broke. Which I duplicated when I unhooked it. If the problem was a fault on a non gfi device in the house or other location. The voltage or fault current would not have made it to the hot tub. Thus I would conclude that the problem was not the poor connection of the EG. Or is that crazy?
 

kwired

Electron manager
Location
NE Nebraska
The hot tub is located on a wood deck with no metal around except a light about 6' away mounted on a wood post 8' away. The voltage reading I was getting was between this metal on this light and the water in the hot tub. Now caint duplicate. Also the light is still wired but hasn't worked in years and off at a switch. I went back several times thru the day hoping to catch a water heater being on or some load different to see if I could duplicate. No luck. Also there is no bond between that light and the spa. All the motors seem to be grounded with grounding electrode and water is grounded, but only by equipment grounding conductor. I think that's ok!

You can and often will get voltage between the EGC and anything at "earth" potential. As I said earlier this happens even when the installation is correctly done as there is voltage drop on grounded conductors when they are carrying current and they will operate at a voltage above "earth ground". Your EGC is also bonded to this grounded conductor back at the service equipment and is going to be at the same potential above "earth". That is why anything within close proximity to things like swimming pools, hot tubs and such need to be bonded together. This way no matter what level they are above "earth" everything within reach of users of this equipment is all the same potential. Everything can theoretically be 1000 volts above earth, but if everything within reach of the user is same potential, they will never feel a thing.

Water in tub could easily be grounded by a metal shell in the heating element section of the plumbing.

If it is a listed packaged tub, you shouldn't need to worry too much about bonding the water - it should already be done somewhere as part of the listing I would hope - you maybe do need to check into it though just to see if it is still intact in some cases like this.
 

Little Bill

Moderator
Staff member
Location
Tennessee NEC:2017
Occupation
Semi-Retired Electrician
When you all keep saying "ground" the water, i think you mean "energizing" the water.

At least that is what is described when doing that for a pool. Or more preciseley, "bonding" the water.

Of course, you are only energizing it if there is leakage somewhere that might cause a difference in potential. Then you're just trying to keep the water on the same potential as anything else close to it.
 

WESL

Member
Location
Wewahitchka, fl
New info from owner. He tells me the conduit feeding the spa got dug into by tractor a while back. He remembers close to where it happened and planning to dig up today. But if it were a bad splice, say the hot was leaking transient into the grounding conductor at splice, wouldn't the GFI breaker see that fault on the branch circuit side. Any situation I can think of on the branch circuit side should be seen by the GFI breaker. Am I correct?
 

kwired

Electron manager
Location
NE Nebraska
When you all keep saying "ground" the water, i think you mean "energizing" the water.

At least that is what is described when doing that for a pool. Or more preciseley, "bonding" the water.

Of course, you are only energizing it if there is leakage somewhere that might cause a difference in potential. Then you're just trying to keep the water on the same potential as anything else close to it.
Absolutely correct.

New info from owner. He tells me the conduit feeding the spa got dug into by tractor a while back. He remembers close to where it happened and planning to dig up today. But if it were a bad splice, say the hot was leaking transient into the grounding conductor at splice, wouldn't the GFI breaker see that fault on the branch circuit side. Any situation I can think of on the branch circuit side should be seen by the GFI breaker. Am I correct?
I would make sure the EGC wasn't compromised (with a meter) and not waste any time looking for a needle in a haystack. EGC is only conductor that could be damaged and not cause GFCI tripping problems and/or equipment operation problems.
 

Jraef

Moderator, OTD
Staff member
Location
San Francisco Bay Area, CA, USA
Occupation
Electrical Engineer
I'd put money on the fact that the ground is/was disconnected, but when you tested it the heater was not on. The heater element is probably leaking, but if the water was already hot when you tested it by removing the ground connection, there would be no current to the heater. But because the ground reference was lost, when the heater did come on, the GFI didn't see it.
 
Last edited:

WESL

Member
Location
Wewahitchka, fl
I have cranked the temp up on the spa and measured the voltage while temp of water was rising and cannot duplicate the voltage i had the first day from water to metal on the nearby light. Also wouldnt the GFI breaker trip if it were the element leaking? Also not 100% on this but my understanding is a 2 pole gfi will still work with no EGC. Also a S/P will. That is why if in an old house with no grounds they make you GFI protect all outlets. A 2 pole GFI monitors between the 2 hot legs and each hot to neutral correct? As for this tub it is only a 3 wire so the neutral is not hooked up in spa, but you get your protection from phase to phase monitoring. Let me know if im wrong anywhere!
 

wdemos

Member
Location
Commerce, Mich.
Water in tub could easily be grounded by a metal shell in the heating element section of the plumbing.

If it is a listed packaged tub, you shouldn't need to worry too much about bonding the water - it should already be done somewhere as part of the listing I would hope - you maybe do need to check into it though just to see if it is still intact in some cases like this.[/QUOTE]

Check to be sure the bonding conductor is still connected to the heater shell it may have been removed during a element change and did not get reconnected. - sorrry Jraef I am slow at typing.
 
Last edited:

GoldDigger

Moderator
Staff member
Location
Placerville, CA, USA
Occupation
Retired PV System Designer
I have cranked the temp up on the spa and measured the voltage while temp of water was rising and cannot duplicate the voltage i had the first day from water to metal on the nearby light. Also wouldnt the GFI breaker trip if it were the element leaking? Also not 100% on this but my understanding is a 2 pole gfi will still work with no EGC. Also a S/P will. That is why if in an old house with no grounds they make you GFI protect all outlets. A 2 pole GFI monitors between the 2 hot legs and each hot to neutral correct? As for this tub it is only a 3 wire so the neutral is not hooked up in spa, but you get your protection from phase to phase monitoring. Let me know if im wrong anywhere!

The 2-pole GFCI will detect leakage from the heater element (or wiring) to ground. If the leakage is small enough, and the water and element shell are not effectively grounded there could still be a significant voltage on the water since there is no ground path. But more important, if the leakage to the water is coming from somewhere other than the GFCI protected circuit, the breaker will not protect against that, since no mismatched current is flowing through that circuit.
Are the heater and the pump both on the same GFCI-protected circuit?
 

WESL

Member
Location
Wewahitchka, fl
they are on the same circuit. It is a package spa fed with a 1 2 pole 60 gfi. I have returned and checked all I know to check. I have turned on almost everything in the residence to create as much load as possible looking for what may could cause a fault current, I have made sure the pumps and heaters are bonded to ground. I have made sure the heaters and pumps are running while checking for a voltage reading I got the other day. I have verified the EGC is continuous from panel to spa. I have not been able to duplicate what caused the 30 volts I read. My theory is the bad connection of the EGC on the first visit was either causing the voltage or was inhibiting the GFI from functioning correctly and now that EGC is cleaned and tight if the fault occurs again should allow GFI to function.
 

Smart $

Esteemed Member
Location
Ohio
Any other docks with electrical equipment nearby. Could easily have been caused by another service having a problem... not the one you are looking at.
 
Status
Not open for further replies.
Top