Brand New install. Fiberglass pool with water bond. LED nicheless light, salt generator, gas heater. Only feel the tingling when salt generator communication wire is plugged into automation. The water coming out of the jets have a charge. Any ideas. Perimeter bonding in place, equipment is bonded also
Pool water energized
- Thread starter 1haz
- Start date
- Status
synchro
Senior Member
- Location
- Chicago, IL
- Occupation
- EE
Is the salt generator powered through a GFCI breaker?
- Location
- Bremerton, Washington
- Occupation
- Master Electrician
What type of comm does the salt generator use?
kwired
Electron manager
- Location
- NE Nebraska
- Occupation
- EC
My question (not familiar with this sort of equipment) is what is it? Seems possible it is some sort of conductivity probe that tests the water to determine if more salt is needed. Best way to do this I would think is two probes in close proximity, with neither one being "grounded". But if a single probe detecting continuity to ground I can see more potential for problems like mentioned.
- Location
- Tennessee NEC:2017
- Occupation
- Semi-Retired Electrician
A salt generator just converts salt to chlorine. Salt is poured into on end of a pool and the return line pulls it back with the water. There is a control cabinet, most can be either 120V or 240V. I don't know all the settings but it controls the time, mixture, and other things while the water runs through the tube/cartridge, which is in-line with the plumbing. The tube/cartridge that is in-line with the plumbing has a communication cable that runs to the control panel.
There should be a lug on the control cabinet to attach the bond wire to. It sounds like there are two things happening.
There is something wrong with the system and the control cabinet is not bonded.
As far as GFCI, I usually either run them off a timer that is fed from a GFCI breaker or run them straight off a GFCI breaker.
With that said, @1haz
How are you feeling the tingling? Are you in the water and touching the deck or ground? Are you just reaching in to the water from the deck or ground?
Any metal ladders, steps, diving boards?
Along with the possibility of the salt control cabinet not being bonded, sounds like there could also be something else missing in the bonding grid.
There should be a lug on the control cabinet to attach the bond wire to. It sounds like there are two things happening.
There is something wrong with the system and the control cabinet is not bonded.
As far as GFCI, I usually either run them off a timer that is fed from a GFCI breaker or run them straight off a GFCI breaker.
With that said, @1haz
How are you feeling the tingling? Are you in the water and touching the deck or ground? Are you just reaching in to the water from the deck or ground?
Any metal ladders, steps, diving boards?
Along with the possibility of the salt control cabinet not being bonded, sounds like there could also be something else missing in the bonding grid.
Last edited:
- Location
- Placerville, CA, USA
- Occupation
- Retired PV System Designer
My first guess is that the electrodes of the salt generator (actually salt cell/chlorine generator, not salt generator) are isolated from the equipotential grid but is referenced to the electronics including the communications port. And the equipment at the other end is at a different potential.
It could be elevated from ground OR they are at remote earth potential while the pool grid is not.
It could be elevated from ground OR they are at remote earth potential while the pool grid is not.
A chlorine generator works by applying low voltage (up to perhaps 35V) DC to a set of plates that the water flows through. This supply should be totally isolated from the power system (in the same vein as a pool-rated lighting transformer).
It sounds to me like there is something amiss in the controller that has caused the DC system to no longer be isolated from the power system.
I’d like to see what “Mystic Pools” has to say about it.
It sounds to me like there is something amiss in the controller that has caused the DC system to no longer be isolated from the power system.
I’d like to see what “Mystic Pools” has to say about it.
Correct on what it is. But it’s a DC power supply. Think of a diode rectifier. That gives us around 180 V though...way too much. So if we use an SCR we can turn the diodes on as needed, like a lamp dimmer. That gives us control over the DC.
Water temperature, current/voltage, and air temperature are common signals that the control box uses to adjust the output. It is possible to tightly control chlorine levels with another sensor but they are expensive and need periodic calibration and replacement so only used on some large commercial pools. The system is so stable even most commercial pools don’t bother doing that.
Either way the DC is between the plates. But it is hardly isolated from the AC power line. So the DC negative and AC neutral are either equal or offset by a constant (half of the DC output) depending on if it’s a half or full wave bridge rectifier. My bet is on half wave.
Three possibilities I can think of. The DC negative in the cable is broke and ground/neutral is now the return. SCR shorted (typical failure mode) and inducing AC into water but surprised it didn’t blow fuses. Or bonding to control panel is missing so it is floating probably around 60 V.
Get your meter out and measure ground to water, etc., to narrow down. Check the chlorine generator manual on testing.
kwired
Electron manager
- Location
- NE Nebraska
- Occupation
- EC
Why wouldn't this be designed to use an isolation transformer before the rectifier? Would never be able to supply from a GFCI as there would always be too much leakage current.
- Location
- Tennessee NEC:2017
- Occupation
- Semi-Retired Electrician
I've installed several of these and all were GFCI protected. I can't speak to how they actually work but do know the GFCI held.
- Location
- Bremerton, Washington
- Occupation
- Master Electrician
I was thinking comm as in remote monitoring.
I’m gonna open mine up and see if I can tell if it has an isolation transformer.
kwired
Electron manager
- Location
- NE Nebraska
- Occupation
- EC
If it were direct rectification with voltage regulation via SCR's and then to a probe in the water you will likely leak enough current to ground to trip a GFCI in the supply circuit. If there were an isolation transformer before rectifier then GFCI doesn't see the secondary current.
synchro
Senior Member
- Location
- Chicago, IL
- Occupation
- EE
An isolation transformer could be quite small if it's used in a switching supply, because these typically run at more than 20 KHz. In this case the transformer would have a ferrite core rather than silicon steel laminations.
I couldn’t find a schematic, but I do have a PCB layout. It shows to and from terminals for “cell circuit transformer”. Labeled ‘V’ and ‘Y’.
Sent from my iPhone using Tapatalk
Forgot to follow up on issue. Issue was water bond not making good contact with the water. Not sure why but added a second water bond and it was gone. The source was coming from the salt cell. This is a fiberglass pool which requires a water bond. Not sure why ALL pools don't require this, there is constant voltage flowing across the metal blades producing chlorine so if the pool does not have a light niche (or even if it does) a water bond imo should be added during plumbing.
- Location
- Tennessee NEC:2017
- Occupation
- Semi-Retired Electrician
All pools do require a water bond. What you have with your pool is no metal ladders, handrails, etc. therefore an intentional bond must be installed. If there was any or all the items I mentioned then just the bonding to them would also be the water bond.
Maybe someone who wires more pools than I do can better answer but the only pools I see with a intended water bond are above ground and fiberglass. Most concrete pools built today do not have handrails, ladders etc. I guess the steel in the shell acts as the water bond but how? Is it due to the concrete being porous?
- Location
- Tennessee NEC:2017
- Occupation
- Semi-Retired Electrician
There are numerous ways to bond the water. Actually, anything metal/conductive that is 9 sq in and in contact with the water can serve as the bond. I've done lots of pools and the majority of the inground pools that I've done use something in the plumbing system that is approved for bonding. The rest had handrails and/or ladders, and these have a lug that gets bonded before the concrete poured. Then when they are mounted and contacting the water, the water is then considered bonded.
If you are seeing (any) pools that have no water bond, then .........
They may have been installed before the water bond requirement
You just didn't see the bond even though it was there
They weren't inspected
Inspector either lazy or lack of knowledge concerning this
- Status