Hot tub

[ptonsparky]

I applaud the ingenuity, but a shunt trip breaker would be a better for that type of stop. As others have said, you still need a local disconnect.

Am I missing something? What exactly does the bond have to do with no ground at the tub? The equipment ground in the 50amp circuit should be connected to the neutral bar in the main panel. The panel bonding has nothing to do with the equipment ground at the tub. If you were to remove the green bonding screw on a working main panel, it would not change any of the grounds of the circuits leaving the panel. The metal box (panel) would not be bonded, and floating. The circuit's equipment grounds would still function.

They will only function if that bond is made. OP is missing that bond somewhere.

 

[Little Bill]

Am I missing something? What exactly does the bond have to do with no ground at the tub? The equipment ground in the 50amp circuit should be connected to the neutral bar in the main panel. The panel bonding has nothing to do with the equipment ground at the tub. If you were to remove the green bonding screw on a working main panel, it would not change any of the grounds of the circuits leaving the panel. The metal box (panel) would not be bonded, and floating. The circuit's equipment grounds would still function.

If the EGCs and neutrals were not bonded in the main panel, how would a fault to ground trip a breaker?

 

[tthh]

The "disconnect" is not really a disconnect, it is a more of a "push-to-test" using a switch that is supposed to trip the GFCI. There needs to be a disconnect that doesn't depend on something tripping.

 

[Dennis Alwon]

There are 2 things going here. There is the emergency off and the tub disconnect

680.41 Emergency Switch for Spas and Hot Tubs. A clearly
labeled emergency shutoff or control switch for the purpose of
stopping the motor(s) that provides power to the recirculation
system and jet system shall be installed at a point readily accessible
to the users and not less than 1.5 m (5 ft) away, adjacent
to, and within sight of the spa or hot tub. This requirement
shall not apply to one-family dwellings.

 

[retirede]

If you were to remove the green bonding screw on a working main panel, it would not change any of the grounds of the circuits leaving the panel. The metal box (panel) would not be bonded, and floating. The circuit's equipment grounds would still function.

That is simply untrue. In that scenario, a ground fault would not trip the OCPD on non-GFCI circuits.

That’s also what is causing the N-G shunt switch to not trip the GFCI in the OP’s case.

 

[tortuga]

I found in the main panel there is a ground coming from the meter through a grounded bushing straight to the cold water ground,

So it sounds like a Supply side / main bonding jumper is in the meter can itself.

no bonding in main panel

This would be consistent with the main bonding jumper being in the meter.

nor not good continuity between main and sub panel !

Is there a wire type EGC between the main and the sub?

I would love to see a one-line of what you have there.

I think we have something like this
Utility ---> Meter can with N/G bond ----> service panel ----> subpanel -----> 240V branch circuit
Since its Chicago the EGC's might be EMT and not the wire type and perhap there is a loose fitting causing the issue.

 

[wwhitney]

That is simply untrue. In that scenario, a ground fault would not trip the OCPD on non-GFCI circuits.

Depends on whether the main panel has grounds and neutrals on separate bars, with the neutral bar isolated from the case, and the MBJ green screw is part of the fault path for every exiting EGC; or whether it has a common terminal bar on which all grounds and neutrals are landed, and the green screw is just bonding the case.

Cheers, Wayne

 

[Little Bill]

So it sounds like a Supply side / main bonding jumper is in the meter can itself.

This would be consistent with the main bonding jumper being in the meter.

Is there a wire type EGC between the main and the sub?


I think we have something like this
Utility ---> Meter can with N/G bond ----> service panel ----> subpanel -----> 240V branch circuit
Since its Chicago the EGC's might be EMT and not the wire type and perhap there is a loose fitting causing the issue.

You can't have a MBJ in a meter since there is no EGC in the meter.

 

[tortuga]

You can't have a MBJ in a meter since there is no EGC in the meter.

I am talking about something like this, where the GEC is brought to the meter can

 

[retirede]

Depends on whether the main panel has grounds and neutrals on separate bars, with the neutral bar isolated from the case, and the MBJ green screw is part of the fault path for every exiting EGC; or whether it has a common terminal bar on which all grounds and neutrals are landed, and the green screw is just bonding the case.

Cheers, Wayne

That’s true - I was presuming the bonding screw was bonding the isolated neutral bar to the can.

 

[Little Bill]

I am talking about something like this, where the GEC is brought to the meter can

That still is not in the meterbase. We are allowed by all but one POCO to land the GEC in the meterbase.
But the MBJ is in the service disconnect (w/OCPD)

 

[Murr72]

So it sounds like a Supply side / main bonding jumper is in the meter can itself.

This would be consistent with the main bonding jumper being in the meter.

Is there a wire type EGC between the main and the sub?


I think we have something like this
Utility ---> Meter can with N/G bond ----> service panel ----> subpanel -----> 240V branch circuit
Since its Chicago the EGC's might be EMT and not the wire type and perhap there is a loose fitting causing the issue.

There is n bonding except for the grounded bushings between meter socket an main panel, so basically the conduit is bonded because they used concentric knock outs, straight to cold water pipe!

 

[Murr72]

No bonding no green screw just bonding of the conduit in between meter and panel

 

[Murr72]

This home is all conduit, I haven’t really researched the home, wether the method of the switch neutral to ground is right or wrong Is not the problem anymore ! To correctly bond the panels to avoid electrical issues in the future and to secure that the GFCI on the hot tub is working properly is the main concern at this point! I have already been convinced from so many posts that it’s not within code to do that although I have passed inspections up in my area doing this! Neutral to ground Tripping the GFCI! So y’all may know the code but whatever I don’t care we’re all learning and I’m still myself after 30 something years in the trade I’m sure I been taught wrong in certain situations but I’m here to learn! I work alone for a company in the service industry and I don’t know everything and I need help from time to time! Thanks for all the feedback

 

[GoldDigger]

Just for information, rather than arguing one way or another about this use of the GFCI:

The normal primary operation of a 240V two pole GFCI uses a single toroid to detect current on the two hot conductors and the neutral and trip if the phasor sum of those currents is more than 6ma. A bond between the EGC and the neutral will NOT activate this primary trip function since no current will normally be flowing in the neutral in the first place is all loads are line to line.
What this technique is exploiting is an entirely separate trip function which uses a second toroid to inject a voltage between neutral and EGC downstream of the GFCI (breaker or receptacle type). Current in this second toroid will trip the GFCI if the resistance between N and EGC on the load side of the breaker is below a certain value. I am not sure what that value is, but apparently the faulty bond between N and EGC at the main panel is preventing the trip, or the GFCI in question does not contain this secondary trip mechanism.
On a 120V circuit with a high enough load connected, an N to EGC switched connection will trip the primary sensor in the GFCI.

 

[kwired]

That is simply untrue. In that scenario, a ground fault would not trip the OCPD on non-GFCI circuits.

That’s also what is causing the N-G shunt switch to not trip the GFCI in the OP’s case.

If your EGC's also land on the neutral bus in the service panel what he said is true. The enclosure would be all that is floating.

If you have the EGC's all landed on a ground bar attached to the enclosure then the enclosure and all the EGC's would be floating.

 

[retirede]

If your EGC's also land on the neutral bus in the service panel what he said is true. The enclosure would be all that is floating.

If you have the EGC's all landed on a ground bar attached to the enclosure then the enclosure and all the EGC's would be floating.

Right…..but with the info we’ve been given, I’m fairly confident the EGC (at least the one to the hot tub) cannot be landed on the neutral bar (or otherwise bonded to neutral). Whether that’s true for only the hot tub circuit, or all GECs, we don’t know - that’s what we’ve encouraged him to investigate. But he only wants to “fix the hot tub wiring.”

 

[kwired]

I am talking about something like this, where the GEC is brought to the meter can

Which is what I think is probably going on in OP's install, they just routed the GEC through the nipple to the service disconnect enclosure without bonding it to anything along the way.

Code compliant or not just isn't that great of an idea - if you are going to the service disconnect with the GEC anyway, just land it there.

I think the main issue here is likely that the EGC's land on separate bus from neutral bus but there is no main bonding jumper in the neutral bus.

That is the issue not allowing the GFCI breaker to trip that triggered all of this conversation, however I don't see that method as an acceptable means of disconnect at the SPA either. I could see it as acceptable method for required emergency stop in 680.41. Dennis brought this up in post 41. That E- stop isn't required to disconnect all ungrounded supply conductors like a disconnecting means is, it just needs to shut down the circulation system - AFAIK was written at a time when it was more common for someone to become stuck to an inlet on the piping system from the suction. Now they use inlets designed so that can't happen very easily and it may not be as important to have such an E stop as it once was.

This is/was more commonly needed for non packaged spas where the circulation controls are remote from the tub area. A packaged spa still needs a disconnect near the unit - it is more of a requirement for those servicing the unit than it is required to be an emergency stop. The packaged units already have controls on them to stop circulation.

 

[Eddie702]

Seems like shorting two wires together when you can't find the breaker

 

[tortuga]

Which is what I think is probably going on in OP's install, they just routed the GEC through the nipple to the service disconnect enclosure without bonding it to anything along the way.

Code compliant or not just isn't that great of an idea - if you are going to the service disconnect with the GEC anyway, just land it there.

The whole concept of having the MBJ start in the meter then hit the bond bushing in the service can got me looking in the code for where the MBJ has to be 100% in the service disconnect. And its 250.24:

(B) Main Bonding Jumper. For a grounded system, an
unspliced main bonding jumper shall be used to connect the
equipment grounding conductor(s) and the service-disconnect
enclosure to the grounded conductor within the enclosure for
each service disconnect in accordance with 250.28.

Also keep in mind this is Chicago and everything would be EMT so probably no wire type EGC's.