480/208V transformer question. Neutral/ground bond

[wwhitney]

The panel response is surprisingly explanatory. But it seems like a low impedance (60 Hz) ground fault path for the SDS will be a low impedance (60 Hz) earth reference, since the primary EGC is earth referenced Is there a need for the SDS to have a low impedance (higher frequencies) earth reference?

I guess the primary feeder EGC could still function as the SDS GEC if it meets the installation requirements for both.

Cheers, Wayne

 

[electrofelon]

The equipment grounding conductor's primary purpose is to provide a low impedance path for
fault current in the event of a ground-fault on the system up to and including the primary of the transformer, not to act as the low
impedance earth reference conductor.
.

Very true. Those electrons performing earth reference duty would not find the EGC acceptable .

 

[don_resqcapt19]

The panel response is surprisingly explanatory. But it seems like a low impedance (60 Hz) ground fault path for the SDS will be a low impedance (60 Hz) earth reference, since the primary EGC is earth referenced Is there a need for the SDS to have a low impedance (higher frequencies) earth reference?

I guess the primary feeder EGC could still function as the SDS GEC if it meets the installation requirements for both.

Cheers, Wayne

I see no technical need for a low impedance connection to earth, and the code doesn't either as there is no specified impedance for that connection in Article 250. The closest thing to one is where you are using a ground rod and the code says if you want to use a single ground rod, the impedance cannot exceed 25 ohms. If it does, you add a second one and you are done. No specific requirement for any low impedance path to the earth.

 

[fastline]

I see no technical need for a low impedance connection to earth, and the code doesn't either as there is no specified impedance for that connection in Article 250. The closest thing to one is where you are using a ground rod and the code says if you want to use a single ground rod, the impedance cannot exceed 25 ohms. If it does, you add a second one and you are done. No specific requirement for any low impedance path to the earth.

I would like to circle back to our specific case if someone can cite specific code locations that 'may' apply so I can make sure and soak up all of it.

Basically, here is what is happening, and I'd like to get suggestions on preferred connections. 480 3 wire+G is provided to a building indoors. Equipment requires 208V 3P (no 1P ). Because machines are big, it is very common to set a transformer for every machine. They commonly use a 480 delta/208-120Y transformer because they are common as dirt.

I will just ask the more general question of how would 'you' make connections? Equipment grounding and personnel protection is of paramount importance. Someone asked if this is an isolation transformer. Yes! By design, primary is isolated from secondary.

In my mind, there MUST be a neutral to case ground in the transformer to create an electron path for any phase to ground. Otherwise the secondary is technically "floating", which is probably likely since I tested odd voltage of legs to ground.

 

[don_resqcapt19]

I would like to circle back to our specific case if someone can cite specific code locations that 'may' apply so I can make sure and soak up all of it.

Basically, here is what is happening, and I'd like to get suggestions on preferred connections. 480 3 wire+G is provided to a building indoors. Equipment requires 208V 3P (no 1P ). Because machines are big, it is very common to set a transformer for every machine. They commonly use a 480 delta/208-120Y transformer because they are common as dirt.

I will just ask the more general question of how would 'you' make connections? Equipment grounding and personnel protection is of paramount importance. Someone asked if this is an isolation transformer. Yes! By design, primary is isolated from secondary.

In my mind, there MUST be a neutral to case ground in the transformer to create an electron path for any phase to ground. Otherwise the secondary is technically "floating", which is probably likely since I tested odd voltage of legs to ground.

The secondary is a 208Y/120 volt system and that is required to be a grounded system per 250.20. You must bond XO to a grounding electrode system and provide a system bonding jumper. The EGC for the secondary circuits originate at the location of the system bonding jumper, but if that is located at the transformer, a supply side bonding jumper would be installed between the transformer and the OCPD. The EGCs would connect to the supply side bonding jumper at the OCPD enclosure.

 

[fastline]

The secondary is a 208Y/120 volt system and that is required to be a grounded system per 250.20. You must bond XO to a grounding electrode system and provide a system bonding jumper. The EGC for the secondary circuits originate at the location of the system bonding jumper, but if that is located at the transformer, a supply side bonding jumper would be installed between the transformer and the OCPD. The EGCs would connect to the supply side bonding jumper at the OCPD enclosure.

I will have to get the transformer open and see what we have. It sounds like either add a conductor from X0 to the first OCPD, or install a jumper inside the transformer? I know the transformer was originally with the machine and reconnected as it was, which is why this was missed, but that may mean the jumper was never in transformer. Can a jumper be made, or should that come from the transformer OEM? I can't believe a massive company with inhouse electrons would not have grounded things properly so I will dig into it.

 

[don_resqcapt19]

I will have to get the transformer open and see what we have. It sounds like either add a conductor from X0 to the first OCPD, or install a jumper inside the transformer? I know the transformer was originally with the machine and reconnected as it was, which is why this was missed, but that may mean the jumper was never in transformer. Can a jumper be made, or should that come from the transformer OEM? I can't believe a massive company with inhouse electrons would not have grounded things properly so I will dig into it.

If you don't install the system bonding jumper and the GEC at the transformer, you will have to add two conductors between the OCPD and the transformer. If the system bonding jumper is at the OCDP, you need both a neutral and a supply side bonding jumper between the OCPD and the transformer. If there are no line to neutral loads, and you install the system bonding jumper at the transformer, you will need to add a supply side bonding jumper between the two points.

System bonding jumpers, where installed at a transformer are typically field supplied and installed. If the system bonding jumper is installed at a panelboard, and the panelboard is suitable for use as service equipment, then the bonding jumper is typically provided by the panelboard OEM.

 

[fastline]

If you don't install the system bonding jumper and the GEC at the transformer, you will have to add two conductors between the OCPD and the transformer. If the system bonding jumper is at the OCDP, you need both a neutral and a supply side bonding jumper between the OCPD and the transformer. If there are no line to neutral loads, and you install the system bonding jumper at the transformer, you will need to add a supply side bonding jumper between the two points.

System bonding jumpers, where installed at a transformer are typically field supplied and installed. If the system bonding jumper is installed at a panelboard, and the panelboard is suitable for use as service equipment, then the bonding jumper is typically provided by the panelboard OEM.

It would seem largely more practical to simply install the X0 bonding jumper in the transformer, yes?

 

[LarryFine]

Way back when Ie

It would seem largely more practical to simply install the X0 bonding jumper in the transformer, yes?

Yes.

 

[don_resqcapt19]

It would seem largely more practical to simply install the X0 bonding jumper in the transformer, yes?

I would always put it there for a system that does not supply line to neutral loads.

If the system supplies line to neutral loads, it goes in the location that results in the shortest grounding electrode conductor length.

 

[fastline]

I am still rather confused how/why this particular machine's transformer doesn't have a bonding jumper. The machine was originally in a big facility with nothing but pros in the house. Would it even be proper to leave a machine totally isolated? I mean, the machine was otherwise certainly equipment grounded, but if the secondary has no reference to ground, it is effectively 'floated'.

It is pretty common with bigger equipment to run 480 to a machine and set a dedicated transformer within feet of it, as was this case. The machine was moved and transformer was pretty much just hooked back up as it was.

No arguments here, just asking the question as we need to set some 'ground rules' so to speaking moving forward.

 

[LarryFine]

No arguments here, just asking the question as we need to set some 'ground rules' so to speaking moving forward.

 

[augie47]

I am in agreement with don (and others)
That said, I have seen installations approved where the transformer was treated as an ungrounded system where the SBJ was omitted (as if you had a 480D to 240D transformer) In those situations you still had a SSBJ and ground detection per 250.21(B). Perhaps that was allowed at the previous location.
A side note: When you actually view the transformer connection you will likely see a copper strap from the core to the case. This is ground on the core and not a SBJ. Folks often see that and think XO is bonded via that strap.

 

[kwired]

Question might be is this actually a 208/120Y secondary. I think most are assuming it is simply because it is 208 line to line, but the fact it appears to be part of some machine and there is no 120 volt loads and OP kind of never mentions any X0 terminal in the transformer it could be possible it is 208 delta secondary. If so then it needs to either be corner grounded or set up with proper ground detection and run as an ungrounded system.

 

[fastline]

Question might be is this actually a 208/120Y secondary. I think most are assuming it is simply because it is 208 line to line, but the fact it appears to be part of some machine and there is no 120 volt loads and OP kind of never mentions any X0 terminal in the transformer it could be possible it is 208 delta secondary. If so then it needs to either be corner grounded or set up with proper ground detection and run as an ungrounded system.

The 'transformer' is 208/120 secondary. However, the 'machine' only requires 208 3P. That is pretty typical in these machines. No line to neutral loads at all. They have a multitude of internal transformers and PSUs that make up to 20 different voltages for different things.

This is why there is no need or even a place for a neutral conductor to the equipment.

 

[don_resqcapt19]

The 'transformer' is 208/120 secondary. However, the 'machine' only requires 208 3P. That is pretty typical in these machines. No line to neutral loads at all. They have a multitude of internal transformers and PSUs that make up to 20 different voltages for different things.

This is why there is no need or even a place for a neutral conductor to the equipment.

Is this transformer internal to one piece of listed equipment?

 

[fastline]

Is this transformer internal to one piece of listed equipment?

No, this is an external Acme transformer. these machines at lower voltage but huge amps, so it becomes practical to get the power to the equipment at 480, then kick it down.

 

[don_resqcapt19]

No, this is an external Acme transformer. these machines at lower voltage but huge amps, so it becomes practical to get the power to the equipment at 480, then kick it down.

OK, then, assuming the transformer has an XO, it is required to be a grounded system per 250.20(B)(1).

 

[kwired]

The 'transformer' is 208/120 secondary. However, the 'machine' only requires 208 3P. That is pretty typical in these machines. No line to neutral loads at all. They have a multitude of internal transformers and PSUs that make up to 20 different voltages for different things.

This is why there is no need or even a place for a neutral conductor to the equipment.

Then it is sounding like it is not grounded. You may have electrodes, conductors, and such all done properly but not a solid connection to the neutral point of the secondary for whatever reason. The resulting measured neutral is a floating position that can change with conditions changes.
If no connected real loads and no ground faults, then the readings will change with capacitance changes on the secondary conductors, a solidly connected neutral however will stabilize capacitance to neutral/ground.