Installing a Transformer for Isolation

[kwired]

I agree that what happens in the transformer stays in the transformer.

If X2-X3 is jumpered, but no conductor is attached, it can be ignored.

Absolutely, one good example is a three phase three unit delta bank. There you would have three X2-X3 connections, but no way you can bond all three together or ground more than one of them, or if intent of system is to be a corner ground you leave all three those points "floating".

 

[kwired]

220416-1029 EDT

No one responded to my previous post.

Why is an isolation transformer requested? What problem is it to solve? What is an isolation transformer? There has to be adequate information to provide any reasonable answer.

For example, an ordinary transformer ( ordinary meaning no DC electrical conductivity from input winding to output winding ) will not prevent a large spike on the input waveform from passing thru, nor will an ordinary transformer prevent capacitive coupling from primary to secondary, nor will it prevent voltage variation on the input from passing thru, etc.

.

That I can also agree with.

Whether there is voltage change or not, most the time we are talking about a "separately derived system" on the secondary, but unless you are utilizing an ungrounded system on either side, they will still essentially have one common conductor in whichever conductor of each system is grounded.

 

[LarryFine]

Absolutely, one good example is a three phase three unit delta bank. There you would have three X2-X3 connections, but no way you can bond all three together or ground more than one of them, or if intent of system is to be a corner ground you leave all three those points "floating".

If there is a compliant way to use a wye secondary as a delta, and leave X0 unterminated, then the same applies here.

 

[kwired]

If there is a compliant way to use a wye secondary as a delta, and leave X0 unterminated, then the same applies here.

Well you could build a wye "bank" out of units with dual volt secondary's and leave all the center taps of each unit floating.

 

[LarryFine]

Well you could build a wye "bank" out of units with dual volt secondary's and leave all the center taps of each unit floating.

That tells us that otherwise-used terminals can be left inside and ignored.

But, must the common point be grounded if it's not needed by the load?

(If the 150v rule would otherwise apply) (While avoiding ground detection)


Does the 150v rule apply to (or, because of) a high-leg delta's A and C phases?


Also, would the same rules apply to an enclosed transformer and one with wire leads?

In other words, if X2 and X3 were external and wire-nutted instead of jumpered lugs inside an enclosure, would that affect their "ignorability"?

 

[electrofelon]

That tells us that otherwise-used terminals can be left inside and ignored.

But, must the common point be grounded if it's not needed by the load?

(If the 150v rule would otherwise apply) (While avoiding ground detection)


Does the 150v rule apply to (or, because of) a high-leg delta's A and C phases?


Also, would the same rules apply to an enclosed transformer and one with wire leads?

In other words, if X2 and X3 were external and wire-nutted instead of jumpered lugs inside an enclosure, would that affect their "ignorability"?

I think my opinion is still that if the connections are available to the electrician, in other words not encapsulated in epoxy or buried in the internals of the transformer, and the system CAN BE grounded to meet 150 v to ground rule then that would be required. If not then it seems like you have to believe that the NEC does not apply to the wiring compartment of a transformer and I can twist and tape, use 18 guage fixture wire, or do whatever I want in there.

 

[wwhitney]

and the system CAN BE grounded to meet 150 v to ground rule

I think the ambiguity here stems from the preceding sentence using the word "system" twice. Here's the start of the section (2017), where I've labeled the two uses:

250.20(B) Alternating-Current Systems of 50 Volts to 1000 Volts. Alternating-current systems (Use 1) of 50 volts to 1000 volts that supply
premises wiring and premises wiring systems (Use 2) shall be grounded under any of the following conditions:
(1) Where the system can be grounded so that the maximum voltage to ground on the ungrounded conductors does not exceed 150 volts.

So if the word "system" in paragraph (1) refers to Use 2 in the opening sentence, then you naturally get the interpretation you have proposed. In fact, you could take that to an extreme and argue that you are not allowed to use a transformer with a 240V 2-wire secondary, as such a secondary could be center-tapped to satisfy 250.20(B)(1).

While if the word "system" in paragraph (2) refers to Use 1 in the opening sentence, you get the interpretation that Larry and I favor, where the "AC system" is defined by which circuit conductors you choose to bring out of the transformer to supply the premises wiring system.

Cheers, Wayne

 

[electrofelon]

I think the ambiguity here stems from the preceding sentence using the word "system" twice. Here's the start of the section (2017), where I've labeled the two uses:

250.20(B) Alternating-Current Systems of 50 Volts to 1000 Volts. Alternating-current systems (Use 1) of 50 volts to 1000 volts that supply
premises wiring and premises wiring systems (Use 2) shall be grounded under any of the following conditions:
(1) Where the system can be grounded so that the maximum voltage to ground on the ungrounded conductors does not exceed 150 volts.

So if the word "system" in paragraph (1) refers to Use 2 in the opening sentence, then you naturally get the interpretation you have proposed. In fact, you could take that to an extreme and argue that you are not allowed to use a transformer with a 240V 2-wire secondary, as such a secondary could be center-tapped to satisfy 250.20(B)(1).

While if the word "system" in paragraph (2) refers to Use 1 in the opening sentence, you get the interpretation that Larry and I favor, where the "AC system" is defined by which circuit conductors you choose to bring out of the transformer to supply the premises wiring system.

Cheers, Wayne

Wayne, Perhaps being a lawyer is your true calling. Have you considered that as a career choice?

 

[wwhitney]

Wayne, Perhaps being a lawyer is your true calling. Have you considered that as a career choice?

I'd rather be a logician. Language would be clearer if there weren't so much overloading (in the computer science sense, where one symbol/pattern has two different meanings that have to be distinguished by context only.)

Seriously, though, does that not reflect the difference in our readings? In which case a clarifying PI could be made to change the word "system" in 250.20(B)(1) to be more specific.

Actually, looking at (B)(2) and (B)(3) they use a different phrasing, "in which the neutral conductor is used as a circuit conductor." Which clearly supports my interpretation and Larry's in those cases.

But then since the language in (B)(1) is different, that supports your interpretation. I'm about ready to cede the point.

Cheers, Wayne

 

[kwired]

I think my opinion is still that if the connections are available to the electrician, in other words not encapsulated in epoxy or buried in the internals of the transformer, and the system CAN BE grounded to meet 150 v to ground rule then that would be required. If not then it seems like you have to believe that the NEC does not apply to the wiring compartment of a transformer and I can twist and tape, use 18 guage fixture wire, or do whatever I want in there.

Now say you have a 480 volt corner grounded and you want a standby generator. Easiest to find unit is 480/277. Does this mean you can't leave the neutral float and ground a phase to match the utility configuration?

If you used three pole breakers on everything (switching the grounded conductor) you could easily connect it to the 480/277 and operate as wye on standby with no problems, but say you didn't switch grounded conductors from the corner ground, then you kind of have no choice but to supply it with a standby system that has a grounded phase conductor.

 

[cosmicelectric]

Thank you all for the abundance of information and deliberate responses. With all of your help, I am confident in properly getting this equipment installed!

 

[Ground Loop Guru]

As an electronics engineer, I've worked in the professional audio world for about 50 years. Audio system noise (the hums and buzzes) is most often, and rightly so, on the premises AC wiring. There is a common perception among people involved with electronic systems, including audio professionals, that an "isolation" transformer will solve their problem. And it can, but not by itself. The origin of their system noise lies in the tiny (less than 1 Volt) differences in "ground" voltage between outlets, including outlets on the same branch circuit. One of the most serious issues is bringing the "ground" voltage at the main-panel neutral-ground bond out to a branch circuit via green wires. The "isolation" transformer can create a separately-derived system where the secondary side neutral-ground bond is at the transformer and the "grounding electrode" is something physically close to the point of use of the power (and NOT connected to the primary-side EGC or conduit). Then this locally-created "ground" is distributed to isolated-ground outlets utilized by the equipment to be "isolated," This often makes a huge difference in lowering noise in the electronic system (say a theatrical sound system or a recording studio). To further lower noise, all conduit wiring on the secondary-side of the transformer should have their L and N conductors (only - not to include green) twisted tightly together. This reduces magnetic induction of noise into the green grounding wire by a factor of about 1,000. I have supervised the installation of several such systems and clients are usually taken aback by how quiet the sound systems are when powered up and at full volume. Anyone wanting to know more about this approach can drop me a line at engineer_bill@verizon.net. I can send copies of papers I've written on the subject. The same techniques should work to reduce random problems in digital data systems as well, but I don't know anywhere it's been implemented.

 

[Speedskater]

'Ground Loop Guru' knows of what he speaks. He has been writing articles, Audio Engineering Society papers and doing seminars for decades.