Question about NEC 240.21 and 450.3

[nsidebottom]

I was curious about protection of transformer secondaries. From reading 240.21 (C) (2), I get the impression that the NEC is more concerned with the actual configuration of the transformer as opposed to the actual number of wires coming off of the transformer.

For instance, if I had a transformer where the primary is a single 120V coil, and then I had two separate single-phase coils would that setup still benefit from the exemption from requiring a secondary OCPD? (not split-phase 120/240, but something like you would commonly see in an appliance -- e.x: 120V primary --> 24V secondary (pair of wires) and another separate 60V secondary (separate pair of wires)).

If the answer here is no, what if I was only using one of the secondary coils, so the transformer was being used as a single phase 2 wire in 2 wire out transformer? Whenever I come across a small transformer, they are usually setup in a way to be able to output multiple different voltages by having individual sets of coils with only one of the secondary's being used. It seems like it would be a pain to have to put OCPD's on the secondary conductors in these instances as they are essentially the same as the systems explicitly called out by the NEC as not needing the OCPD.

 

[wwhitney]

From reading 240.21 (C) (2), I get the impression that the NEC is more concerned with the actual configuration of the transformer as opposed to the actual number of wires coming off of the transformer.

I think you mean 240.21(C)(1), "Protection by Primary Overcurrent Device"?

If you have a transformer with multiple independent secondary coils that are not interconnected, I would say the language in 240.21(C)(1) permits you to treat each secondary coil separately. Certainly the physics does.

Cheers, Wayne

 

[nsidebottom]

I think you mean 240.21(C)(1), "Protection by Primary Overcurrent Device"?

If you have a transformer with multiple independent secondary coils that are not interconnected, I would say the language in 240.21(C)(1) permits you to treat each secondary coil separately. Certainly the physics does.

Cheers, Wayne

Thanks Wayne!

You are correct it was 240.21(C)(1), "Protection by Primary Overcurrent Device". I figured that I should be good since electrically they are equivalent, but I wasn't sure . . . ZI wanted to be cautious as sometimes the NEC can be pretty silly with getting unnecessarily pedantic -- a MechE I work with was really upset when I told him we aren't allowed to secure wires to raceways even if those wires are otherwise allowed to run alongside the raceway solely due to the NEC's concern that people would put too much weight on the raceway. He said something along the lines of "if they're worried about a few tiny wires on the outside of steel tubing I'm surprised they aren't worried about a fly or mosquito landing on it too!". Funny guy haha.

 

[wwhitney]

I wanted to be cautious as sometimes the NEC can be pretty silly with getting unnecessarily pedantic

Then you may want to wait for other opinions--my reading of 240.21(C)(1) is certainly motivated by the underlying physics. It does use the word "the" in one place where it would be better if it said "a", to coincide with the reading of "a" as "one of one or more" rather than "a" as "one and only one".

a MechE I work with was really upset when I told him we aren't allowed to secure wires to raceways even if those wires are otherwise allowed to run alongside the raceway

300.11(C) has two provisions that may allow this. 300.11(C)(2) might apply to the MechE's wires. Not sure what raceways might be "identified" as a means of support for 300.11(C)(1).

Cheers, Wayne

 

[jim dungar]

Certainly the physics does.

Really?

How much could you overload one of the secondary windings before the primary device opens, if the other winding is minimally loaded?

 

[wwhitney]

How much could you overload one of the secondary windings before the primary device opens, if the other winding is minimally loaded?

Good question, but that would be covered by article 450, not 240.21. Section 240.21(C) is concerned only about protecting the secondary conductors. Which as far as I can see, when you have a transformer with multiple independent secondary coils/circuits, with no interconnection between them, can be determined on a secondary circuit by secondary circuit basis.

Cheers, Wayne

 

[wwhitney]

How much could you overload one of the secondary windings before the primary device opens, if the other winding is minimally loaded?

Thinking about your question some more, seems like the NEC lets you end up with a situation where you can overload a transformer without overloading any OCPD. An example:

Take a 4.8 KVA transformer (to make the math simpler) with 240V 2-wire primary and 120/240V 3-wire secondary. The rated current on both the primary and secondary is 20A. So supply the transformer with #12 Cu on a 20A OCPD, that 100% primary OCPD protection satisfies 450.3(B) without regard to secondary protection. On the secondary side run two sets of secondary conductors, each set #12 and at most 25', each set landing on a 20A OCPD. That satisfies 240.21(C)(6). Then load the same 120V leg of each set of secondaries to 16A, with 0A on the other legs. That puts 16A on the primary and 32A through one of the two secondary coils, which is only rated for 20A.

But I don't see any NEC violation?

Cheers, Wayne

 

[tortuga]

Take a 4.8 KVA transformer (to make the math simpler) with 240V 2-wire primary and 120/240V 3-wire secondary. The rated current on both the primary and secondary is 20A. So supply the transformer with #12 Cu on a 20A OCPD, that 100% primary OCPD protection satisfies 450.3(B) without regard to secondary protection. On the secondary side run two sets of secondary conductors, each set #12 and at most 25', each set landing on a 20A OCPD. That satisfies 240.21(C)(6). Then load the same 120V leg of each set of secondaries to 16A, with 0A on the other legs. That puts 16A on the primary and 32A through one of the two secondary coils, which is only rated for 20A.

So your saying tap one and the same 120V secondary with two 16A 120V circuits on a 20A OCPD?
Each coil would have a rated current no?

 

[wwhitney]

So your saying tap one and the same 120V secondary with two 16A 120V circuits on a 20A OCPD?
Each coil would have a rated current no?

If the secondary terminals are A, N, B, then I'm saying run two sets of secondary conductors to them, each with 20A OCPD, say A1, N1, B1 and A2, N2, B2.

If we try to draw 16A A1-B1 and 16A A2-B2, the primary current is 32A, and the 20A primary OCPD will trip. But if we try to draw 16A A1-N1 and 16A A2-N2, the primary current is only 16A, under the 20A primary OCPD. But one half of the secondary coil (the A-N half) will see 32A, over its 20A rated current.

Cheers, Wayne

 

[nsidebottom]

Then you may want to wait for other opinions--my reading of 240.21(C)(1) is certainly motivated by the underlying physics. It does use the word "the" in one place where it would be better if it said "a", to coincide with the reading of "a" as "one of one or more" rather than "a" as "one and only one".


300.11(C) has two provisions that may allow this. 300.11(C)(2) might apply to the MechE's wires. Not sure what raceways might be "identified" as a means of support for 300.11(C)(1).

Cheers, Wayne

Thanks again Wayne.

Just to clarify, and make sure I'm not missing anything, you were referring to 300.13 (C) Raceways Used as Means of Support? From what I understood, you are essentially only allowed to do this for lighting or in the case where you are running control wires to a piece of equipment which the wires in the raceway you are tying to is feeding (like powering a motor with 480VAC in the raceway, and strapping a 12VDC circuit to feed the motor controller to the outside of the raceway.)

Not sure if this is the case as I am relatively new to the NEC and would be extremely happy if there was an exception as that would really help me out. At the moment, I am trying to run a 16AWG 4C cable from a pair of generator controllers to an ATS for auto-start triggering and being able to strap it to the outside of the conduit would make things a lot cleaner. This whole system is inside a conex container and it's looking like I will need to suspend it from the ceiling if I can't run it on the outside of the conduit.

Best,

Nicolas

 

[tortuga]

If the secondary terminals are A, N, B, then I'm saying run two sets of secondary conductors to them, each with 20A OCPD, say A1, N1, B1 and A2, N2, B2.

If we try to draw 16A A1-B1 and 16A A2-B2, the primary current is 32A, and the 20A primary OCPD will trip. But if we try to draw 16A A1-N1 and 16A A2-N2, the primary current is only 16A, under the 20A primary OCPD. But one half of the secondary coil (the A-N half) will see 32A, over its 20A rated current.

Cheers, Wayne

Your very good at spotting holes in the code, thats a sharp point however do you not think in the A-N half example you provided might be a violation becasue that one coil is not protected at its 20A rating? 20A x = 40A.

 

[wwhitney]

Your very good at spotting holes in the code, thats a sharp point however do you not think in the A-N half example you provided might be a violation becasue that one coil is not protected at its 20A rating? 20A x = 40A.

What provision in Article 450 is violated?

Cheers, Wayne