Matching Transformer Primary Conductors to Primary OCPD

[david luchini]

To reiterate post #17 is a simpler way: if the limits in 240.21(B)(3) are supposed to protect tap conductors as well as 240.21(B)(2) does, then the voltage ratio needs to be the worst case voltage ratio. That's a more succinct version of my original post.

Cheers, Wayne

And to reiterate post #9: The Primary-to-secondary voltage ratio of a 480-208/120V transformer is 2.3.

 

[wwhitney]

And to reiterate post #9: The Primary-to-secondary voltage ratio of a 480-208/120V transformer is 2.3.

This is true for the Article 100 definition of "Voltage (of a circuit)," so would definitely apply to the phrase "primary circuit to secondary circuit voltage ratio."

But as the discussion above demonstrated, this answer of 2.3 is non-conservative relative to the standard set by 240.21(B)(2). We could instead apply the definition of "Voltage (Nominal)" and get a ratio of "480V to 208/120V". That's not a typical ratio, which is fitting given the complexities discussed. Taking the higher ratio of 480V to 120V is what the physics guides us toward.

So you've got two choices of interpretation: the NEC means 2.3 in your example, and 240.21(B)(3) is just less conservative than 240.21(B)(2), intentionally or unintentionally. Or the phrase is somewhat ambiguous, given the multiple definitions of voltage in Article 100 for different contexts, and so after some thought we choose the interpretation that is physics-based.

Proceed as you like, but when it's up to me, I'll make the conservative choice, and advocate for that. In the meantime, I should prepare a 2029 NEC PI. What term do you think would clearly indicate the ratio of 4.0 in your example?

My first thought is "worst case secondary-to-primary current ratio." Although looking at all the uses of the word ratio in the 2017 NEC, for some reason the NEC always refers to voltage ratios of transformers, rather than current ratios. So I wonder if "worst case primary-to-secondary voltage ratio" would be sufficiently clear and more likely to be accepted.

Cheers Wayne

 

[david luchini]

Proceed as you like, but when it's up to me, I'll make the conservative choice, and advocate for that. In the meantime, I should prepare a 2029 NEC PI. What term do you think would clearly indicate the ratio of 4.0 in your example?

My first thought is "worst case secondary-to-primary current ratio." Although looking at all the uses of the word ratio in the 2017 NEC, for some reason the NEC always refers to voltage ratios of transformers, rather than current ratios. So I wonder if "worst case primary-to-secondary voltage ratio" would be sufficiently clear and more likely to be accepted.

Cheers Wayne

You can do as you like. I don't see the phrase "worst case primary-to-secondary voltage ratio" anywhere in the Code, however. Nor do I see any reason to inject such a concept into the Code instead of using it's plain language.

It seems to me you are making an argument for not allowing secondary conductors to be protected by primary ocpd's when there is a multi-voltage transformer secondary. I think the Code has already addressed that.

 

[wwhitney]

You can do as you like. I don't see the phrase "worst case primary-to-secondary voltage ratio" anywhere in the Code, however. Nor do I see any reason to inject such a concept into the Code instead of using it's plain language.

And you don't see any problem with the idea that 240.21(B)(3) is less conservative than 240.21(B)(2)?

There isn't any existing phrase that covers the issue raised here. But this issue is sort of a corollary to 240.4(F), in that it arises precisely when 240.4(F) does not permit the primary OCPD to protect the secondary conductors.

[Although speaking of 240.4(F), any reason it shouldn't also allow primary OCPD to protect secondary conductors for a wye-wye transformer? My understanding of the physics is that the primary currents will be a direct multiple of the secondary currents, just like with 2-wire / 2-wire and delta / delta.]

Cheers, Wayne

 

[david luchini]

And you don't see any problem with the idea that 240.21(B)(3) is less conservative than 240.21(B)(2)?

No. I can't imagine that this would be an issue in a real world scenario.

 

[wwhitney]

No.

OK, that explains your position in this argument. For me, it's obviously undesirable to have a 3:1 ratio in most applications, then a 5.2:1 ratio for delta-wye transformers, and a 6:1 ratio for single phase transformers with center-tapped secondaries. I can't believe that's the intent.

Cheers, Wayne

 

[topgone]

Back to the OP, it seems the idea is akin to placing the cart before the horse, or is it not? The method I follow tells me to choose the OCPD based on the load current possible first and then choose the conductor size that will have a much higher current rating than the overcurrent protection device. You not only get your install code compliant, but it also does away with circuitous computation.

 

[electrofelon]

I do agree with the OP that there does seems to be an illogical discrepancy between what we are allowed to do with motors and what we are allowed to do with transformers as far as OCPD sizing. IMO the length allowance in 240.21(B)(3) should be greatly extended.

 

[david luchini]

OK, that explains your position in this argument. For me, it's obviously undesirable to have a 3:1 ratio in most applications, then a 5.2:1 ratio for delta-wye transformers, and a 6:1 ratio for single phase transformers with center-tapped secondaries. I can't believe that's the intent.

Just wait until you hear about the 10' tap rule.

 

[Jim_SWFL]

Whoa.. this went in a much different direction than I planned, but interesting discussion and perspectives.

I do agree with the OP that there does seems to be an illogical discrepancy between what we are allowed to do with motors and what we are allowed to do with transformers as far as OCPD sizing. IMO the length allowance in 240.21(B)(3) should be greatly extended.

By having protection on both sides of the xfrmr, conductors are protected from overload, and the inrush is not really an issue for them; same as with a motor circuit.

I believe I have solved the problem by specifying an adjustable trip that will permit the inrush current of 8 or 9x FLA for .1 seconds, with a long-time setting about 15% above the FLA of the transformer.

This keeps us in a 600A breaker and conductors instead of 700A. I'd still like to have conductors sized for 520A (100% FLA), but this is a reasonable compromise.

 

[jaggedben]

Back to the OP, it seems the idea is akin to placing the cart before the horse, or is it not? The method I follow tells me to choose the OCPD based on the load current possible first and then choose the conductor size that will have a much higher current rating than the overcurrent protection device. You not only get your install code compliant, but it also does away with circuitous computation.

If you want to take advantage of 240.4(B) then this is backwards. You determine the required conductor ampacity first, (which is generally based on nameplate and subject to a code rule in a particular article). Then you choose a conductor with a rating (after adjustment and correction, if necessary) that meets or exceeds the required ampacity. Then you choose the OCPD rating based on the conductor ampacity, which can be the next one higher under 80A, or in special cases like this one even higher. It's no more or less circuitous than the other way, just more correct.

 

[wwhitney]

Just wait until you hear about the 10' tap rule.

Same argument applies--when interpreting 240.21(C)(2), for example, we should use the worst case voltage ratio to get a consistent 10:1 ratio and match the level of protection provided by 240.21(B)(1). [The statement of mine you quoted was implicitly in the context of 25' tap rules.]

Cheers, Wayne

 

[ramsy]

The physics say that if I have a 36kVA, 208V, 3ph load on the secondary of a 480-208/120V transformer, the secondary current will be 100A, and the primary current will be 43.3A. The Secondary-to-Primary current ratio of the transformer is 2.3. The Primary-to-Secondary voltage ratio of the transformer is 2.3.

Thank you for your patience to provide this example exercise.

I see Ground/ing voltages 277/120 at same 2.3 ratio, regardless of wiring xfmr primary xo isolated from secondary XO.

 

[wwhitney]

Thank you for your patience to provide this example exercise.

Yet the case described is not the worst case for the ratio of secondary currents to primary currents.

Cheers, Wayne

 

[ramsy]

not the worst case

Oddball transformers that don't add up at same ratio to Ground tell me to bail out, without engineering supervision, or stamped plans to take complete responsibility.

You could be that indemnified AHJ, or a private architect with errors & omission insurance.

 

[wwhitney]

Oddball transformers

No, I was referring to a the normal delta-wye 480V : 208Y/120V transformer under discussion in your quote. The worst case current ratio is 4.0, the ratio of 480V : 120V.

Cheers, Wayne

 

[ramsy]

worst case current ratio is 4.0, the ratio of 480V : 120V.

Not a design consideration, since not possible.

Please advise how 480v:120v fault or current exists.

I'm having visions of 480v propagating in 120v system by shorting 480v legs, faulting to 277v, blow open 120v loads, or all the above.

 

[wwhitney]

Not a design consideration, since not possible.

See post #10, the post just after the one you initially quoted today.

Even if the 480D : 208Y/120V transformer only supplies 3 wire 3 phase secondary loads, any secondary L-N or L-G fault will be 120V. The corresponding 480V primary conductors will only see 1/4 of the current of the 120V load or fault.

Cheers, Wayne

 

[ramsy]

secondary L-N or L-G fault will be 120V.

Having another vision of 480D ground fault, 240v current ratio=2 @ primary wire OCP, and 208Y ground fault, 120v current ratio=1.73 @ secondary wire OCP.

With both OCP, 480/120=4 and my earlier Pri/2nd Neutral ratios are not used by 240.21, nor by equations in post #10.

 

[wwhitney]

Having another vision

Sorry, your last post was a little too cryptic for me to decipher.

But I'll repeat from post #10 that if you have a 480D : 208Y/120V transformer, with a 100A 120V L-N load on the secondary, the primary current will be 25A on two of the three line conductors. That's a 4:1 ratio.

Cheers, Wayne