Transformer Tap rule

[jaggedben]

... The code still tells us to use the voltage ratio of the transformer, rather than the voltage ratio of an individual pair of windings.
...

Sorry, but you made that up. The code says "primary-to-secondary voltage ratio". That's the entire phrase. It's ambiguous as to whether that refers to windings or total assembly.

In my opinion a person is quite justified in interpreting it as the referring to the total assembly, because the code essentially never requires consideration of equipment internals, especially when not explicitly stated. It never even occured to me to think otherwise before this thread. But, nonetheless, the calculation result under that interpretation gives a different (lower) level of protection to the secondary conductors of a delta-wye than an wye-wye.

 

[jim dungar]

Somebody is not keep it simple.

If you are told that you have a 30kVA transformer with a 208Y/120V secondary what voltage have you used, for your whole career, to determine your conductors.
How many of you have been using the L-N voltage? How many the L-L voltage?

If you are trying to use 120V because that is the coil across a single winding, are you also considering that a single winding would only be rated 1/3 of the total kVA, or 10kVA?

 

[wwhitney]

Somebody is not keep it simple.

It should be as simple as possible while still doing what it needs to do.

For the 240.21(B)(1) 10' tap rule, the minimum tap conductor size means that a 10x overload will cause the upstream OCPD to see its rated current. For the 240.21(B)(2) 25' tap rule, the minimum tap conductor size means that a 3x overload will cause the upstream OCPD to see its rated current.

That seems like the obvious principle that should hold for the 240.21(C) secondary conductor conductor rules. So the phrase "primary-to-secondary voltage ratio" is just a proxy for "current transformation ratio." When there are multiple primary and secondary voltages, the choice of voltage that should be made is so that the resulting ratio matches the worst case current transformation ratio.

Cheers, Wayne

PS For 240.21(C)(2) and (6), the transformer rating doesn't directly enter the calculation of minimum conductor size; all you need are the primary OCPD size and the "primary-to-secondary voltage ratio."

 

[jim dungar]

It should be as simple as possible while still doing what it needs to do.

For the 240.21(B)(1) 10' tap rule, the minimum tap conductor size means that a 10x overload will cause the upstream OCPD to see its rated current. For the 240.21(B)(2) 25' tap rule, the minimum tap conductor size means that a 3x overload will cause the upstream OCPD to see its rated current.

That seems like the obvious principle that should hold for the 240.21(C) secondary conductor conductor rules. So the phrase "primary-to-secondary voltage ratio" is just a proxy for "current transformation ratio." When there are multiple primary and secondary voltages, the choice of voltage that should be made is so that the resulting ratio matches the worst case current transformation ratio.

Cheers, Wayne

PS For 240.21(C)(2) and (6), the transformer rating doesn't directly enter the calculation of minimum conductor size; all you need are the primary OCPD size and the "primary-to-secondary voltage ratio."

You didn't answer my question.
For all the years you have been in the trade, have you used the secondary L-L or the L-N voltage when selecting secondary conductors? I didn't ask about any code section pertaining to protective devices.

 

[wwhitney]

For all the years you have been in the trade, have you used the secondary L-L or the L-N voltage when selecting secondary conductors?

3 phase computations can be done with either the L-L or L-N voltage. Personally, I find it simpler to use and think about the L-N voltage and a factor of 3, rather than the L-L voltage and a factor of sqrt(3).

Cheers, Wayne

 

[wwhitney]

Here's a simple example of the issue:

Say you have a 100A OCPD protecting a spare unloaded 2-wire 240V feeder. You need to supply a single 20A 120V fused disconnect 20' away. You could:

(a) Use a 2-wire 240V to 2-wire 120V transformer, and run secondary conductors to the 20A 120V disconnect. The secondary conductor minimum ampacity under 240.21(C)(6) is 100A * (240V / 120V) / 3 = 67A.

(b) Use a 2-wire 240V to 3-wire 120V/240V transformer. Under one theory above, now you can say the secondary voltage is 240V, even though the load is 120V. And the 240.21(C)(6) computation would become 100A * (240V / 240V) / 3= 33A.

How does the theory in (b) make any sense? None of the currents in the circuit are affected by the unused 240V secondary supply. The ability of the 100A primary OCPD to protect the secondary conductors between the transformer and fused disconnect is the same in each case. Allowing 33A secondary conductors in case (b) is contrary to the level of safety established by case (a) above and by 240.21(B)(2).

Cheers, Wayne

 

[jaggedben]

... I didn't ask about any code section pertaining to protective devices.

Then you're off topic. The OP was about 240.21 (C)(6). The subject is about a minimum code requirement given an primary OCPD, without necessarily knowing load or transformer kVA.

I imagine the scenario under discussion is fairly academic since as far as I've noticed I've never looked at a facility where a transformer was used for a secondary load of less than a third of the primary circuit VA. I'm sure multiple secondary circuits from a single XF have been done somewhere, but I also imagine plenty of people who install transformers could go a career without encountering a scenario where this discussion matters.

 

[jim dungar]

Then you're off topic. The OP was about 240.21 (C)(6). The subject is about a minimum code requirement given an primary OCPD, without necessarily knowing load or transformer kVA.

I imagine the scenario under discussion is fairly academic since as far as I've noticed I've never looked at a facility where a transformer was used for a secondary load of less than a third of the primary circuit VA. I'm sure multiple secondary circuits from a single XF have been done somewhere, but I also imagine plenty of people who install transformers could go a career without encountering a scenario where this discussion matters.

I know the topic.
I know transformers.
I have installed small feeders off of large transformers more than once, especially back in the 80s, before secondary main breakers were required.

I know that the industry standard is to use the L-L voltage on the output of the transformer when determine things like transformer size based on load VA, conductor sizing based on transformer output when not using calculated load, and for 450.3 secondary protective device sizing.

Is it possible to use some other methodology, sure it it, but it is not what electricians are typically taught. I don't think I have seen Ugly's and other industry references, or any manufacturer catalogs or selection tables use anything except L-L voltages when showing output currents.

As I said earlier, if you are trying to determine the current in a single Wye winding, using L-N voltages, you should not be using any three phase values including the overall transformer size.

 

[wwhitney]

I know the topic.

Great, so would you be willing to comment on my post #26?

I think your point so far is that "industry standard usage means that the phrase "primary-to-secondary voltage ratio" definitely refers to the ratio of the largest primary and secondary voltages present." If so, then I have no reason to disagree with that. However, that means that the NEC is using the wrong terminology for what I infer is the intention, yielding a non-conservative ratio to use when applying 240.21(C)(6) and the like.

So current code language aside, I would love to get your take on (a) what ratio makes the most sense to use in 240.21(C)(6) and (b) what terminology would most simply and correctly describe that ratio, for any transformer topology.

Thanks,
Wayne

 

[jaggedben]

I know the topic.
I know transformers.
I have installed small feeders off of large transformers more than once, especially back in the 80s, before secondary main breakers were required.

I know that the industry standard is to use the L-L voltage on the output of the transformer when determine things like transformer size based on load VA, conductor sizing based on transformer output when not using calculated load, and for 450.3 secondary protective device sizing.

Is it possible to use some other methodology, sure it it, but it is not what electricians are typically taught. I don't think I have seen Ugly's and other industry references, or any manufacturer catalogs or selection tables use anything except L-L voltages when showing output currents.

And?
Litterally no one has cast any aspersions on anyone for using L-L voltages when following 240.21(C)(6). Because it seems like an obvious way to read the code. The question that has been raised is whether that code section actually provides the intended measure of safety when interpreted that way.

All the tasks you mentioned in your second paragraph are irrelevant to 240.21(C)(6).

As I said earlier, if you are trying to determine the current in a single Wye winding, using L-N voltages, you should not be using any three phase values including the overall transformer size.

No one had suggested doing that. What had been suggested is that the current in a single wye winding might sensibly be the relevant number for 240.21(C)(6).

 

[jim dungar]

And?
Litterally no one has cast any aspersions on anyone for using L-L voltages when following 240.21(C)(6). Because it seems like an obvious way to read the code. The question that has been raised is whether that code section actually provides the intended measure of safety when interpreted that way.

All the tasks you mentioned in your second paragraph are irrelevant to 240.21(C)(6).



No one had suggested doing that. What had been suggested is that the current in a single wye winding might sensibly be the relevant number for 240.21(C)(6).

I wanted a simple answer.
You seem to want to make it difficult.

The industry standard practice is to use the L-L secondary voltage, as evidenced by almost every manufacturers tables showing secondary amps.

So a 208Y/120V secondary would have a voltage ratio of 2.03 based on a 480V primary. This tells us that 2.03 amps on the secondary is equal to 1A on the primary. NEC 240.21 seems to be okay with protecting limited length conductors that are at least 1/3 of the upstream protection. So effectively the secondary must be at least .77 of the primary device.

 

[jaggedben]

I wanted a simple answer.
You seem to want to make it difficult.

...

I'm not motivated to make it difficult. The 1/3 part of the code requirement is arbitrary anyway. I just think a valid point has been raised that using the L-L ratio doesn't provide the same protection as other 1/3 requirements in 240.21.

 

[jim dungar]

I'm not motivated to make it difficult. The 1/3 part of the code requirement is arbitrary anyway. I just think a valid point has been raised that using the L-L ratio doesn't provide the same protection as other 1/3 requirements in 240.21.

I don't see where there is a difference in protection. I don't think the point being raised is valid.

To me it looks like people are pulling numbers to justify a position which may not be a real life situation.

Is the concern that a protective sized per 240.21(C)(6) will not protect a 100% load connected to only one L-N?
Then how do you justify the 125% secondary value found in 450.3?

 

[wwhitney]

Then how do you justify the 125% secondary value found in 450.3?

Values in 450.3 have nothing to do with conductor protection; they are only about transformer protection. Article 240 covers conductor protection.

I don't see where there is a difference in protection.

Post #26 clearly and simply demonstrates the difference in protection.

The overall picture is that a conductor with an ampacity of 33A (to pick a roundish number) may see the following currents when the upstream OCPD is seeing its rated current:

(a) 35A per 240.4(B) when the conductor length is unlimited and the OCPD is at the source of supply as per the first sentence of 240.21.
(b) 100A per 240.21(B)(2) when the conductor length is limited to 25'.
(c) 200A per 240.21(C)(6) when the conductor length is limited to 25', and we use a strict L-L secondary voltage convention interpreting 240.21(C(6). [Or for the 3 phase delta-wye case, it would be 173A.]
(d) 100A per 240.21(C)(6) when the conductor length is limited to 25' and we use an L-N secondary voltage convention interpreting 240.21(C)(6) when appropriate.

Clearly case (c) differs from case (b). And parity with case (b) suggests that case (d) is what should be required by 240.21(C)(6), not case (c).

Cheers, Wayne

 

[jim dungar]

Values in 450.3 have nothing to do with conductor protection; they are only about transformer protection. Article 240 covers conductor protection.


Post #26 clearly and simply demonstrates the difference in protection.

The overall picture is that a conductor with an ampacity of 33A (to pick a roundish number) may see the following currents when the upstream OCPD is seeing its rated current:

(a) 35A per 240.4(B) when the conductor length is unlimited and the OCPD is at the source of supply as per the first sentence of 240.21.
(b) 100A per 240.21(B)(2) when the conductor length is limited to 25'.
(c) 200A per 240.21(C)(6) when the conductor length is limited to 25', and we use a strict L-L secondary voltage convention interpreting 240.21(C(6). [Or for the 3 phase delta-wye case, it would be 173A.]
(d) 100A per 240.21(C)(6) when the conductor length is limited to 25' and we use an L-N secondary voltage convention interpreting 240.21(C)(6) when appropriate.

Clearly case (c) differs from case (b). And parity with case (b) suggests that case (d) is what should be required by 240.21(C)(6), not case (c).

Cheers, Wayne

First the values in 240.21 are most likely arbitrary rather than being based on any actual science. The fact a transformer exists needs to be considered when making the comparison.

In your example (c) the secondary conductor of 33A can only be considered as being protected if the primary device is not greater than 42A. Given the transformer ratio the secondary current would be limited to 84A. How did you come up with the 200A figure?

 

[jaggedben]

First the values in 240.21 are most likely arbitrary rather than being based on any actual science.

Fair enough, but does that mean one should cut them down by more than 50% by not paying attention to transformer topology? That's a pretty big factor, it's more than 2/3 the difference between the 25ft and 10ft rules.

The fact a transformer exists needs to be considered when making the comparison.

Isn't that what the sections in 240.21(C) tell us how to do? This is a discussion about how to interpret them.

 

[jaggedben]

I don't see where there is a difference in protection. I don't think the point being raised is valid.
...

Please go back and read the first page of the thread, it's all there. The OP thought the voltage ratio is that of the windings, not the L-L voltages. (Btw, what about a transformer with no L-L voltage, as in post #26.?

 

[wwhitney]

In your example (c) the secondary conductor of 33A can only be considered as being protected if the primary device is not greater than 42A. Given the transformer ratio the secondary current would be limited to 84A. How did you come up with the 200A figure?

It's all in post #26, but to recap:

A 100A primary OCPD for a 240V : 240/120V single phase transformer. If we say this transformer has a 1:1 primary-to-secondary voltage ratio, then 240.21(C)(6) gives a minimum secondary conductor ampacity of 33A (for lengths up 25'). But if the secondary conductor is connected only to a 120V load, it would take 200A through the secondary conductor before the 100A primary OCPD sees 100A.

Not sure what calculation you used that came to 42A?

Cheers, Wayne

 

[jaggedben]

It's all in post #26, but to recap:

A 100A primary OCPD for a 240V : 240/120V single phase transformer. If we say this transformer has a 1:1 primary-to-secondary voltage ratio, then 240.21(C)(6) gives a minimum secondary conductor ampacity of 33A (for lengths up 25'). But if the secondary conductor is connected only to a 120V load, it would take 200A through the secondary conductor before the 100A primary OCPD sees 100A.

Not sure what calculation you used that came to 42A?

Cheers, Wayne

I think we should be more concerned with a line to ground fault than with how the load is connected.

It occurs to me that for a given impedance of the fault, the lower L-N voltage may effectively cancel the difference in when the primary OCPD will trip. Not that I've modeled any of this mathematically.

 

[jim dungar]

It's all in post #26, but to recap:

A 100A primary OCPD for a 240V : 240/120V single phase transformer. If we say this transformer has a 1:1 primary-to-secondary voltage ratio, then 240.21(C)(6) gives a minimum secondary conductor ampacity of 33A (for lengths up 25'). But if the secondary conductor is connected only to a 120V load, it would take 200A through the secondary conductor before the 100A primary OCPD sees 100A.

If the primary protective device is 100A how is it going to allow twice as much current to flow through it simply because you are now using only 1/2 of a secondary winding?

The physics of the transformer hasn't changed. The two halves of the secondary are in parallel series 100A flowing through a secondary windings will cause 100A to flow through the primary winding. It does not matter if that secondary current is from L1-L2 or from L1-N.

Correction: windings are in series.