Exceeding Transformer Capacity

[Npstewart]

This is a ridiculous question but where would I find the requirement for a transformer to be sized in accordance with its load. Its not anywhere in 450 that I can find.

Having a very friendly discussion with a colleague whom says the kVA rating of a transformer can be exceeded and he referenced a couple tables in 450 (450.3 (A)/(B)). My position is that it cant be exceeded and the tables referenced in 450 are for over-current protection only.

 

[templdl]

This is a ridiculous question but where would I find the requirement for a transformer to be sized in accordance with its load. Its not anywhere in 450 that I can find.

Having a very friendly discussion with a colleague whom says the kVA rating of a transformer can be exceeded and he referenced a couple tables in 450 (450.3 (A)/(B)). My position is that it cant be exceeded and the tables referenced in 450 are for over-current protection only.

Let's start with the transformer FLA at its nameplate rating. The transformer's ration is based upon a 40cdegC ambient plus the temperature rise of 150, 115, or 80degC+ 30deg is required for hot spot allowance.
Then consider the actual ambient temperature which is often lower than its rating. Then consider the termperasture rise of the transformer which for a dry type transformer is 150, 115, or 80degC rise. Knowing that the common insulation class is 220degC and that the transformer is based upon 40degC ambient and the actual average ambient is 25degC right out of the shoot you have 15degC that works in you favor without exceeding the 220degC insulation class which allows for addition capacity.
The if you transformer is a 115degC rise you have an additional 35degC of capacity without exceeding 220degC and 70degC with an 80degC rise transformer. Just think that you have an extra 15deg to work with in a25deg ambient and a wooping extra 70degC with and 80degC rise transformer without exceeding the 220deC insulation class.
As such a transformer can often be loaded greater that its name plate rating without exceeding the insulation class of 220degC.
As such would it be appropriate to reference art 110 for listing and labeling which the transformer would be applied with in the manufactures design rating?
Also, you also must consider that any addional ampacity may exceed the rating of the bus/conductor as included in the original design of the transformer's nameplate rating in addition to the temperature rating of the terminals.

 

[Tony S]

From experience:

On two occasions following Tx failures a single transformer has taken the full load of a board normally fed by two units with the bus-coupler open.
One occasion the remaining unit being loaded to 130% FLC. The other occasion the remaining unit took 125% FLC.

We fastened 15? fans and baffles to the cooling fins so instead of being ONAN they became ONAF. Like all of our transformers they had oil over temperature trips.
Both ran like that for several months while repairs were carried out on the failed units. That was over a decade ago, as far as I know the two abused units are still in service.

As an aside both the transformers we overloaded were made in the early 60?s, I wonder if a modern unit would take the abuse?:?

 

[Smart $]

This is a ridiculous question but where would I find the requirement for a transformer to be sized in accordance with its load. Its not anywhere in 450 that I can find.

Having a very friendly discussion with a colleague whom says the kVA rating of a transformer can be exceeded and he referenced a couple tables in 450 (450.3 (A)/(B)). My position is that it cant be exceeded and the tables referenced in 450 are for over-current protection only.

Well, if we temporarily forget that Code is fairly conservative on calculated load values, a transformer can be properly rated under Code and be overloaded.

Being properly rated is a cascading of requirements rather than a direct correlation. First our conductor and ocpd ratings are based on calculated load. The conductor size and ocpd rating* cannot be less than noncontinuous plus 125% continuous portions of calculated load and the ampacity cannot be less than calculated load.

*Permitted to be upsized to next greater standard rating at 800A or less.​

When we correlate the preceding with transformer protection, we see the rating stipulation of 125%** is used. As such, a non-continuous load at full calculated load could "overload" a transformer at maximum permitted protection rating for less than three hours.

**Permitted to be upsized to next greater standard rating... but not limited to 800A.​

Next with no consideration for continuous load and no stipulation concerning the use of 100%-rated breakers... a full continuous load on a 100%-rated breaker used to also protect a transformer at maximum would overload the transformer by 25%.

Using an upsized breaker rating increases the potential of having a problem configuration.

 

[iwire]

This is a ridiculous question but where would I find the requirement for a transformer to be sized in accordance with its load. Its not anywhere in 450 that I can find.

Having a very friendly discussion with a colleague whom says the kVA rating of a transformer can be exceeded and he referenced a couple tables in 450 (450.3 (A)/(B)). My position is that it cant be exceeded and the tables referenced in 450 are for over-current protection only.

In my opinion an inspector would be within their authority to cite 110.3(B).

Transformers are listed equipment with a label stating their rating.

 

[Smart $]

In my opinion an inspector would be within their authority to cite 110.3(B).

Transformers are listed equipment with a label stating their rating.

That and 110.2.

However, I think the basic issue being emphasized is Code does not explicitly require transformers be rated for the calculated load they supply.

 

[iwire]

That and 110.2.

However, I think the basic issue being emphasized is Code does not explicitly require transformers be rated for the calculated load they supply.

I don't see any need for explicitly and I bet my local inspectors don't either.

 

[Smart $]

I don't see any need for explicitly and I bet my local inspectors don't either.

I assumed you'd say as much and perhaps so regarding your local inspectors. So exactly how would the inspectors apply a transformer rating to the load supplied: use total connected load, Article 220 calculated load, factor continuous loads, etc.?

 

[iwire]

I assumed you'd say as much and perhaps so regarding your local inspectors.

Of course, you are a legend. :lol:

So exactly how would the inspectors apply a transformer rating to the load supplied: use total connected load, Article 220 calculated load, factor continuous loads, etc.?

I can't help but think this proves there are stupid questions.


(Edit: OK, that was a poor way to express myself but I shoot from the hip.)

 

[Smart $]

Of course, you are a legend. :lol:



I can't help but think this proves there are stupid questions.


(Edit: OK, that was a poor way to express myself but I shoot from the hip.)

Not a poor way of expressing you are full of yourself. :thumbsup:

 

[petersonra]

I don't see anything in the code that requires a transformer be sized to the calculated load.

There are explicit requirements for sizing the OCPD and the conductors, but not the transformer itself.

The xfmr instructions do not say anything about sizing the transformer to the calculated load on any xfmr I have have ever seen.

The calculated load has little to do with the actual load the transformer sees. It is often 2X (or more) what the actual load is.

Having said that, it is not unusual for local inspectors to make up requirements that do not actually exist in the code.

 

[kingpb]

I don't see anything in the code that requires a transformer be sized to the calculated load.

Repeat after me:

The NEC is not a design guide. :dunce:The NEC is not a design guide. :dunce:The NEC is not a design guide. :dunce:The NEC is not a design guide.:dunce: The NEC is not a design guide. :dunce: The NEC is not a design guide.:dunce: The NEC is not a design guide.:dunce:


:roll:

 

[iwire]

Repeat after me:

The NEC is not a design guide. :dunce:The NEC is not a design guide. :dunce:The NEC is not a design guide. :dunce:The NEC is not a design guide.:dunce: The NEC is not a design guide. :dunce: The NEC is not a design guide.:dunce:

:lol:

Take a look at 210.50 through 210.70 and it seems it is.

 

[iwire]

Not a poor way of expressing you are full of yourself. :thumbsup:

Yes I am, as are all the regulars here.

But your question was a set up. You are aware as I am that it can't be answered.

That being the case 90.4 takes over and we will use the method the AHJ decides is appropriate.

Kind of commonsense but I know that is in limited supply.

 

[petersonra]

Yes I am, as are all the regulars here.

But your question was a set up. You are aware as I am that it can't be answered.

That being the case 90.4 takes over and we will use the method the AHJ decides is appropriate.

Kind of commonsense but I know that is in limited supply.

I would point out the the inspector is not the AHJ and has no authority whatsoever on his own to make up requirements not found in whatever code requirements are adopted.

i would also point out the 90.4 explicitly says the AHJ has this power, not a mere inspector. I am not aware of any AHJ that has delegated any code making powers to any inspector.

 

[Npstewart]

I should have checked back on this thread quicker after posting it.

Some background on this discussion; We are doing a very small renovation in a commercial office space with a existing 30 kVA transformer in an existing electrical room. When the existing load is added to the additional loads, the new total load is approximately 38 kVA.

When I opened section 450, I kind of expected to see something similar to what is says for generators, (ie. The transformer shall be sized sufficiently to handle the loads calculated in accordance with 220).

This appears to be more of a discretionary issue with room for interpretation. However, if we were talking about sizing a panel, feeder, or MOCP, there would be much less grey area here.

 

[jim dungar]

Take a look at 210.50 through 210.70 and it seems it is.

While the NEC absolutely contains some design guidelines, it is not a 'design manual for the untrained'. :slaphead:
It should come as no surprise, that there are many design questions that cannot be answered solely by use of the NEC.:jawdrop:

 

[steve66]

As a side note, I am in the habit of adding an extra 25% to continuous loads. I keep forgetting that this extra 25% only applies to circuit breakers and their conductors, and not transformers.

It appears there is no need to add the extra 25% "demand factor" for continuous loads when sizing a transformer. Or at least, there is no code requirement for adding 25% to continuous loads for sizing a transformer.

Can we all agree on that??

Steve

 

[iwire]

It appears there is no need to add the extra 25% "demand factor" for continuous loads when sizing a transformer. Or at least, there is no code requirement for adding 25% to continuous loads for sizing a transformer.

Can we all agree on that??

I can agree on that.

 

[petersonra]

As a side note, I am in the habit of adding an extra 25% to continuous loads. I keep forgetting that this extra 25% only applies to circuit breakers and their conductors, and not transformers.

It appears there is no need to add the extra 25% "demand factor" for continuous loads when sizing a transformer. Or at least, there is no code requirement for adding 25% to continuous loads for sizing a transformer.

Can we all agree on that??

Steve

IIRC, the 25% thing only applies to ampacity of conductors and not directly to sizing OCPD.