Sizing Transformers

[ssandoval]

Just wondering what the easiest way is to size a transformer. For example, in a commercial/ educational environment, if I needed to install a 3 phase 120/208v panel with 225 amps worth of service would I just use the formula for calculating load side current, I= kVA/(208 x1.73) or for my case, kVA= 208 x 225a x 1.73 ? I came up with about 81 kVA. What am I missing?

 

[iwire]

If your supplying a 225 amp 208Y/120 panel from a 480 feeder the typical transformer size is 75 kVA.

The primary side breaker is usually between 100 to 150 amps.

The rated secondary current is 208 amps at 208 volts

The rated primary current is 90 amps @ 480

 

[ssandoval]

If your supplying a 225 amp 208Y/120 panel from a 480 feeder the typical transformer size is 75 kVA.

The primary side breaker is usually between 100 to 150 amps.

The rated secondary current is 208 amps at 208 volts

The rated primary current is 90 amps @ 480

Thanks for the feedback. I left out the fact that I already knew what size transformer I needed because, as you stated, this is a fairly typical application. What I'm trying to figure out is, from a design standpoint, how do they calculate this rating. For example, we have 3 phase/208-120v/ 100a load centers all over the school district that have 22.5 kVA xfmrs feeding them. If you do the math, a 22.5 should only be good for 62 amps. I have yet to get anyone from our engineering firm to tell me how you can put a 100 amp main in a panel that has a transformer that is only good for 62 amps? I'm sure I'm missing something...

 

[drbond24]

It's because the panel rating isn't the concern, only the actual load. The panel could be rated for a bazillion amps, but you could feed it with a 22.5 kVA transformer if you know the actual load is only 60 amps. The only requirement for the panel at that point is that is has to be rated for at least 60 amps. It can be larger.

 

[iwire]

Two things come to my mind.

1) The transformer must have a higher rating then the calculated load.

2) Table 450.3(B) provides the requirements for transformer over current protection.

With over current protection on both sides of your typical delta Wye transformers the primary side device can be 250% above the rated load current and the secondary side device can be 125% above the transformer rating and can be rolled up to the next standard size over current device.

 

[ssandoval]

Keep them coming. All of this is good advice. In response to drbond24, there's nothing to keep maint personel from adding breakers, and therefore load, other than running out of spaces. Sounds like I need to get with our design/ eng firm to request that if they are going to spec 100 amp load centers then they need to be sure the xfmrs will support future expansion right?

 

[iwire]

there's nothing to keep maint personel from adding breakers, and therefore load, other than running out of spaces.

The NEC expects qualified personal to be working on the systems.

See Article 100 "Qualified Person"

No one should be adding load without knowing that the supply is adequate.

Sounds like I need to get with our design/ eng firm to request that if they are going to spec 100 amp load centers then they need to be sure the xfmrs will support future expansion right?

That is certainly an option if your paying the bill. :smile:

 

[jim dungar]

Keep them coming. All of this is good advice. In response to drbond24, there's nothing to keep maint personel from adding breakers, and therefore load, other than running out of spaces. Sounds like I need to get with our design/ eng firm to request that if they are going to spec 100 amp load centers then they need to be sure the xfmrs will support future expansion right?

It would be better, and more cost effective, to train your personnel not to add load without confirming the system can handle it.

 

[ssandoval]

A couple of things here, unfortunately in our system "maintenance" and "qualified" are mutually exclusive, even though they are state licensed. The other thing is that there are many times that circuits have to be added for whatever reasons, so we are trying to utilize the spares that are there. The problem seems to be that our engineers are not sizing the xfmr large enough to facilitate future expansion. If I have a 30 ckt load center then I darn sure ought to be able to have thirty circuits without overloading the xfmr, I would think.

 

[iwire]

A couple of things here, unfortunately in our system "maintenance" and "qualified" are mutually exclusive, even though they are state licensed.

I understand the real world but I was giving you NEC reasons. :smile:

The problem seems to be that our engineers are not sizing the xfmr large enough to facilitate future expansion.

I am sure the engineers would be more than happy to provide more capacity, I will even say it is likely the first round of drawings had more capacity but when put out to bid and prices came back some 'value engineering' had to be done and not providing capacity for the future is an easy place to save money.

Not to mention having 'over sized' transformers everywhere wastes power.

If I have a 30 ckt load center then I darn sure ought to be able to have thirty circuits without overloading the xfmr, I would think.

You can never assume that.

 

[drbond24]

What I'm trying to figure out is, from a design standpoint, how do they calculate this rating.

I'm not sure anyone answered this directly, although the answer may have become obvious through the course of the conversation.

A designer calculates the actual DEMAND load that needs to be supplied by the transformer. What you are doing is looking at the CONNECTED load, which is completely different. One the designer knows how much current is actually needed, they use that number to size the transformer.

For example, you used the number 225 A to get 81 kVA. In reality, that 225 A may only be using 100 A, in which case the transformer does not need to be as large as 81 kVA. A 50 kVA would be plenty. Yes, it is a good idea to allow some room for expansion, but as iwire has already pointed out that costs money and tends to get cut out of the project later when the budget gets tight.

 

[zbang]

If I have a 30 ckt load center then I darn sure ought to be able to have thirty circuits without overloading the xfmr, I would think.

That's the "I still have checks left" method of cash management .

Yes, expansion capacity is good. No, most people don't bother to check availability past the panel's main.

 

[Wes G]

Remember that transformers are most efficient when fully loaded. If you have access to the data, recalculate the load. If not, put an ammeter on the incomming lines at time of peak use and observe how much power is being used. In a school, I generally like to see between 80 and 85% load at peak times, which leaves plenty of room for the unexpected and for modest expansion. Also compare load with feeder capacity.