desing substation

[juanmp]

I need to design an electrical substation , but i have 150 motors with continuos application (aprox load 640 kVA), but other 50 motors are intermitent application (aprox load 260 kVA) and other loads.

All motors are three-phase with normal torque characteristics, the values of full-load currents of 430.150 NEC table.

How can I consider the intermitent application motors for the substation electrical design (operating capacity)? Do I have to consider the load at a 100% or I have to consider any other value, according to NEC?

 

[LarryFine]

Dang! I wanted to see how to stop a substation from singing!

 

[jtester]

Look at 430.24. If the Exceptions apply you might be able to diversify the load, otherwise they are all added together. Also look at 430.26. These both apply to conductors, but they'd be a reasonalbe approach to a substation.

Jim T

 

[bob]

I need to design an electrical substation , but i have 150 motors with continuos application (aprox load 640 kVA), but other 50 motors are intermitent application (aprox load 260 kVA) and other loads.

Since you have 640 kva of continuous load, your minimum transformer would be at least a 750 kva transformer. As a guide, table 450.3B allows you to fuse the transformer at 125% for primary protection. If you assume that the
max load can be 125% of rated load, 750 x 125% = 938 kva. Total connected load is 640 + 260 = 900 kva. Therefore it appears that that a 750 kva transformer would carry the load. I assume that this is a permanent installation. I would want to consider a 1000 kva transformer. You need to include projected future additional loads. I don't know difference in cost between a 750 and 1000 kva transformer, but the 1000 kva would solve you problems.

 

[kingpb]

Increasing the size to accomdate growth is certainly a good idea. The cost from going to 1000KVA from a 750KVA may be as much as $7500.

If you look at th other thread regarding fans on a transformer, this is a perfect application where you need more KVA rated, then you will actually need running. Therefore, fans to get the additional KVA may be the best alternative.

 

[jtester]

Increasing the size of the transformer is much more expensive than just the purchase price. The losses incurred by using a larger than necessary transforemer likely would far exceed even the increase in purchase price.

The original post was questioning whether he had to add all motors together even though some were intermittent. 430.26 could allow for significant diversity if the motors don't all run at once. The AHJ has the say so, but a good argument could be made based on past experience, on limitations based on production, etc.

Jim T

 

[coulter]

There is always 220.21. I have used that a lot. Of course I have always felt one should be able to defend the choices.

 

[coulter]

...The losses incurred by using a larger than necessary transforemer likely would far exceed even the increase in purchase price. ...

jt - Educate me a little here. I'm missing the point. Usually you have pretty good points and I generally get them. This one is eluding me.

 

[jtester]

Coulter

When you evaluate losses in a transformer, you can break them down into 2 kinds, no load and load losses. This helps evaluate the total cost of owning the transformer.

No load losses are imposed on the system 24/7 and are energizing losses. They are constant. Load losses are obviously a function of transformer loading and are variable.

A 1000 kva transformer is 33% larger than a 750 kva, as proposed in the above suggestion. Along with the increase in KVA will be much larger no load losses. Since they are always imposed on the system, the cost of them becomes very significant.

There wil be a decrease in load losses at a given kva loading, because the 1000 kva windings are probably larger than the 750 kva windings, but since most transformers aren't loaded anywhere near their maximum, the difference will be less than one would imagine.

Jim T

 

[coulter]

jt -

A 1000 kva transformer is 33% larger than a 750 kva, as proposed in the above suggestion.

We're on the same page. That is the kind of differences I was thinking of - as opposed to a 20MVA for a 1000kVA

No load losses are imposed on the system 24/7 and are energizing losses. They are constant. Load losses are obviously a function of transformer loading and are variable.

This could be where I am having trouble. I'm translating the first into "magnetizing current" and the second into I^2R losses. I'm paying for the I^2R cause that's real power. The magnetizing current is 90deg out and makes VARS. As long as my pf is not in the toilet (better thatn .85 lagging), I'm not paying for the VARS.

...There wil be a decrease in load losses at a given kva loading, because the 1000 kva windings are probably larger than the 750 kva windings...

Looking at the specs of the so called "energy efficient" xfmrs, It appears the main differences are:
1. more iron in the core so as to not drive as close to saturation.

2. more copper (or aluminum) in the windings to cut the I^2R losses.

This is the second place I suspect I am goofed up, cause this sounds like a larger xfmr.

... but since most transformers aren't loaded anywhere near their maximum, the difference will be less than one would imagine.

Yes, I would have expected that to be true.

carl

 

[jtester]

Coulter

Let me try to answer your questions

1. "This could be where I am having trouble. I'm translating the first into "magnetizing current" and the second into I^2R losses. I'm paying for the I^2R cause that's real power. The magnetizing current is 90deg out and makes VARS. As long as my pf is not in the toilet (better thatn .85 lagging), I'm not paying for the VARS."

The magnetizing losses still have a resistive component. Whether the current is in phase, out of phase, or whatever, it still creates heat when it flows thru the primary windings. Those are the no load losses, and are extremely variable at the transformer design by adding core material for example. A manufacturer can vary those by probably 50% in magnitude through primary winding material and core components. They have amorphous steel that produce very low no load losses.

2. "Looking at the specs of the so called "energy efficient" xfmrs, It appears the main differences are:
1. more iron in the core so as to not drive as close to saturation.

2. more copper (or aluminum) in the windings to cut the I^2R losses."

My point is made to more of a utility specification, a university or other large transformer purchaser, particularly of medium voltage transformers. There I can tell a manufacturer that I'll purchase future transformers based on the lowest total owning cost and it will be determined by (purchase price + $6.50 x no load losses + $3.25 x load losses), or whatever I determine the loss values to be. If you buy enough transformers each vendor will taylor their design around the lowest total owning cost. The combinations are almost limitless.

If you are buying only a few, and not waiting for them to be produced at the factory on an individual basis, you have 2 basic choices, standard or high efficiency.

Jim T

 

[bonding jumper]

Dang! I wanted to see how to stop a substation from singing!

I'm gonna start a thread about transformers humming.

A lot of responses here for someone who dosn't respond to his own questions. Must have stumbled upon this forum and forgets how to stumble back.