The load calculated for a (480X208) transformer for multifamily using the optional method (220.84)is significantly smaller than the load calculated using the regular method. It is close but it would allow a smaller transformer (750 vs. 1000).
Are there any disadvantages going with the smaller size?

The advantages are smaller physical size and heat output.

This is a design issue. Are you refering to the POCO xfr?? Our utility co's here routinely install 'small' transformers (13.2 to 480/13.2 to 120/208) on their supply side.

That said, do you have any capacity for future expansion, if you opt for the smaller unit that you reference as 'close'?

IMHO, a good design would allow for future capacity, unless this resi is being built with no possibility for any significant future electrical requirements. (Such as resi apts)

I assumed he was talking about a customer owned transformer.
This may be a Scott question.
At say 600 KVA, is a 1000 KVA more efficient than a 750 (less heat)
I also guess there are higher efficiency transformers for a few extra bucks.
I don't "spec" them I only know how to inspect them

The downside that I could see would be if the actual load turned out to be the higher value and you have to go back and replace everything (on your nickel).
If push came to shove, all that the customer's attorney would have to do is point out that the value determined in the Code was indeed greater than the one that you used and claim that you 'cherry picked' another calculation method that let you use the cheapest transformer.

On the other side of the coin, transformers will handle a larger load than they are nameplated at for brief periods of time and you might be okay based on the load profile of the building...but if everyone starts cooking and turns up their thermostats on the same cold winter day you might be up the proverbial creek.

CGW,
I would never under-spec something like a transformer, especially where it has multiple loads on it, as you state.
You always need a certain amount of operating capacity in a transformer, over and above what your maximum demand calcs tell you, to make sure you don't cause it to overheat or indeed over-saturate.

Thanks for the input.
The question concerns a dry type transformer in the building (the service is 480V and the transformer is for a feeder that serves all the apartments and the 120/208 house loads).
One of the main questions is - do we trust the 220.84 calculation when it is significantly higher than the 220.40 calculation?
Including house load:
220.40 - 945 KVA
220.84 - 664 KVA
750KVA transformer or 1000KVA transformer?

IMHO, 220.40 would be my choice.

1000KVA transformer.

(Keep in mind that it will all come back to the design professional, in the event of any issues)

Several years ago the US government passed an 'Energy Act'. One of the most visible outcomes was the 'killing off' of incandescent lamps, another was the requirement that small power transformers (e.g. non-utility installations) be designed with 'ideal' efficiency at 35% loading.

So a 1000kVA transformer would be most efficient at 350kVA, but it may still be less efficient than a 350kVA unit loaded to 100%.

Have you done a loss (cost of ownership) analysis yet?

Smaller transformers require smaller wire and possibly equipment.

The step-wise introduction of LEDs and less consumptive PCs...now replaced by laptops... means that you would go small.

You're burning kW even at idle. The rare heavy loads can be sustained for hours on end for such transformers.

Be certain that ventilation is good and pure. If it is not so then you have to go large.

You would not believe how hot some e-rooms get because the design disregarded heat-rejection from a dry-type transformer.

In Australia POCOs uusually close spec their trannys,have seen older units rated at 2mva,to 100kva tolerate 200% overloads on winter evenings for 12hrs or more,these are oil cooled with fins,of corse chinese stuff wont cop this for very long.

Ditto for American Pocos.

Transformers burn juice just at idle. An idle that runs 24/7 -- 365.

Consequently, every Poco I've known undersizes its transformers. It's a small matter to swap out to a larger unit -- since the pads and wires are pre-engineered to take a bump to the next rating up.

Normally, EUSERC Pocos use loop/ ring Primaries. This means that they can cut-out a given transformer by switching left and right -- and then hoist the old out and plop on a new one. You'd be surprised at how fast they can do this.

Star Primaries go even quicker.

Distribution losses -- in juice -- run beyond 10% of all power generated -- in most systems. So cutting that drain down is a BIG deal for all Pocos.

BTW, they monitor the loading on their system at every level constantly. It's the primary way that they spot cheats, too. Unmetered power pops up as a power drain akin to a bleeding hot, something that is looked for at all times.

The juice consumed by even one home would have any Poco tearing its hair out -- as kilowatts evaporating into the ground/ sky from one Primary loop/ star is too much to bear.