With all due respect winnie, the only thing Sahib is trying to do is to twist what Iwire said about the load on the transformer into something about the size load it can manage.
The rest is distraction to cover his original mistake.
I agree with your observation of what Sahib is doing, but I think he has a different basis for his misunderstanding.
As I see it, he is unwilling to admit to his part in a language misunderstanding, rather than trying to correct it. Rather than try to figure out why we are not communicating, Sahib is telling us that we don't understand the physics of transformers.
I believe that he is counting as part of the transformer the wires leading from the supply and the wires leading to the load... and counting power delivered directly from the source to the load as part of the transformer rating. In his language, if the transformer as configured is delivering 15kVA, then it is a '15kVA autotransformer'.
All the rest of us are only counting the windings and core, and saying that any power directly delivered by the source to the load is _not_ part of the transformer rating.
Imagine I sell you a box called a 'voltage raising unit' (not specified to be a transformer). It is rated to go from 208V input to 240V output, with 47A rated output. It has 2 input terminals that you connect to a 208V supply, and it has 2 output terminals.
What would you call the VA rating of this 'voltage raising unit' box?
Now when you open the box you find that internally it has a 240V to 32V transformer connected in a voltage boost configuration. This _component_ of the voltage raising box has a 1.5kVA rating.
I will affirm that a _transformer_ consists of the coils and core, and that any loading on that transformer is from voltage applied across the coil terminals and current flowing through the coils. I do not count power going 'around' the transformer as part of the transformer rating.
-Jon