Newton Brawn wrote ...

A geometry and voltage gradient problem...

1 - Make a drawing scale 1:10 ( the core will be 10X bigger, the insulation between layers will be 10X thicker, the mag wire diameter wil be 10X bigger. This will you see the gaps and voltages between the turns, between two adjacent layers, between the first layer and the magnetic core, between the last turn and core, between the fist and last turn.

2 - Calculate the max voltage stressing the insulation that is locate between two layers .
( if the transformer voltage ratio is 30V per turn, and each layer has 20 turns, the voltage produced by a layer is 600V and between two adjacent layers will be 1200V. Considering the very small amount of air bettwen wire and insulation film I may suggest use TWO polyester (Mylar) 0.1mm thick film to provide the proper Insulation between layers. If you consider to wind more turns per layer the insullation thickness shall be resized. Going for more than 30 turn per layer is NOT good practice. )

3 - Calculate the creepage voltage over the insulators as Mylar, polyester PVC, polyetilene, ( the kreepage gradient shall not be more than 600V per mm).

4- Calculate the grad in the air gaps, as last turn and core. It shall be not more than 1.5-2kV per mm.

After winding of the coils a impregnation with Alkid vernish will provide mechanical stability and avoid all coils colapse.

Following these rules a dry transformer can be built and work continuosly for years without problems. The same tansform may be imerse in oil with advantage of beter safe factor



I also agree that the wire shalll NOT be more than #34 AWG. For me the limit is #32AWG Evem with experience in winding coils,