1. single filament winding for the 2A3's. Not good if well centered cathodes/filaments are not available( and they are *NOT* )

2. The phase splitting is done with an autoformer. I have not liked the sound of this sort of thing in the amps I have experimented on. A grid choke, functioning like a grid choke with balanced drive has worked *MUCH* better. I had a finger in the design of the MQ device, and abandoned the project. It is good on paper and the bench, but I found its sonics lacking.

3. I think the S-230-Q is a better choice of output TX for class A PP 2A3's. 6k6 instead of 5k for the S-240. The 230 is not available from MQ though...neither is the 20-20 Plus series in this a-a load.

4. B+ is a bit on the high side for that load if Class A bias is desired. It also seems to sound better if the operating point is a bit more conservative. Certainly the finals last longer.

There is the group build forum here, if you wish to take on a project with group support.
regards,
Douglas