The SR45 as we've done it runs a power supply current of 60mADC, so that is the minimum current rating for the power supply choke. The Hammond 1255J is rated 30mADC I believe, and the Magnequest BCP-15 is rated 50mADC, so neither one is acceptable in the power supply.
As I have posted before, the operating point of the 45 in this circuit is optimized for the 3000 ohm load. This is a perfectly good operating point for the 45, and sounds very good. Other operating points are possible, but would require adjustment of the plate voltage and current to optimally match the impedance, and this would require re-design of the shunt regulator as well.
Incidentally, for anyone else thinking of it, the old Hammond power transformer is not up to the task - you need the PT-2 that was standard on the Paramour II.
Also incidentally, I am not entirely happy with this design - we put it together to try out the concept, and it does indeed sound extremely good, but I'd prefer to set it up with a few changes. Some are for better reliability, and others are for better performance; DC power to the filament is an example. I have a design in mind and will be making a prototype when I get some time to spend on it. This new design will most likely use a 3K or 4K transformer impedance.