If threaded under power could the die or dieholder make contact with the shoulder and then as forward progress is prevented even if momenterrialy partially destroy the thread and reduce the diamenter?
I would never use a solid die by choice to produce a new thread, I consider them as "die nuts" good only for running down a thread if it needs restoring. I always start with a split die opened up in the stock for the first pass and then close up and check the fit of the nut. Single pointing on the lathe is best, and using an internal threading bar behind the workpiece with the lathe in reverse, it is easy to cut the thread right up to a shoulder.
Some joy. I ordered a split die and taper/plug tap from Drill Services Horley which arrived yesterday.
This appears to have solved most of my issues. I turned the stock to 4.98 and a finished thread came out at 4.97 which of course is a bit tight for the tapped holes, but at least I have some adjustment now. I am reluctant to say the first die was faulty because I don't know, it's possible that the helpful posts have caused me to adjust my technique, either way I can proceed and I think I may well remake the safety valve columns to ensure thread consistency and also try some Colphos. I do get the impression that the brass is a bit stcky and can clog the die. Never stop learning do you. Thanks to all of you who helped.