Locking out advance with welds

View attachment 1714996061
The inner shaft with the spirals is a bit loose to the top hat due to clearances. Welding will very likely, pull the top hat to a side, or make a wobble. And pulling to side, changes when spark triggers, advancing on near side, retard on far side. The set screws as suggested and use of dial indicator can null out run out. On VR distributors, welding may result in magnetization, and that can mess with sensor signal. VR sensing is inaccurate enough in standard form. I use tab wheels with significantly increased diameter, and digital outputs for accurate timing reference. I found that blueprinting distributors helped, improving the sensing technology with optical tab sensing is far superior, and cost effective, with features not possible with VR.

You can say nonsense and that is OK. My wife points out your white Dart, on occasions when it is stalled at a busy intersections in M'boro. She looked at me and said you have given me cars to drive for over 40 years, they are not new, but I have never been stalled.
Ha! I'll eventually fix the gas gauge in that thing. I'm one of those eternal gasoline optimists that is forever coasting into stations...sometimes.. OK, often, I miss.
Back to the original issue. If the weights are slid all the way out, and evenly bottomed in their respective slots when you weld them, the relationship of the upper and lower half's cannot change.
It would be more likely there be a varience from the factory using two different springs, than two even welds.
Been doing them this way since the 70s and have had zero issues, ever