Dean,
In my case the valve stems were going to be about 1.700" long and I had originally designed them with some extra length on the end for a center drill and tailstock support. One of the few problems I have with the Sherline lathe is that without a tailstock extender, the tee slots on the crosslide are such that if the tool post is in the slot nearest the chuck, there isn't enough travel of the tailstock barrel to cover the width of the crossslide. With the toolpost closest to the tailstock then the left side of the crosslide will tangle with the chuck more easily. They make an extender for the tailstock but its one accessory I haven't bought yet. In this case though I had the piece of 1/8 drill rod...all nice and ground and polished and straight, which the stems need to be. As I had noted in a previous post the valves will not need to be heat treated so the silver solder will not be compromised by that...though I will heat up the bottom end of the stems and quench them to add some hardness there where they will engage the tappets. Finally the only pulling on the soldered joint should be from the valve springs which shouldn't be anywhere near enough force to strain the soldered joint. If it hadn't been for my failure on the first attempt, the whole process for making both valves should have been quicker that turning them from a solid piece, especially considering that the turned stems would have needed some extra time to polish them and work out any slight variations in diameter. Basically I think this two part process is a little more foolproof for the average guy like me, but ultimately time will tell.
Bill