I asked about merging the BB/CS and BB/ST junctions last week. I got a lot of helpful answers on- and off-list. Thanks
One idea that came from those many answers was to to experiment with improving the internal fillet of the brazes. I reckon if I can get a decent internal fillet, then I reduce the need for big glob tacks outside the joint, and it probably gives a all-around grip on the joint anyway.
I swear I heard that some people put rings of brass inside the tubes to assist building a (big) internal fillet. Definitely it's a technique using silver for fork ends, for example.
I turned down a mandrel and used it to wind lengths of Cycle Design brass into springs that fit into the interior of the seat tube and expand against the edges.
I fluxed it up and fitted it to a BB shell (good tight miters). I heated from the outside, but didn't get any brass flow-out. I heated the tube longer and hotter than I would have tolerated if I was brazing normally. Still no flow.
So I cut the tube off (it was tacked to the shell) to inspect. No brass melt at all. I probably should have expected this. Even with the brass touching the walls of the red-hot tubes, there simply isn't enough heat transfer to bring the brass to melting temperature. I guess a scientist would say that I was only heating the internal brass via conduction. Missing out on convection and radiation which I'd get on an outside joint.
Still curious if I heard right that people do manage to make this work.
Anyway, so that experiment is over. And now I'm left with a shell and a tacked-on ST stub. So I worked on deliberately building an internal fillet (from the outside).
Remember these were quite tight miters. What I did was get a close, hot flame, get it all cherry, dab (much less than normal) amount of brass, and then heat it and walk it along the miter edge. I notice the edge sucking in the brass immediately. And it appeared that after that first drink, trying to jam subsequent amount of brass in, failed. All I managed to do was cook some of the outer flux. (I was using a harsher, closer flame, like when I tack than when I braze a fillet). Then I did the other side, and this time tried to avoid forcing in more brass after the golden moment at the start.
Results was a small, even fillet inside the circumference of the tube. I +think+ this is probably good. I can see how I could manage to cram a whole lot more brass in there. You can see on one side of the tube that the overheating drew some of the brass up the seat tube. This was from me trying to coax more brass into a reluctant, plugged miter joint. The other side doesn't show this.
Attaching a picture of those results.