The "sink" in the top of these indicates that they did bond, but the fact that they're all off-center means they're marginal, at best. Also note that almost all of them have some kind of lead ring bonded to the case mouth. That is NOT desirable, and causes problems during swaging. So far, this liquid flux is looking like a failure, other than the fact that the bond is stronger.
Some examples of a process I've been playing with, to try to save some of the out-of-round rejects, that are otherwise on-weight and bonded well.
Instead of feeding the jacket and core into the die base-first, I start by feeding it nose-first (if it will enter the die). Then, I turn it over and run it back in base-first. And, finally, nose-first again.
After the abysmal results from yesterday, I may have quite a few of these to experiment with.
I know everyone wants to see some serious accuracy testing, but I need to work a few more bugs out before I can commit to a large batch again. Trust me, I want to shoot a bunch of these even more
than you want to see the photos.