Stuff doesn't break from just sitting around, so there is a real good argument that vibration was the cause.

Your engine is on a shock mount from the sound of it. Shock mounts work by letting the engine vibrate without as much connection to the "solid" rest of the airplane. So no matter how much more or how little more the engine was vibrating, it was moving more than it would had it been rigidly nailed down to the airframe. So there was "more" vibartion induced movement from the engine in your setup, or at least more than less.

If you ran all the runs on the ground you actually added "some more" shock to what the broken parts probably would have seen had the airplane been in the air. So there was "more" again. It's really worthwhile to not break an engine in on the airplane, especially if you do the breakin with the airplane sitting on the ground.

Whenever there are flexible parts in a system, they are there for a reason. And a lot of the materials that're flexible get lots less flexible when cold. So there was "more" again. It sounds like the soft parts in the present system weren't really appropriate for the operating conditions they saw. So there is another "more".

I wish there was a picture to look at, but whatever..... If you can, you might re-engineer one of the joints (like where it broke would be a good joint to use) to use some of the very flexible, blue tuned-pipe connector hose that's available in our hobby. It looks like huge fuel tubing. It's even somewhat see-through like fuel tubing.

I'd bet that what happened was just an accumulation of small additions of "more".

If you can use the hose, it'd be worthwhile to see if you can make the joint longer. Just a few more millimeters of length can do wonders to increase the flexibility.