jarsanb,
thank you!
You got me.
It's truly hard to interpret this a bit more detailed from the picture and of course it might be as well that the weld by itself could have weaken the smaller tube's wall for an amount, of that even the reduced cross section was torn in that area of dilution. This again would eventually explain the "regularity" of the path of cracking.
But yes, I would agree with you, and as well as described in your second post this detail has caused me additionally to ask the question. There is sure a "trace" of a "positive - negative" reflection in the path of crack which looks like a kind of "serration", please see also the embedded picture.
But nonetheless it looks a bit strange for me.
It might be interesting to know what the thickness of the smaller tube was. When having a look upon the gouge in the smaller tube, it could have a higher wall thickness since it doesn't appear to have been broken through its wall thickness cross section.
This again could cause us to consider a higher heat dissipation into the smaller tube and this again leads me to cite you (I hope you don't mind) "...looks like this could have been severe lack of fusion along that bottom leg...".
But to repeat myself, just as a humble personal interpretation of the low resolution picture.
Perhaps evwelder will finally point us in the right direction.
Thanks and best regards,
Stephan