There is a huge modulus gradient there. Highest stresses are carried on the extreme surfaces and you have a discontinuous surface there inboard that goes from forged tubing to forged tubing weld HAZ to cut plate. There should be cyclical majority compression there under braking. You are lucky you were not in a spin traveling backwards under full braking where the spindle/wheel could have been pushed under the car as the spin came around.
No big deal, but I would suspect this could be made easier to manufacture consistently by designing a single plate to join three tubes where the tube perimeter surfaces have no welds on outside edges. everything preheated in the jig and welds started mid length each tube and run outboard towards end of tube.
Alternatively, two mandrelled tubes and then a taco gusset plate to form the third tube inboard.
Either way, maybe consider no extremity loaded surface with a weld zone in cyclical loading?
No big deal, but I would suspect this could be made easier to manufacture consistently by designing a single plate to join three tubes where the tube perimeter surfaces have no welds on outside edges. everything preheated in the jig and welds started mid length each tube and run outboard towards end of tube.
Alternatively, two mandrelled tubes and then a taco gusset plate to form the third tube inboard.
Either way, maybe consider no extremity loaded surface with a weld zone in cyclical loading?