I support efforts to improve conductivity and efficiency, but there’s been discussion in this thread about whether we should be pushing this emitter much harder. In terms of actual emitter current, the bond wires blew at 9.4A in koef3’s CW test.
It’s not clear what portion of the tail current actually passes through the emitter in the FT03, but some folks, like yourself, are measuring over 9A at the tail. keof3 considered 7.4A emitter current a safe maximum and efficiency sweet spot; anything higher overheats the bond wires. This can be seen as a (slightly) steeper rise in Vf above 7500mA in keof3’s chart titled “LED TEST”.
At some point, the emitter becomes a hybrid LED/incandescent light source and this is a prelude to no light at all 
.