I have fiddled with this some. Brazing is a form of welding where two dissimilar metals form a molecular bond at a melted interface. The welded metal is usually steel or cast iron and just barely reaches a pliable temperature and the bonding metal is bronze rod which is completely brought to melting temperature. Obviously ,the rod needs to melt at a lower temperature than the stock.
Cymbals are a candidate for fusion welding where a bronze rod of a predictable melting point could be flowed into the cymbal crack, which would just reach a melting point. I've done lots of this successfully with cast iron fusion and managed to avoid subsequent spidering and cracking during cooling but my initial attempts at doing A. Zildjian cymbals met with success on the fusion side but there was subsequent cracking around the weld.
There are two things here. I was working with a few samples of newly developed rod, that came in a variety of alloys. I only tried a few of them , so perhaps I got it too hot and my cooling technique was faulty---I should have preheated the entire cymbal better and wrapped it in heavy insulation to cool.
There are now some very sophisticated welding techniques unavailable to me----laser welding, tig and others that might do a good job. I was using a very fine tipped torch,oxy / acetylene.------but I have thought about this quite a bit and it seems to me that the way to go is to duplicate the method of manufacture---by melting the affected area and do what was once called blacksmith welding, which would hammer the cymbal to weld it under controlled forge and torch heat----basically resetting the bronze crystals along and around the crack and then retempering. low tech but a lot of work!!