This brings up another very important factor that needs to be understood, namely, what was the basis for choosing “800°F” as the defining temperature in the first place. Take a look at Fig. 2, which shows a “thermometer” scale with the melting temps for a variety of different pure metals.
Notice that near the bottom of the thermometer is a group of metals (zinc, lead, cadmium, tin, etc.), and that there is a significant temperature gap between zinc (787°F) and aluminum (near 1220°F). It was therefore felt to be convenient to pick a nice “even number” such as 800F (just above the melting temp of zinc), to officially define brazing.
Thus, 800F merely represented a number “pulled out of the air”, representing a convenient, even number somewhere between the melting point of zinc and aluminum, for the purpose of creating a written definition, and had no real metallurgical significance beyond that.
The metric world, however, found such a number to be inconvenient, and obviously would want to choose to define brazing a bit differently, while still placing their chosen temperature somewhere in that gap between the melting temps of zinc and aluminum.
But, in order to present a “unified” picture to the world, a compromise was needed in how to define the temperature that would be used to separate brazing and soldering. In subsequent meetings of the AWS Committee on Definitions, Symbols, and Metric Practice, it was first of all agreed that the primary system for measurement should be metric (SI), and then, that 450°C would be the temperature to be used in the definition of brazing, as shown in Fig. 3.
I am personally a bit surprised that because the line that separates soldering and brazing is a completely artificial line, merely picked “somewhere between the melting points of zinc and aluminum” for the convenience of a definition, why wasn’t 500C the final, agreed-to temperature selected instead of 450°C, just as the earlier AWS committee here in the US had chosen the nice, even number of 800°F?