Normalizing

Ed this is my first post to your forum,a lot of good answers.
I have a question that has been bugging me. Some smiths say to let the blade cool to a black heat in a dark space then normalize again. I notice you say cool to the air temperature before normalizing again. What happens between black heat and air temperature? I want to get it rigth, a lot of time is spent making a knife to mess up something simple.
THANKS B L Howard

First of all, Welcome to the forums! Glad to have you with us.
If your working with plain carbon steels, such as the 10 series, you can get away with only letting the steel cool to black heat. However, with alloy type steels(52100 & 5160) you will realize a reduction in cutting ability if you don't let the blade cool completely. Without getting long winded, it has to do with the type of stresses that occur when elements such as chromium are present. Some elements require a slower, or longer cool down cycle to achieve full benifit of the stress relieving step.
Allowing the plain carbon steels to cool completely doesn't give any noticable improvements, but it doesn't hurt either. Since I work with 52100 and 5160 a lot, I simple adopted the method for plain carbon steels too. This way if I do things the same, I don't mess up an alloy steel blade by not letting it cool.
Hope this helps!

Thanks for your answer Ed. Another question, what about heat treating L6? The L6 I bought from Adrimral steel has .8% chromnium the same as 5160. do you heat treat it like 52100 & 5160, and wait 24 hous between hardening cycles? How about 01, I know 01 has less chromnium .5%.
B L howard

L-6 will realize some benifit from multiple quench, the difference is kinda subtle, but it is there. As for the O-1, I've played with it and haven't found any improvement from the multiple quench. I would heat treat the L-6 just like 5160 or 52100, and do a single quench with the O-1. I would recommend that any type of oil/water hardening steel be given a triple temper, it just makes good sense to ensure you've got everything converted in the matrix.