Annealing, Normalizing, multiple quenching and tempering: carbides and stuff...

Alarion

Alarion

Joined
Nov 26, 2001
Messages
1,375
I've read some posts by Kevin Cashen where he explains stuff about how a wrong temper, or annealing can actually weaken the steel, instead of refining the grain, especially with hypereutectoid steels like,m say, 1095 or W-2.
So, I was wondering hwich is the correct way of doing it.
From Kevin's posts it seemed that mutliple tempers were good if the ones following the first, made at about normalizing temp, were made with a quick enough heating of the blade.
I try to heat slowly to avoid thermal stress, but may be I'm doing something wrong here, then, encouraging big, fat carbide boundaries around the grains...
Can somebody (Kevin?) explain some more?
Thanks.
 
Kevin and a bunch of others are at the hammer-in at Ashokan so your stuck with me ! There isn't much reason for slow heating with something simple in shape and small like a knife. If you are worrying about stresses then preheat to 1200F then bring quickly to hardening temperature." multiple tempers " should read "multiple normalizing ". The easiest way to get complete carbide grain boundaries is to anneal holding it for a good time at temperature then very slowly cooling [ as in the turned off furnace] !!!
 
Good question Alarion. I hope Mete and the other metallurgists can go into this some more. (Oh Kevin, where art thou!)

There is a process for grain refinement where you quickly heat then cool the steel over several cycles to achieve super fine martensite grain. The temperature changes are sudden and are accomplished with molten salt. So slow ramp ups are probably unnecessary on a blade. If your furnace heats up slowly (Paragon, Evenheat etc.) I would charge directly into full ausentizing temperature, a preheated furnace.

I think what you're getting at has a lot to do with where the carbides are going, not just grain size. When new carbides form due to heat treat, they can form in the grain boundaries, weakening the steel. So I expect a heat treat that keeps new carbide development to a minimum might be a good goal, I don't know. I'm looking for input on how to achieve that.

Mete, can you clarify your last sentence? I'm pretty sure that's a bad thing, which is how you meant it, but anybody reading this not understanding that carbide in the boundaries is not good might walk away from this thinking they just learned a cool new way to heat treat to get lots of carbide in their grain boundaries...

I can see it now "our new mall ninja streat fighter has been xengineered for maximum grain size and boundary carbides for superior stiffness"...
 
Carbide in the grain boundaries = BRITTLE ! Slow cool from the hardening temperature gives lots of time for the carbides to diffuse to the grain boundaries. The hypereutectiod steels like O-1 and W-2 have a tendency to do this - don't encourage it ! The carbides are going to form in any case but we have to try to get them to be distributed evenly ...Once grain boundary carbides are formed a bit of work has to be done to remove them. ..The other cause of brittleness we see commented on here is caused by over heating [large grains] especially the edge or tip often seen when HTing with a torch.
 
So, assuming I'm working with W-2, to refine grain once I'm finished working with the steel, should I normalize and then quench before doing the actual hardening?


I'd normalize and let cool, then normalize and quench, normalize and quench and then go to critical, keep it there to soak a little, and then quench for hardening.
Is this ok?

Working with C70 (similar to 1075) this shouldn't be necessary, as it's not an hypereutectoid steel?
 
Alarion said:
So, assuming I'm working with W-2, to refine grain once I'm finished working with the steel, should I normalize and then quench before doing the actual hardening?


I'd normalize and let cool, then normalize and quench, normalize and quench and then go to critical, keep it there to soak a little, and then quench for hardening.
Is this ok?

Working with C70 (similar to 1075) this shouldn't be necessary, as it's not an hypereutectoid steel?
Click to expand...


Very good, specific question. Since the folks ought to be back from Ashokan, perhaps someone could answer it?
 
If you are forging - always - normalize after forging. This makes the structure the same throughout the piece. The term normalize implies heating above critical and air cooling. BTW ,W-2 has vanadium which gives a finer grain than W-1. The normalizing cycles reduce grain size .You could also quench in these cycles to produce martensite....This is all done to refine grain and is independant of the type of steel.1075 is nice to work with as you would normally not have a problem of grain boundary carbides as you do in hypereutectiod steels.
 
When working with W2, I forge it, normalize once, anneal, then perform all the grinding, drilling and the like. I then normalize, three times, letting it air cool between cycles. I have had good results when hardening long thin blades using this cycling technique. Don Hanson passed this on to me.

Fred
 
mete said:
If you are forging - always - normalize after forging. This makes the structure the same throughout the piece. The term normalize implies heating above critical and air cooling. BTW ,W-2 has vanadium which gives a finer grain than W-1. The normalizing cycles reduce grain size .You could also quench in these cycles to produce martensite....This is all done to refine grain and is independant of the type of steel.1075 is nice to work with as you would normally not have a problem of grain boundary carbides as you do in hypereutectiod steels.
Click to expand...


Thanks again Mete. That's what I wanted to know.
:thumbup:
 
You must log in or register to reply here.
Back
Top