Grain size in "annealed" steel

M

Makermook

Joined
Feb 28, 2011
Messages
131
I'm getting deeper into the science of bladesmithing, and I'm well aware that the high heats involved in forging lead to grain growth, which needs to be addressed prior to heat treating. But I also make some knives by stock removal, using primarily O1 from Flatground.com (if that is important...)

In my stock removal work, the temps obviously don't get that high, but I wonder about the condition of the steel when it comes to me as "annealed". Does that mean that the steel, after being fabricated, rolled out and ground to dimension, has grain structure that is as fine as it gets? Or is it still worthwhile to go through normalization and my own annealing to further refine the grain size?

This is something that never occurred to me before I started forging. Any insight? And thanks in advance.

Mark
 
From what I know about metallurgy, annealing basically relieves the stresses in the microstructure. It is accomplished by heating the steal above the austenizing temperature then SLOWLY cooling it. The "slowly" is important because the quenching process is really how you control grain size. In the case of annealing, since the heated steal is cooled slowly, the grains will be large, resulting in a very soft, ductile steel.

Here is a good explanation of annealing from Wiki: There are three stages in the annealing process, with the first being the recovery phase, which results in softening of the metal through removal of crystal defects (the primary type of which is the linear defect called a dislocation) and the internal stresses which they cause. Recovery phase covers all annealing phenomena that occur before the appearance of new strain-free grains. The second phase is recrystallization, where new strain-free grains nucleate and grow to replace those deformed by internal stresses. If annealing is allowed to continue once recrystallization has been completed, grain growth will occur, in which the microstructure starts to coarsen and may cause the metal to have less than satisfactory mechanical properties.
 
Industrial practice for carbon steels will normally supply fine grained, annealed and spheroidized material. It will be in much better shape for hardening than we smiths can usually achieve with our facilities.

John
 
What my friend John said. :thumbup:

But there are a few misconceptions here, one is that forging temps lead to grain growth. Such temps (at least proper forging temperatures) certainly could lead to grain growth if things were static, but proper forging is actually a balance of temperature and a rate of deformation that introduces enough strain energy to continually nucleate fresh new grains. This process is often referred to as "dynamic recrystallization".

Another bit of confusion seems to be about rate of cooling and grain size. Grain size is outlined by the austenite grains themselves, so pearlite, martensite and other products produced by cooling will occur within the framework of the prior austenite grain boundaries. The size of the austenite grains is determined by the maximum temperature not necessarily by the rate of cooling. So if you were to heat say to 1450F and then cool extremely slowly the grain size would be no different. The lamellar spacing within the pearlite would be wider though. Rate of cooling does not change the size of the grain that the pearlite forms from, but it does affect how coarse or fine that pearlite will be. It is all bout maximum temp, just as some mistakenly believe that soaking, or reasonable times at temperature cause grain growth, and this is not so, as long as you have control of the temperature.

Spheroidizing is another form of annealing that is entirely sub-critical and thus has no effect on grain size at all, since it is accomplished below recrystallization temperature no new austenite grains are formed that could be larger than the last ones you made.

The quenching process is not the best place to control grain size, since if your grain isn't all taken care of at this point, the normalizing wasn't what is should have been and problems that should have been taken care of then are now being handled a little late in the game. It is possible, and even preferable, to set up your grain size and carbide condition before the anneal, use a sub critical anneal so no big changes occur other than softening the steel and then nailing the hardening to get very good properties with much less hassle.
 
John White said:
Industrial practice for carbon steels will normally supply fine grained, annealed and spheroidized material. It will be in much better shape for hardening than we smiths can usually achieve with our facilities.

John
Click to expand...

Gotcha. Thanks.
 
Kevin R. Cashen said:
What my friend John said. :thumbup:

But there are a few misconceptions here, one is that forging temps lead to grain growth. Such temps (at least proper forging temperatures) certainly could lead to grain growth if things were static, but proper forging is actually a balance of temperature and a rate of deformation that introduces enough strain energy to continually nucleate fresh new grains. This process is often referred to as "dynamic recrystallization".

Another bit of confusion seems to be about rate of cooling and grain size. Grain size is outlined by the austenite grains themselves, so pearlite, martensite and other products produced by cooling will occur within the framework of the prior austenite grain boundaries. The size of the austenite grains is determined by the maximum temperature not necessarily by the rate of cooling. So if you were to heat say to 1450F and then cool extremely slowly the grain size would be no different. The lamellar spacing within the pearlite would be wider though. Rate of cooling does not change the size of the grain that the pearlite forms from, but it does affect how coarse or fine that pearlite will be. It is all bout maximum temp, just as some mistakenly believe that soaking, or reasonable times at temperature cause grain growth, and this is not so, as long as you have control of the temperature.

Spheroidizing is another form of annealing that is entirely sub-critical and thus has no effect on grain size at all, since it is accomplished below recrystallization temperature no new austenite grains are formed that could be larger than the last ones you made.

The quenching process is not the best place to control grain size, since if your grain isn't all taken care of at this point, the normalizing wasn't what is should have been and problems that should have been taken care of then are now being handled a little late in the game. It is possible, and even preferable, to set up your grain size and carbide condition before the anneal, use a sub critical anneal so no big changes occur other than softening the steel and then nailing the hardening to get very good properties with much less hassle.
Click to expand...

As I understand it, it is the normalization cycles that help reduce grain size, not necessarily the full annealing. Heat to just above critical (not all the way up to forging temps, just high enough for austenization to happen), allow to air cool, repeat two or three more times. The austenization followed by very SLOW cooling would anneal the steel and make it as "soft" as possible, but contribute no more to grain size reduction than a simple normalizing cycle.

Do I have it about right?
 
I found this while doing some research for grad school classes. Also as a Marine, i am always looking for the best possible knife steel.
the metal is called Ytterbium. It is used to improve the grain refinement, strength, and other mechanical properties of stainless steel. Some ytterbium alloys have been used in dentistry. Essentially, i am assuming "mechanical" properties mean that it allows for greater ability to resists breakage via bending and other similar type stresses. Anyone heard of this before? If so please chime in.
 
Hm. im thinking Ytterbium might be too rare to acquire since it has relatively few uses and its consideration for use in portable Xray equipment. I will do some more digging and post my findings here folks.
 
You must log in or register to reply here.
Back
Top