What's Inside?

[Lin Rhea]

As you know, I and Karl, and some others have been preachin about grain growth and grain reduction. I want to show an illustration of the way steel changes due to the manner in which it's heated and cooled.

Saturday I and a bunch of others went to Mountain View, Arkansas to a shop vist at Jim Crowell's. Some of you know him and know that he's been a Master Bladesmith for a number of years. I was doing a forging demo and some questiones were raised about grain reduction etc and we decided now was a good time to get a look at the inside of the steel.

So, I took a bar of knife steel, in this case 1084, and forged a point on both ends to the same geometry. I just taperd and thinned them a little to roughly immitate a blades taper.

I heated one up and thermocylcled it just like I would a blade, then quenched it. I wrapped a rag around it and hanging about an inch over the anvil, I broke it.

Next I heated the other pointed end and talked a little and more or less immitated the careless approach. The blade got a little too hot for normal circumstances, although not sparkling or in any way ruined. I then quenched it without thermocylcling. Then I broke it.

Here is the visible difference.

   

[DANA HOLMAN]

Thanks Lin
that is good info. Question I have is, is thermocylcling the same as normalizing? The way I read different post from other people it is the same, also after tempering the blade, do I need to draw back the spine? I thought the tempering would draw it back enough, maybe I've been doing it wrong, if I have not drawed the spine back what would the steel look like?
Thanks for any help
Dana

[Scott Roush]

nice demo there Lin...

[kbaknife]

Thanks a bunch, Lin.
Nice demo and one to take to heart!!
It shouldn't take very much imagination to understand that the large grain piece would have minimal strength, as compared to the tightly grained piece.

Dana, when "thermal cycling" to reduce grain size, it is often done three or so times, gradually reducing the heat each time, and only cooling back to a black heat before re-introducing the steel to the forge for the subsequent heat.
Say you do it three times.
The first time you might go as high as your forging temps.
The second time, not nearly so high.
The third time, just above non-mag, but not as high as the second time.
These steps are what are referred to as "reducing heats".
That way you don't neutralize what you gained on the previous step.
Remember, let cool between heats only to a black heat, where the steel has lost its color, before putting back in the forge for the subsequent heat. So you're only cooling to about 900 degrees or so.
"Normalizing" would be to let cool all the way back to room temp.

There would be no need to draw back the spine unless it was hardened. Right?
And, you may not want a soft spine.
It all depends on what you want that blade to do.

Didn't mean to take over, Lin, I was just on break from knife assembly.    

(I might also mention that maybe the way I do it is not exactly the same way Lin or others do it.
There are often more than one way to get to the same point. Remember that we deal with so many different types of forges/heat sources, steels, quenchants, training, attitudes, etc. That's one of the reasons why so many different recipes of knife making exist, and it's so easy to get conflicting information.
But one thing that does not differ, is grain growth, or the need to reduce it.
It is also, by Lin's wonderful example here, a variable in your own knife making that is easily demonstrable.)

[DANA HOLMAN]

Thanks Karl

[gudspelr]

Quick question from some of the stuff mentioned...  If the thermal cycling is done as explained by Karl, is there a need to normalize?  In my really basic understanding of this stuff, normalizing did what I'm hearing thermal cycling does.  What changes by allowing the steel to cool all the way vs. just getting to black heat?  Thanks for any help and thanks Lin for putting this into perspective with that picture.

Jeremy

[Ragnarok Forge]

Good question Jeremy.  I normalize three times before the hardening quench.  When I break a blade I get very fine grain in the steel.

[Montauks]

great info, great read... thanks guys

[amar911]

A picture paints a thousand words, but the words sure helped explain Lin's pictures. Thanks to everyone who contributed to this thread.

Allan

[kbaknife]

I am going out a little ways on a limb here and am going to quote my friend, Kevin Cashen, who has helped countless blade smiths with his educational expertise on metallurgy.
One of the problems often associated with "normalizing" and "thermal cycling" is simply a matter or word usage.
Thermal cycling a few times before quenching to harden is NOT normalizing, if only going back to where the magnet sticks again. I do the same thing frequently, and it is just going through a cycling of heat, i.e. "thermal cycling".
 I personally perform the final step as dictated here by Kevin, in that I quench my steel for full martensite on my last thermal cycling heat, and then do a sub-critical spherodizing cycle in my heat treating oven so as not to disturb any of the grain refinement I just accomplished.
This way, my steel never goes above non-magetic again until ready to harden, yet is still quite machineable.
This should be done on any steel above .8% Carbon.

Kevin Cashen:

"Industry, for the most part, has one set of definitions for “normalizing” while there can be as many definitions among bladesmiths as there are bladesmiths. The idea behind normalizing, as has been stated is to evenly redistribute carbon and structures and resulting stresses in the steel. Grain refinement is often more a reference to homogenous grain size, whatever that size may be, and for good reason since uniformity can often be more important than actual size. True normalizing requires heats high enough to put the material in full solution and thus is often in the 1600F – 1700F range. Air cooling is important for actual normalizing in order to keep things into solution evenly on cooling, and to be certain things cool uniformly. So to actually normalize blades the first heat should be well above critical, but as mentioned never “yellow” or “white” hot.

Bladesmiths then follow this heat which equalized things with a heats to refine that condition further.

  These heats are almost universally referred to as “normalizing” by bladesmiths but should actually be called thermal cycles to be more accurate.
 
 The next heat will be to make smaller grains from the evenly sized ones and is done just to critical and then air cooled. Some will quench at this point instead, I often do, to increase the grain refinement, a quench will replace around two air coolings with careful heating.

For any of these refinements, you really only need to air cool to Ar1 (when the magnet sticks again) before reheating, but if you quench you will need to go to room temperature in order to see gains, otherwise you will only be reheating the same internal condition (austenite) again.

Often smiths finish up on a dull red final heat. This can create subgrains, like grain seeds that will sprout into new, finer grains on the next heating. But if one is careful not to allow the blade to go nonmagnetic this dull red heating can also substitute for an anneal, and is preferable for steel with more than .8% carbon.

Annealing is for the expressed purpose of softening the steel and can be a full (lamellar) anneal, where you heat to critical and then slow cool (wood ash, vermiculite oven etc…). Or you can got with a sub-critical anneal which takes us back to keeping the heat below nonmagnetic and softening the steel that way, this is called spheroidizing. Spheroidizing is highly recommended for anything over .8% carbon and also helps keep everything you did in the normalizing, while full annealing tends to undo some of it."

[Lin Rhea]

I  suppose in this, as in other complicate areas, I have to simplify things so as not to bog myself down in the technical stuff. On some level, I know there is sience envolved, but I work much better on a more basic, some would say simple, level.

When I can actually see the results, as in the picture above, I know all the technical explanations are true and I believe and follow it. Following it is the important part, in my opinion.

Some of the old masters did not have the sientific data at their dispoal, they just knew what worked and the better ones FOLLOWED RELIGEOUSLY the proven methods. In fact some cultures tied religion into the craft giving credit to the gods when methods succeeded. Through the process of elimination, they discovered what pleased the "gods". The basic results were that the craft, complete with methods, was preserved.

[gudspelr]

Thanks for the explanations-makes sense.


Jeremy