Zach,

Let's take this by stages.

First, forging. Industry standards for 1084 ( and 15n20/ 1084 by extension) needs to be forged between 1500 and 2150. Do not forge below 1500. The idea is to move the steel quickly at the high temps, thus lessening the time for de-carb, and refining the forging( removing hammer tracks, etc) at lessor temps, where the steel doesn't dent as easily.

After forging, the 1st normalization is done at 1600 ( 1500's OK,too) with an air cool to ambient, to relieve stress, and begin grain refinement.

Soaking at 1600 for 6-10 minutes before air cooling will allow the carbon to diffuse evenly throughout the steel. At this temperature and duration, significant grain growth, scaling, and decarb are not a problem.

Subsequent grain refinement is the next step. Grain refinement, by subsequent normalization cycles (thermal cycles) are usually done in three steps, say 1500, air cool to "black heat," ( appx 800), 1450, cool to black heat, 1425, air cool to ambient. Three thermal cycle normalizations will give adequately small grain. As has been noted, grain refinement can be taken to extremes. Smaller grain in simple carbon steels like 1084 makes the steel shallower hardening.

At this point, the steel is stress relieved, and has small grain, with well distributed carbon,and is in the form of pearlite.

It is now that you want to anneal, and the process you are following, 1475, cool at 50 degrees per hour to 900, will spheroidize the steel, as well as leave it in it's softest state.

Now you grind. If you feel that you might have re-stressed the blade while grinding, you might want to do one stress relief cycle at 1150-1175, and air cool. (Thermal cycles at below critical do not change phases, and do not create new grain, but merely relieve any added stress.)

To austenize for hardening, take to 1500 and soak for 8-10 minutes, and quench in fast oil.

The soak is to let the carbon spheres dissolve, and evenly distribute in solution. It seems counter-intuitive, but nicely spheroidized carbon takes a bit longer to diffuse.

At 1500, for the length of time you're holding, if you are working in a reducing atmosphere, as in a Fogg-type propane forge, there will be no significant scaling, de-carb,( or grain growth.) If you are in an electric oven, scaling can be adequately reduced by argon injection, or use of an anti-scaling compound such as Brownell's ATP-641.

To review,

1)Forge

2)Normalize

a) for stress relief

b)for grain further grain refinement

3)anneal ( and spheroidize)

4)grind

5)stress relieve(optional)

6)austenize and harden

7)temper

Considering what you're doing at each stage will show why the sequence followed is appropriate.

Caution: This is my current, somewhat practical understanding of the processes and effects. I'm hoping someone more knowledgable will also post up, and hopefully correct my (mis)understandings.

John

John,

Thanks for the detailed response.

In fact I wrote my sequence out incorrectly. I did normalize, then stress relieve, THEN anneal.

I'm surprised about the recommended temperature range for forging. Where do you get this kind of information, originally?

THANKS!

Zack

Zach,

You're welcome.

The area where I'm least certain in this discussion is the question of how much actual spheroidizing is achieved in 1084 by the annealing process or by my short-cut of heating to 1200 for an hour and air cooling. I note that drilling with small bits is much improved.

As to forging ranges, I probably first encountered specific temp ranges in one of Kevin Cashens discussions. Check out his excellent heat-treating sections with links to common forging steels on his website,www.cashenblades.com

Here's a very short discussion with carbon content chart also,here

In any case, to refer back specifically to your original question, soaking times at various working temps, significant grain growth or scaling do not occur at the lengths of time we're working with, say 5-15 minutes, until around 1700 roughly.

John

John and Zach, the ASM Heat Treater’s Guide states for 1084 “Forging: Heat to 2150F (1175C). Do not forge below 1500F (815C)” I also have spec sheets on individual runs of steel that coincide with the ASM numbers. Add to this my personal experience on the effects, or lack thereof, and I am confident that at least the initial parts of the forging process should be in this range. One clear benefit, which is not as critical with damascus since it has already see plenty of higher temperature cycles, is the homogenizing effect that the high temperature diffusion and cycling will have upon the carbide segregation and banding that can be present in these 10XX series steels. The idea behind forging is to shape the steel as much as possible in the shortest time at heat.

But since we are talking about damascus here, it is worth noting that the steel has already seen numerous cycles at temperatures much higher than 1500F or even 2150F, if this was a problem that normalizing couldn’t fix then all damascus must be pretty bad stuff due to the very nature of its creation. And yet I can assure folks that well made damascus is every bit as good as either of its parent steels.

There is a lot more concern about grain and overheating in forging than there needs to be. One must remember that there are three distinct sets of grains in every recrystallization heat cycle, there are the initial austenite grains, then there is the phase formed in cooling, followed by an entirely new set of austenite grains on reheat. This leaves more than enough potential to redo, or undo anything that has been done in just a couple of cycles. So if you managed to get teeny tiny grains in forging just one good normalizing heat will reset the whole game, and vice versa, if you have large grains from forging, just one good heat will reset things to a much better situation. Moreover plastic deformation (forging) is a very inhomogeneous way of affecting the internal structures of the steel when compared to the very thorough and even effect of heat. So it is really much better just to get the job done and avoid the decarb, scaling and other hassles and then handle the internals of the steel with thermal treatments.

A more interesting examination is to search for the source data for the low temperature forging numbers, as the only place I have found this concept is among bladesmiths or the anecdotal examples of mysterious old metalworkers from elsewhere given by bladesmiths. There are folks who still believe in edge packing, I also have a horse shoe hanging next to my shop door, I just don’t look to it much for answers about my steel or help in forging.

On the sequence, I am glad to see the correction on the normalizing/cycling/annealing. Normalizing is a very powerful thermal treatment, perhaps the most important, and anything done before it will pretty much be undone, if it is done correctly. Annealing is the last thing one does before grinding or other machining since it sets things up for those operations. Stress relieving is better used after the heavy machining or grinding to prepare the blade for a trouble free hardening. It is a much more subtle treatment aimed specifically at strain energy and not much else, so it would be rather redundant and in consequential immediately after the anneal.

John, Kevin. Thanks very much to both of you. My understanding is more complete. Kevin, your last post in particular confirmed a number of suspicions I have long held about thermal treatments, sequence, etc. Much obliged.

There's something that John said which has pointed out another question to me:

"If you feel that you might have re-stressed the blade while grinding, you might want to do one stress relief cycle..."

How does one evaluate this? I mean, it seems like the safe thing to do would be to give it a stress relief cycle anyway--no harm done, right? But how does one evaluate whether or not the blade has been stressed during grinding, and what is the nature of that stress?

Thanks,

Zack

Zack,

Certainly no harm done with a stress relief after grinding,( except for lost time and and expense! )

One possible stress cause related to grinding would be over-heating during grinding. Prevention would include grinding with bare hands so as to monitor heat, grinding with sharp belts, and frequent water cooling.

A second source of "stress" would be grinding from a mis-shapen forging, where much more has to be ground from one side. Prevention would be better forging.

In each case, the purpose of specific "heat-treatment" steps refers back to what you're trying to "repair" or what you're trying to prepare the steel to do next.

I'm much happier working the steel, now that I'm assembling a mental picture of what's going on internally at each step. Much more to learn, though,

John

Ah, ok that sounds about right. Thanks.

Yeah, I forge as evenly as possible, down to matching the number of blows per heat on each side of the blade. Generally speaking, that leaves me with very even grinding. Also, I have always ground with bare hands, and I never overheat the steel.

I couldn't agree more about assembling a mental picture. Even if it doesn't change the specific steps I'm taking, I always feel much more secure when I have a conceptual understanding of WHY I'm doing each step.

Cheers,

Zack

|quoted:

…One possible stress cause related to grinding would be over-heating during grinding. Prevention would include grinding with bare hands so as to monitor heat, grinding with sharp belts, and frequent water cooling…

There is no “thumbs up” emoticon for this forum, or I would give this two of them John. I am adamant with my students that we heat treat at the forge- NOT ON THE GRINDER! It is a poor practice to turn your steel blue on the grinder, it is bad for the steel, it is bad for your belts and it is bad for you! Even if there were no effects on the blade or belts, handling a 500F+ piece of steel while trying to do something as tricky as a good grind line is not only self-defeating, it is dangerous.

It has been very difficult to convince many makers of this being a problem because they can’t readily see the effects, and some well-meaning, but misguided, smiths have insisted there is no effect since the blade has yet to be heat treated. One cannot see strain energy, not even with a microscope (the effects perhaps, but not the stored energy). During the normalizing and annealing operations one puts so much effort into relieving as much stress/strain as you can and to keep what is left as perfectly balanced as possible. After all of that effort why do we want to lay into the steel on a grinder in such a heavy handed way that we effectively undo all we have done in that area?

Sharp belts cut more than they deform but as they dull the opposite is true, and that deformation is no different than cold working one side of the blade. When viewed under a microscope, you would be surprised at the amount of pulling and smearing that occurs at the bottom of a grind scratch.

Anybody who doubts the effects of overheating by laying into a grind heavier than necessary obviously has never ran a surface grinder. I have scrapped pieces in my learning curve on the surface grinder by warping them beyond repair by making colors on the steel; just think of the amount of out of balance stored energy required to do that, and it was nowhere near as purple or blue as what I have seen many make their blade on a grinder.