How do you forge a katana WS

HOME

1 of 5

ABOUT WALTER

http://www.waltersorrells.com/blades/katana%20forging.htm

HOME

ABOUT WALTER

HOW THEY'RE MADE

NEW & AVAILABLE

BLADESMIITHING VIDEOS

GALLERY

PHILOSOPHY

LEARN SOME STUFF

HOW TO ORDER

How Do You Forge A Katana? Let’s start with definitions. Forging is a process by which steel is shaped by being heated until it gets soft and then hammered to shape. The katana is the traditional Japanese long blade, with a cutting edge of 24 inches or more. It has a single edge and an obvious curvature. However, perhaps the single most distinctive feature of the Japanese blade is metallurgical: the blade is selectively hardened so that the spine is springy and resilient, while the edge is harder and less easily dull ed. Traditional Japanese smiths make their own steel through a forge-welding process which is outside the scope of this article. For the purposes of this demonstration, I'll just start with a bar of modern steel -- in this case, 1050 high carbon steel. (I'll do another article soon about forge-welding, in which I'll demonstrate how various welding schemes including high-l ayer forge-folded steel, san-mai, cable damascus, etc. But, as I say, that's outside the scope of this article.) Typically I’ll start with a bar which is 1 inch wide, 3/8 of an inch thick and 30 inc hes long. Because the forging process draws out the steel, this is plenty of metal to form a katana of approximately 36 inches total length (27 inches of blade and 9 inches of tang). (Just as a side note, the illustrations for this article will show several different swords in different stages of the forging process, so don’t be surprised if the sword sudden ly seems to take on a different shape as it makes it’s way through the process.) I begin by thinning the stock down, making a preform from which the final sword will be forged. The Japanese refer to this preform as a "sunobe." The preform will continue to be rectangular in cross-section, but it will be tapered both in width and depth. It’s possible to forge straight from bar to blade. But it’s a bad idea. There’s too much steel to move in order to build in the proper taper and you’ll either end up with a twisty blade or one with insufficient taper. A Japanese-style blade tapers both toward the tip of the blade and toward the end of the tang. The junctu re between tang and blade (referred to in Japanese as the "machi") is the thickest part of blade.

12/15/2008 10:38 PM

HOME

2 of 5

ABOUT WALTER

http://www.waltersorrells.com/blades/katana%20forging.htm

Hammering the sunobe or preform to shape by hand takes me about two hours. Using my hydraul ic forge press cuts that down a little. But at a certain point I have to turn off the press and forge by hand to get the shape I need. I begin by heating the end of the b ar which will form the tip and then hammering vigorously to thin it out. I move slowly u p the blade (usually making several passes, each of which is more precise and less vigorous) until I have the dimensions I’m looking for. Every sword is different, but generally the preform will end up just under 1 by 3/8 inches at the machi, tapering to about 5/8 by 1/4 inches at the tip. It’s very important that the preform be forged carefully. I have to have a very clear idea in my mind of what the final blade will look like...and how much steel will be required to get me there. If there are any significant irregularities, it will be difficult – if not impossible – to form the final blade shape. The most serious flaws are overly deep hammer dings and overly narrow spots. Unlike twists and bends, they can't be fixed. Both of these flaws are usually caused by getting impatient and hammering too aggressively. Forging the sunobe usually draws the bar out by four to six inches. What was previous ly a 30 inch bar is now close to three feet long. The final forging process will length en the blade more. With a katana, it will add about three more inches to the blade and a couple of inches to the tang. Therefore if I want a 27 inch blade, I’ll forge the preform from the machi to the kissaki (tip) to a length of about 24 inches, leaving an additional five or six inches of steel for the tang. Once the sunobe is complete, I begin forging the blade to its final dimensions. First I forge the tip. Then I begin forging the bevels, starting with the main bevel. I work four t o six inches at a time, forging from the tip up to the machi. If the blade is a single beve l (hira zukuri) type, then I just get the bevels established and move on. If the blade is a d ouble beveled (shinogi zukuri) type, then I begin with the main bevels, then work the secondary bevels. Once I’ve got my four-to-six inch section properly shaped, I’ll begin to work my way on up the blade. (There is also a third very minor bevel on the spine of the blade. I forge this, too, but not with a great deal of precision. It’s so small that it’s barely within the fairly broad tolerances of the forging operation.) Shown here is the forged blank. The overall shape of the blade has been established and the blade is still covered with black fire scale from the forging. Notice that it is more or less straight. The characteristic curve of the katana will be formed during heat treating.

Crucial to proper forging is working both sides equally. I forge one side, reheat, th en forge the other, reheat and move back to the first side again. If you don’t stick pretty scrupulously to this back and forth forging method, you’ll cause the blade to corkscrew. Additionally, you’ll forge stresses into the blade which will cause a variety of problems later. Because the metal always moves away from the hammer, the blade also bows significantly as forging progresses. It’s necessary to continually make corrections, straightening and restraightening the blade as you go. Once all the bevels for the blade are formed, I’ll usually hot-cut the tang off of wh atever remains of the bar stock. Then I’ll let the blade cool for a while. Once it’s cool en ough for me to hold (at least with my Kevlar gloves) I’ll flip it around so I’m holding it by the tip and forge out the tang. At this point I’ll carefully check for places that are too thick, as well as for curves, twists and kinks, cork-screwing, asymmetry in the tip and so on. I’ll correct everything as best I can, making the blade as flat and straight as possible.

12/15/2008 10:38 PM

HOME

3 of 5

ABOUT WALTER

http://www.waltersorrells.com/blades/katana%20forging.htm

One absolute rule of blade-making is that the better a job you do at each stage, the less additional work you have to do later. Cutting corners always wastes time later on. This is especially true at the correction phase. The closer you can get the forged blade to its final dimensions, the less grinding, filing, scraping, etc. that you'll have to do later. This second part of the forging process takes another couple of hours, give or take. Once the forging is complete, I move on to heat treating the blade. A brief digression: a lot of people don’t realize it, but most steel is capable of forming into any one of several different crystalline structures, each of which has significantly different levels of hardness, toughn ess and flexibility. Changing from any given crystalline structure to another takes a lot of heat. The purpose of heat treating is to optimize the steel’s microstructure for whatever tool its going to be made into. Knives must be hard and s harp. Springs must be flexible and resilient. Swords – because they are long and because t hey whack into things with great force – must be tough, even at some expense to edge-holding ability. The first heat treating process I’ll go through is called normalization. Normalization is intended to reduce the grain size of the steel. The tighter the grain of the steel, t he tougher it is and the better it holds an edge. To normalize 1050 steel, I heat it to around 1600 degrees, then let it air cool. I repeat this process three times. Once the normalization cycle is finished, I take the rough-forged blade to my belt grinder and knock off all the scale. In theory you could use the belt grinder to grind the bl ade to its final shape. But in practice, it’s really easy to muck things up with a grinder. So once the scale is gone, I'll true up the blade using a metal file. At this point I ’ll be leaving the edge itself about a tenth of an inch thick because the stress of hardening is quite violent. A thin edge is more likely to crack. Hand filing is extremely laborious (it generally takes a good deal longer than forging the blade itself). But doing it by hand allows you to even out ripples and other forging imperfections much more accurately you can with a grinder. Once I’m happy with the lines of the blade, I’ll move to the most exciting portion of the sword making enterprise: hardening. Another digression: The traditional Japanese blade is selectively hardened. About ten centuries ago, Japanese smiths developed an ingenious metallurgical process (used no place else on the planet) which gave Japanese swords their unique combination of strength and cutting ability. In modern metallurgical terms, we say that the edge of the Japanese-style blade is composed of a very hard (but relatively brittle) crystalline structure called martensite, while the spine is composed of softer but tougher structures calle d pearlite and ferrite. Martensite is formed in carbon steels by heating the steel to what is known as its "critical point" – a temperature at which the steel begins to assume a more plastic structure called austenite. If allowed to cool slowly, the austenite will precipitate into pearlite again. However, when high carbon steel is quenched in a cooling medium such as water, the st eel forms a martensitic (hard) structure. It is this peculiarity which is exploited by the Japanese smiths to form the differentially hardened blade. What follows is a variant of the ancient technique of "clay coating" which was developed centuries ago in Japan. First I coat the spine of the blade with clay. I leave the edge uncoated. I then heat the blade to critical temperature (around 1500 degrees Fahrenhe it for 1050 steel) and then quench the blade in water. What happens is that the clay on the spine of the blade acts as a heat sink so that the spine cools much more slowly than does the edge – too slowly to cause it to harden. In metallurgical terms: the edge steel converts from austenite to martensite while the spine converts to pearlite and ferrite. Bottom line: hard edge, soft back. When properly polished, the transition between the hard and soft steel is visible to the

12/15/2008 10:38 PM

HOME

4 of 5

ABOUT WALTER

http://www.waltersorrells.com/blades/katana%20forging.htm

naked eye. Depending on how the clay is placed on the blade, then, different designs can be made in the steel. This feature of the Japanese blade is known as the "hamon." Shown here is a blade covered with clay, ready for heat treating. This particular cl ay design will form a straight or "suguha" hamon with small lines called "ashi" extending into the hardened area of the blade.

Great care has to be taken in heating the blade before the quench. Heat the blade too much and it will likely crack during the quench . Heat the blade too little and it won’t harden at all. I run the blade back and forth through my forge at a high heat, slowly bringing the steel up to temperature. One of the peculiarities of steel is that it stops being magnetic at about critical temperature. So I frequentl y touch the steel with a magnet to see if it’s gotten non-magnetic – an indicator that it's ready to be quench ed . If you don’t keep your eyes peeled, however, you’ll over-heat portions of the steel. In fact, I only do quenching at night, when my eye can tell me which parts of the blade are too hot or too cold. One of the fascinating aspects of the quenching process is that in the four seconds that it takes to harden, the blade takes on a radical curve. It goes into the water straight and comes out with the characteristic curve of the Japanese sword. This blade is fresh out of the heat treating process. As you can see it went in more or less straight and came out with a curve of over an inch. Magic!

After quenching, I heat the blade to about 400 degrees F by immersing it in a hot oil bath. This is called tempering. (People frequently confuse hardening with tempering. Tempering is not the hardening process. Tempering actually softens the steel slightly and relieves stresses in the steel. In so doing it makes the steel far less brittle and therefore less likely to crack during use. I do three separate tempering cycles of an hour apiece. Failure to perform any one of these heat treating operations correctly will compromise the blade – usually fatally. It will either dull quickly or crack under use. Either is un acceptable. Once the blade has made it through tempering, the blade is shaped to its final dimens ions. I’ll use a belt grinder to thin out the oversize edge. But once that's done, I perfor m all the final profiling with water stones. This is a laborious (and, frankly, pretty damn dul l) process. Like filing the blade prior to heat-treating, it’s also a very demanding process. One moment of inattention and you can mess up the geometry of the blade significantly. It won’t usually cause fatal problems – but if you don’t pay attention, a single slip of the wrist could easily cause an hour of extra work. Once I’ve established the profile of the blade on the 180 grit Japanese water stone, I’ll move to progressively higher grit stones. At a certain point – and it depends on my mood, the type of sword, the phase of the moon, and so on – I’ll change to wet-or-dry sandpaper. This blade is shown after being worked on the first stone. The hamon is already visible. (It's that dark, smudgy part along the edge.)

12/15/2008 10:38 PM

HOME

5 of 5

ABOUT WALTER

http://www.waltersorrells.com/blades/katana%20forging.htm

At some point you move from profiling to polishing – though exactly where the former ends and the latter begins is somewhat nebulous. The point is that eventually I have the geometry of the blade finished and I’m focusing simply on refining the surface appear ance of the steel. Once this happens, you move from substance to cosmetics. If I wanted a sword purely for cutting purposes with no cosmetic considerations whatsoever, I could stop here. It's already wicked sharp. The main thing I’m trying to do in the polishing phase is to reveal the hamon in all its detail. I do this by using progressively finer and finer grades of sandpaper (quitting at around 2000 grit). A proper heat treatment and a proper polish will result in a clearly visible h amon. After I reach the final grit, I etch the blade's surface lightly with an extremely mild acid. After several polishing and etching cycles the hamon really pops out and become visible. I'll then do a final polish on the blade. The sequence of the etching/polishing cycles an d the type of abrasives used change from blade to blade. I use wide variety of commercial metal polishes like Pikal and Flitz as well as extremely fine silicon carbide and aluminum oxide abrasives. In some cases I'll also rub on a substance called nugu i -- iron oxide (Fe 2O3) suspend ed in oil -- which acts to heighten visual contrasts in the steel. The final stages of polishing change from blade to blade depending on the steel, the hamon and various ot her factors. And that’s it. All told, somewhere in the neighborhood of forty to sixty hours of work. Here's the final product.

If you're interested in purchasing an existing blade from the gallery or ordering a custom piece, click the ORDER INFORMATION button. If this article leaves you hung ering for more detailed information about making Japanese style blades, check our BLADESMITHING VIDEOS page.

12/15/2008 10:38 PM