Forge Build Pt IV


Building a simple forge torch

Disclaimer: building a torch is DANGEROUS. I am not an authority on combustible gas and am not responsible for any errors or accidents. This is simply to document my build, for my burner, and not to be taken as an instructional guide. If you wish to undertake a torch build of your own, do your research, take precautions, check fittings, and understand the risks involved.

This is actually the second torch I've built, in the same style and model as my previous one, reusing some of the same parts in fact, but upgraded and replaced a few of the worn out components. 

First off let's lay down a few distinctions. A solid fuel (coal/charcoal/coke powered) forge operates largely off the principles of correct air blowing and containment of heat to get the coals hot enough, whereas an efficient propane forge simply needs the most flame contained well enough to heat the workpiece. 

So (among other variables), bigger, hotter flame = bigger, hotter forge. Pretty simple. For this build, I decided not to do a complicated burner build, but simply improve on my current model, which is very easy to make.

Gas torches require a combustible gas under high pressure to be mixed with air, which is then ignited. Under the propane torch umbrella are two main types; blown burners and venturi burners. Blown burners make use of gas as well as air pumped into the torch shaft, so the air, as well as the propane, is forced in. This is more efficient and economical than a venturi burner but is more complicated to build. I'll likely try a blown burner in the future but it's nowhere near a necessity now.

Note: another bladesmith messaged me to say that it's actually entirely possible to convert a venturi torch into a blown burner. I am interested in this but that upgrade will be for another time

The venturi burner's namesake is the venturi effect. The basic idea of the venturi effect is that when gas flows from a wider area of pipe, through a choke, or restricted section of a pipe (like a funnel), the pressure decreases, and velocity increases inside the restricted area. Not only increasing the speed of the gas down the shaft, it also sucks air into the intake. 

So, a venturi burner is a torch that works by shooting a stream of propane down into a cone, into a tighter shaft, and then expelled into the forge where it is burned. Oxygen is sucked in and mixes while in the shaft, where they combust in the forge, producing the precious flame. The end of the torch is also a flare, slowing down the gasflow once more so that it ignites as it is expelled. 

Basic construction is cone, pipe, then cone, with some way to expel propane down the first cone. 

The method I used is known as the Ron Reil style burner (I actually didn't know the name before a follower filled me in, a huge help for more research), what I ended up piecing together after research on various forums and threads. I cherry-picked different principles and this is what I came up with. 

I started with three black iron pieces; two reducing couplings on either end of an 8" X 3/4" nipple pipe, threaded at both ends. I drilled two holes through the intake coupling, 1/2" diameter. 

The propane piping is set up like this: a regulator and hose goes from the propane tank and screwed onto an adaptor, which screws onto a brass nipple, which is inserted perpendicular to the intake coupling and capped on the end. The brass nipple has a #52 hole drilled in the center, this orifice is for ejecting the propane into the torch shaft.  


Note: make sure you clean out the burrs after drilling the orifice in the brass nipple. If you don't, they will clog the hole and inhibit the propane flow. I did this carefully using some needle files and a wire brush. 


The view down the brass nipple pipe: you can see the orifice drilled in the center. Check it inside and out for burrs. 

The individual dimensions of the brass pieces aren't actually all that important: I just went to the local Ace, explained what I needed along the lines of the above, and they put the pieces together for me.

Just a word of warning for those of you who (like me) are not familiar at all with piping: make sure the thread sizes match. When I first built this torch, I went back and forth to the hardware store several times before I realized there wasn't one standard thread size. Who knew, right? 

To summarize, the basic setup is like so: Three black iron componenets: two bell couplers on either end of a 3/4" nipple. One end expells the flame into the forge, the other is the intake and supports the brass piping. 
The propane goes from the tank through a hose, through an adaptor, into a brass nipple piece which is inserted perpendicularly into the intake coupler and capped on the other end. The propane is expelled through a hole in the center of the brass nipple piece, into the black iron shaft, and expelled out the flare at the other end where it is burned. 

Now, where the brass pipe is inserted into the intake coupler, you will notice that there isn't anything holding it still: the pipe will be relatively free to rotate, which you don't want happening. I marked the opposite side from the orifice with a sharpie so that I know when it is oriented correctly. It could be possible to redesign the torch so that the brass pipe is impossible to turn, but I kept it as is, and make sure to double check the orientation before I light the forge. The ability to rotate the pipe also enables me to help fine tune the burning, to get 100% combustion. 


As time goes along I will be making upgrades, tweaks, and repairs to the burner, but this is a very solid, efficient, and working torch that is the perfect size for simple knife forging.

With that, I've completed (for now) the new forge build. The entire series can be found on my site here. If you have any questions or comments, please shoot me an email, I would love to hear them! I am not an expert, and there is loads that I have left to learn.


A very valuable resource that I found only after completing the torch can be found here. I believe it is written by Ron Reil himself, the pioneer of this style of burner. It is very clearly a go-to resource if you plan on building a torch of your own. 

Forge Build Pt III


Forge Build: insulation and clay


Last time we talked about building the main forge body, the frame of the forge. Now it's time to work on the interior. 

The overall principle for the interior is that the end goal is, of course, to contain the heat pumped out by the torch.

Given that, many people build their forge entirely using bricks for the interior. This is a very good way to do it. For small-scale bladesmiths, a circular interior and small size is optimal, as opposed to blacksmiths who need to simply blast a ton of heat onto thick steel. We don't need all that much, but we do need efficiency in a small space. 

Thus, most bladesmiths build a forge using high-temp ceramic wool insulation. That's what I used for this build.

I ordered Kaowool (this one, to be precise) off Amazon, a little more than I needed, for under $50. A friend of mine, Will Freeman, sent me some satanite clay to line the forge (we'll get to that in a minute). 

Ceramic wool is an insulating blanket, it helps keep the heat contained and keeps the frame from overheating. I'm writing this article post-completion, and have used the forge many times. So far, the outside of the forge has not exceeded 400 degrees Fahrenheit, so it does a good job. 

Note: Ceramic wool is made of ceramic fibers, which guess what: NOT HEALTHY for your lungs. Wear a respirator when working it. I suited up with glasses, hoodie, and gloves too; fibers stick in your skin like cactus if you're not careful.

I've only ordered ceramic blanket twice, and both times it came with a simple box cutter to cut it with. This works just fine.

It took a little bit of planning, but I got the wool cut to the right length, rolled it up, and cut a hole in the center to fit the torch flare. Because the blanket is a few inches thick, the tip of the flare and the interior surface of the blanket ended up flush, which is excellent because if the torch extended into the forge, it would overheat too easily and degrade much faster. 

I placed it so the seam ended up at the bottom of the forge. Later on I'm planning on putting some firebricks on the bottom, especially if I intend to use the forge for forge welding: flux from this process dissolves (for lack of a better word) the forge lining. In any case, potential issues with insulation coming apart down the road are far easier if the seam is on the bottom, than on the sides or top where the edges could fall in. 



Once the roll was in place, I trimmed it to be flush with the edges of the forge body. 

The doors were a bit trickier, and took some creative cutting to fit the wool in just right, but I got it eventually. The basic idea was cutting a disc (or a bagel, for the front) for the back, then cutting a strip to line the edge just like the main body of the forge. The main thing is to make sure you have as few separate pieces of blanket as possible,just so there aren't weird gaps between pieces, and so that it's structurally sound. 



Once the inside was completely lined with insulator, it was time to coat it with refractory mortar. Refractory is a type of high-temp clay, and ceramic wool needs to be coated so the fibers don't get in your lungs. 

Now, uncoated wool is a long-term health problem. You won't notice anything immediately if you use an uncoated forge a few times, but the fibers will build up in your lungs.

Coating the wool with refractory both keeps it safe to use, but also keep the interior sturdy and in place. The refractory I used is called Satanite, and can be attained from a few different suppliers. On a side note, satanite is also one of the more common clays used for clay quenching, so it's worth getting a good amount. Rutlands, a common clay that's also widely used for quenching, falls apart a little too easily to coat a forge, so I would stay away from it for that purpose. 

I'm not sure the exact water to clay ratio, but the goal is a sour cream like consistency. I just mixed in bit after bit till I got what I liked. You don't need anything in particular to coat it with; I just put on disposable gloves and slathered it on by hand. The clay doesn't have to be particularly thick; I did just one coat, and put it a little thicker at the insulator's seams. 

Side note: because I had gloves on and they were covered with clay, I wasn't able to get any photos of this process, but it's pretty self explanatory

I also packed extra around the mouth of the torch, for extra security and structure. 

At this point we're pretty much done. It's best to wait at least overnight for the clay to dry, and though it was still slightly damp the next day I fired up the forge anyway with no issues. It works like a dream: gets up to heat in no time, the doors swivel perfectly, and the amount of flame is perfectly variable!


You'll notice that I've saved one of the more important parts of this build for last: the torch, which will be next week. I planned on publishing that article much sooner, but due to it being a little more particular and with smaller room for error than the other parts, it's taking longer to write, so bear with me. I should be able to get it done by next week. 

Forge Build Pt II


Building The Forge Body 

 The forge body is the main component, holding the parts and housing the insulator. If you have access to a welder, it's simple to design and build the body to your exact needs, but many of us use reclaimed materials, usually a tank of some kind. Old propane tanks are a popular choice. 

As you know from last week's article, I used an old fire extinguisher tank that I picked up a few months ago at an antique store. It's compact, with a handle at the bottom and a tight mouth at the top. 


6519a9b3-e8d0-46a6-b4dd-389b9a3b4285 (1).jpg


I started by cutting off both ends just below the weld seams using an angle grinder, and bolted on hinges to "reattach". This way I had two doors for the forge. On the one end I would have the original mouth of the tank, and this could open up further in case I had larger projects to work on such as axes or hammers.
The opposite side could open as well, in case I needed to work longer projects, and would keep it closed otherwise to conserve as much heat as possible. 




The bolts will later be covered with insulating blanket and refractory




A small side note: make sure you mark out properly which side is up, down, and exactly where the hinges should be. Of highest importance is to make sure you don't have everything lopsided or unbalanced, but also double check you have the hinges on the correct side. Because I am right handed, I have the hinge on my left when I'm facing the mouth of the forge. 

Next is the base plate. Because the body is circular, I bolted a flat plate of steel to the bottom. 

I like to keep my forge on top of a few cinderblocks, held up by a cart. I don't like to keep it permanently fixed anywhere, as I'm constantly moving and rearranging things in the workshop. Because of this, I used long bolts and kept them extended: the forge is placed on top of the cinderblocks and bolts go in the opening, this ensures that the forge is not bumped off its stand. 




The final step in the forge body is setting up the torch.

Now there are two things I did not do in this particular rendition that I intend to do in the future; first, is build proper angling for the torch. 

For peak efficiency, the torch is best built at a slight tangent to the wall of the forge, rather than directly down. An angled torch will shoot the flame along the wall of the forge, creating a swirling flame. This mixes, spreads, and contains the heat for as long as possible.

I intend to modify the torch better in the future but for now, this runs efficiently enough for me not to have any problems, especially as I don't do very long of workpieces.

Second is the method of fixture. I punched a hole through the forge roof, slid in the flare, and pinched down the edges of the punched hole to fix it in place. Obviously welding, or a better mechanical fixture is superior, but I was in a rush to get this going for certain projects and this is extremely sturdy for now, and will require only simple modification for a more firm structure in the future. Later insulation and clay structuring cemented it further, but that'll be a topic for next week.

7cb387ac-efe6-456c-b88d-c246665748ca (1).jpg


You can see the top is threaded; the torch shaft will later screw into this component.

The flare protrudes a bit into the forge, which had me a little worried that it would stick too far into the interior and overheat (disintegrating over time) but later when I installed the insulation it ended up being perfect distance.

A possibly superior way would be to cut the hole into the body and place the flare on top rather than inserted into it. Inserted as it is now puts the tip of the flare nearly into the forge interior, which means it gets red hot and at higher temperatures will disintigrate faster. This means I'll have to replace the flare months down the road but this isn't much of a worry, especially as once the insulation is installed it will be flush with the walls of the forge. 


So that's the forge body setup!  

I think this is the seventh forge I've build, possible eighth.

Definitely more compact, tight, and efficient than any other rendition to date.

If you have built your own forge and have photos of the setup, I would love to see what you made it out of and how it was done! There are always ways to improve. 

Next week will be the insulation, and following that an overview of the torch build (spoiler: it's a Ron Reil style burner).

Forge Build Pt I


Planning The Forge Build

Hey all! It's been a bit of time since I last wrote. But, with a bit of equipment needing upgrading, it's time to get some more blade talk out there. 

I forge all my knives using a propane forge I built a few years ago, a fairly redneck one but it worked extremely well. Despite this, it's starting to fall apart and there are some things it lacks that I could use.

So, it's time for another forge build. 

This series of articles will cover everything from the bare frame, to insulation, to building the venturi burners. 


The build of a propane forge usually revolves around the main "frame" piece, the steel containment. I like to browse antique stores a lot (it's a bit of an obsession), and on one of my journeys a few months ago, I found an antique fire extinguisher.

Did you know, by the way, that fire extinguishers used to be based off a soda-acid reaction in order to build up pressure? Who knew! Turns out those vinegar and baking soda volcanoes you made as a kid actually had an application. 

I picked it up originally intending to use it for a quench tank, but never put it to that use. Instead, it's the perfect size and shape for a propane forge.

Having a fire extinguisher made into a forge is a bit poetic, don't you think?


The forge will have hinged doors on both ends, the front end (the top of the extinguisher) having a small opening, but with the ability to open wider (see sketched diagram above), and the back end completely closed unless the door is opened. This will allow me to contain more heat, or open the forge for wider projects.

Inside the tank will be ceramic wool (an insulator), coated with a refractory, and layered along the bottom with firebricks.
Assuming I can build it right, I'll have a few torches inserted nearly vertical, but at a slight angle to swirl the flame. This swirl helps the efficiency of the forge as well as contains more heat.

The torches will be simple venturi burners; we'll get to the details later when I begin building them. 

Feet for the forge will be welded or bolted on, and it'll be fixed atop an old barbeque cart, as I like to be able to move the forge around. 



So, there we are.  As far as tools go, forges are actually extremely simple. A propane torch pumps out flame, all you gotta do is build a structure that is shaped so that this heat is most efficiently contained and controlled, and the workpiece supported. Under this basic principle of course, there are a lot of nuances that help work together for an efficient, convenient forge. 

You'll notice that I didn't draw up full blueprints and my plan isn't exactly complete, but that's partially how I work: I like to start with a rough idea of a plan, start following it, and modify as problems or opportunities arise. And if there's one thing I've learned, it's that problems or opportunities always arise. 

On Trying Something New

When you started making knives, the entire process was foreign and new. You learned to get better at forging, grinding, finishing, fitting, woodworking, shaping and all by jumping right in and applying the book knowledge you gained by reading or watching videos. 

When you made your first knife, regardless of how pretty it turned out, you learned something. Usually something you can apply to make the next knife better, but often you'll find out a trick that made some part of the process easier or better somehow.

Now because this was your first knife, it was new. Because it was new, you learned something. This principle carries on whether you've made five knives or five hundred. 

So what I like to do is every once in a while, is build what I call a "Novel Project".

The Novel Project is something new or unfamiliar to me. It could be a Katana, it could be a chefs knife, it could be a folding knife.

By now I have the basic know-how and experience to have a good idea of what to do in each step, but having not done it before forces me out of my comfort zone and to find new methods to get around unique challenges, especially those that are unique to me and to the tools available. 

The best thing about a Novel Project, is if I return to try making it again at a future date, the second time around is far better than the first. This is both because of the experience and knowledge gained the first time around, as well as my natural improvement from other projects over time. I only fully realized this when I went looking at pictures of some of my old work. These are all the daggers I've made, four in all, with roughly a year in between each. 



The photo sequence is a little wonky, but it's quite clear in which order they were made. Bottom left, top left, top right, and finally bottom right

Now I talk about using a Novel Project to get better, to stretch your skill muscles and to gain experience. And that's all true, but to be really honest, my motivating factor for them is to have fun. That's really all it is. Usually a Novel Project is spurred on by sudden inspiration, and it's really a revival of my first love for the art. Everything else is a bonus, as great a bonus as they are. 

Of course it's good to stick with one style for an amount of time; my preferred one was small bowies, but this is about the benefits of doing a Novel Project. 

Now to continue my earlier point: the first novel project in a certain type (say a chef knife) dips my toes in the water, and often is a complete failure. The second one usually is fairly passable, but not sellable. The third, I can refine the points and mistakes I made last time. The fourth, I can focus on the details and trying to excel. At this point, the novel project has become familiar, and if I like it I'll continue doing it, but it's no longer a novel project. About this time I'll get the bug to try something different, say a katana. 

Over time this all adds up. Every time you do something it becomes a little more familiar, and you gain a little more skill. It's very much like a video game; as you get better at the game and play it more often, you unlock new characters, and harder and harder levels. So it is with knife making. As you make knives, if the attempt fails, you've learned enough to correct it next time, and as you get better, you gain the skills to attempt new novel projects that would have been impossible a dozen knives ago. 

This, coupled with the attempt to make each knife better than the last, is the real way to improvement. After all though, it's all about fun. We put far more effort into the job when we enjoy it. 

Thoughts on Design

I had been working on a collaboration piece with a friend, Timothy Artymko, which proved to be a bit of a challenge. The blade is a san mai (three laminated layers) bowie style with a lot of character. The thing was, it was a little different in style than I'm used to working. This one had distinct curves and a wide blade with a large belly. 

My job was the handle. 

I started working on a fairly basic handle, a pretty standard shape that I use with some variation on a lot of my blades, and got it all the way to where I had it polished out and fit up. I was going for a takedown style so it wasn't (thankfully) permanently fixed to the blade. 

I just didn't like it. Something seemed.... off. Like the handle was too skinny for the blade, even though it felt very natural and useful in the hand. It fit my grip but not the blade. 


And so even though I felt that it was "oh well", I didn't finish it. Just put it in the "almost done" drawer and worked on other projects. Which I felt bad about delaying on, but I didn't feel the life in that. 

So, after finishing another blade, I decided to push myself and get back to it. I thought "Okay, so this handle seems too slim and fast looking. The blade looks thick and curvy. I need a handle that reflects that."

With a bit of tweaking I went with some more dramatic curves, and some more tipping down of the butt, going for a curve that reflects the dip of the blade's chill. Because the blade's features are fairly dramatic, I went with some dramatic curves and corners in the hand. 

As many practitioners of the martial arts know, "The weapon is simply an extension of the wielder." So too, the handle is simply an extension of the blade. It should reflect it. 

This new handle was a huge improvement, you can see this even with the second one in extremely rough shape. I'm not saying that it could be better still, there are always little bits that could be made better, but I am very happy with the profile on this one. 

See if you can identify what features made improvements from one to another. There are often invisible lines you can imagine which point out either errors or successes in making the design flow from blade to handle. 


Heat Treating part VII

Summary and Miscellanous

This is the last of the articles series on basic heat treating. I'll be putting them all together in a PDF for E-book form shortly, and I'll keep you updated on the progress of this. With the close of this series, I'll be starting a new one next week, and haven't yet settled on a subject. If there's anything you're curious about and would like to have addressed, shoot me an email!

Today we're running through a basic summary of the past six articles. 

The first element to understanding heat treating is that of hardness, and toughness. A hard knife will stay sharp (hold an edge) but is brittle, a tough knife will not break but will easily dull. The harder a knife is, the less tough it will be. The tougher a knife is, the less hard it will be. The point of heat treating is to get at just the right point between the two extremes. Coincidentally, this is often yielding in a springy steel, one that will not take a set when bent. 

Now the second element in understanding heat treating is at the molecular level. 
Iron is an element, carbon is also an element. Steel is an alloy; a mix of the two, and often other elements. Steel requires a certain amount of carbon in order to be properly heat treated, generally between .40% and 1% Carbon by weight. This is known as high carbon steel. 

High carbon steel, at room temperature, is generally in a small, tight cubic formation called Body Centered Cubic (BCC). When heated past critical temperature, which can be tested by using a magnet (it loses magnetic attraction at this point), the atoms switch formation into face centered cubic (FCC) which is also a cube formation but slightly larger and more open. Carbon atoms, which interspersed between BCC cubes, now move inside the FCC cubes.

If cooled extremely slowly over many hours, the FCC slowly changes back to BCC and the carbon atoms slip out into nice, easy locations, and the entire piece of steel moves to the easiest, most "relaxed" state. This is known as annealing, getting the steel dead soft.

If cooled at a normal pace (air cooled), much the same happens but the steel retains some hardness. Doing this several times (heating then cooling to black), each time will refine the grain structure of the steel. 

If cooled at a very rapid pace, using a quenchant liquid such as oil or water, the carbon does not slip out in time. The structure becomes very rigid, tough to move in relation to one another but when they do it's at the breaking point. 

Normalizing is done, as mentioned above, by heating a bit above critical and allowing to cool until it's completely lost its color. Why? Steel is made up of "grains"; granules of atoms that are all linked together in a cubic formation oriented the same way. A bordering grain will have its atomic orientation a slightly different way. This border is a weak point. The larger the grains, the weaker the steel. You can see the grain structure in the broken surface of steel, ranging from milky looking to coarse sand. 

Normalizing then, is to refine; that is, multiply and shrink these grains. The more a steel is kept at a malleable temperature and forged, the larger these grains grow. Now, when the steel is heated past critical, FCC grains begin to form at the borders of BCC grains. Once it's been completely converted to FCC, the steel is left to cool and now BCC grains form at the border of the FCC ones. This is repeated several times, each time because there is little time to keep growing, the grains get smaller and smaller. 

Hardening is done by heating to critical and cooling very rapidly, using a quenching like water or oil. The rapid cooling entraps carbon atoms inside cubes too small for them, and a lot of rigidy is put over the structure of steel. The atoms are hard to move in relation to each other (thus it will not dent or bend) but when they do, it's by breaking. 

Tempering is done by heating anywhere in between 350 F to 450 F or so for an extended period of time; this eases up on the steel's pressure, toughening it and making it less brittle. 


Often during a quench, the steel becomes warped due to uneven cooling or stirring in the quenchant. It is unwise to try to straighten it directly after a quench, or even putting pressure after tempering. I've found that by far the most effective and least risky way is to make a jig to bend it slightly in the opposite direction of the warp, and temper like this. The heating makes it slightly more plastic and more likely to take a set, and it is far more controlled then trying to bend it after tempering. 

Differential Heat Treating

As mentioned above, a good blade is one that can hold an edge and that is tough enough to withstand shock in use. One way to get the best of both worlds is to leave a spine that is soft and tough, and so can take much stress, and an edge that remains hard, although brittle, so it can keep an edge. 

This can be done in one of two stages. During tempering, and during the quench. 

If done during tempering, the process is simple. Using something like a torch, the smith heats up the spine only, to a bluish temper color, avoiding heating the edge as much as possible. 

If done during hardening, we can again break it down into two methods, both based on a simple principle which is this: if the steel is heated to critical temperature, and cooled within a certain time frame (i.e., if the steel cools within 3 seconds, it hardens, if it takes 4 or more, it remains soft), it will harden. 

Based off that principle, the first method is by heating the edge only to critical, then quenching. Thus, only the edge hardens. 

The second, you can simply heat up the entire blade but only quench the edge in the quenching liquid. This means the edge cools within the required timeframe but the spine does not. Similarly, a method was developed by the Japanese called Yaki Ire- the clay quench. An insulating clay is applied over the blade's spine, and the blade is heated and quenched in full. The clay slows the cooling rate of the steel underneath it, and so that steel cools too slowly to harden. The benefit of a clay quench is that when polished and/or etched, the steel exhibits a border line between the soft and hard steels, which is often breathtakingly beautiful. 

This covers and summarizes all the articles so far. I apologize for today's article being rather late, but I'm glad it's finally out. Once again, suggestions for future articles are welcomed, shoot me an email, or if you have any questions I would be glad to help out!

An Intro to Heat Treating, Part VI

Differential Heat Treating

As mentioned way back in the first article of this series, there are two basic qualities of steel. Hardness/Brittleness, Toughness/Softness. A harder steel will be less tough, and a tough steel will be less hard (capable of holding an edge). 

But what if you could get the best of both worlds? This is the thinking behind differential heat treatment.

If the edge is harder than the spine, this should increase performance a good deal.

However, a disclaimer is required. Tempering is still very much necessary even with a very tough spine. Most fully hardened steels are too brittle to hold up as an edge; no, they will not bend and dull, but it can chip. Tempering (though possibly at a lower temperature than normal) is still required. 

 There are two main ways to do a differential heat treat. One is in hardening, the other in tempering. 

Tempering is fairly simple, and I touched on this in last week's article. Differential tempering is done by heating the spine, while leaving the edge cool. Usually this is done with a torch, carefully heating the spine to a nice blue color or so (around 600 Fahrenheit). It's risky to do this in air however, as the heat can very easily creep down towards the cutting edge, ruining the temper. 

The other way is by differential hardening. There are several ways to do this. Now, the basic principle is this: if a steel must be cooled from a certain temperature in a certain timeframe, it will be hardened. To keep a portion of steel from hardening, you simply need to either slow down the cooling, or not get the steel hot enough. 

So, firstly, we can heat only the edge to critical temperature. This can be done with something like an oxyacetylene torch; heat up the edge, test with a magnet, heat some more, and quench. 

A second method would be to do an edge quench. This is done by heating the entire blade, and only holding the edge of the steel in the quenchant. With experience, this can be done with great results, but beware too much heat creeping down from the spine. 

A third method would be to quench the entire blade, keeping in mind cooling rates, temperature, and mass of the blade, and remove the blade quickly enough. With enough mass on the spine, and little enough thickness on the edge, it is possible to use timing and calculation to produce an "autohamon"; where the edge is hardened and the spine remains soft.

The fourth and most famous method is to use an insulating clay in a "clay quench". This method, used in traditional Japanese bladesmithing, is to coat the spine in an insulating clay, heat up the blade, and quench. The edge hardens normally while the clay slows the cooling of the spine. This leaves the spine soft while the edge is hard. With careful controlling of temperature and especially the clay, absolutely insane hamons can be produced.

Hamon 刃文

When a blade is differentially hardened, there will be a section of soft steel and a section of hardened steel. If the steel is properly polished and/or etched, the border between the hard and soft steel while become visible as a line of varying appearance. This visible line is called a "hamon", and is highly prized. It can appear either as a dark line separating a gray section from a silver, or it itself can be a ghostly white flame flitting through the steel. Collectors of Japanese swords deal quite extensively in the shapes and characteristics of hamons.

Bringing out a hamon again is based on a fairly simple principle: one area of the steel is hard, another is soft.
There are two standard ways then, to bring out a hamon. The first is the traditional Japanese method: polishing. Learning to do this perfectly takes years and many expensive natural abrasive stones, but it can be replicated by hand to a lesser extent. Exact methods are beyond the scope of this article. However, the idea is using various abrasive stones to rub away at the surface of the steel: the softer steel will wear off at a faster rate than the harder, and proper application can bring out some very stark contrast. 

The second method is by acid etching: acid eats away at the softer steel at a different rate than the harder, and a different rate at the border line itself (hamon). The ideal acid is diluted ferric chloride (circuit board etchant), but boiling vinegar or even lemon juice will work. Everyone has slightly different steps in etching; a black oxide layer forms on the surface, and it must be cleaned of periodically throughout the etching process. The surface finish pre-etching, longer etching, more or less frequent cleaning, acid concentration, and cleaning type will all have some effect on the appearance of the hamon. 

Deep-hardening and Shallow-hardening steels

In the world of differential hardening, there's an important factor to keep in mind: steel. Basic carbon and tool steels can be divided (so much dividing in this article!) into shallow and deep hardening steels. The difference is based on required cooling rates. 

For example, we take a made-up steel, Steel D (D for Deep hardening), in order to harden, must cool from 1600 F to 400 F in five seconds. If it takes six seconds to cool this far, the steel will not harden. However, five seconds is quite a long time. A plain blade in oil can cool this entire rate in three seconds. So if we apply clay, which slows the cooling of the spine by one second, we still have a blade completely cooled in four seconds. Thus, the clay has no effect. The blade has hardened through and through. 

Steel S, (for Shallow hardening) takes three seconds at the same temperatures to cool. If we retard the cooling by one second using clay, the surface underneath the clay takes four seconds, exceeding the upper limit required for hardening. And so, this area remains soft. A thicker chunk of shallow hardening steel, even if quenched without clay, may differentially harden due to the extra mass and residual heat, or the surface may even harden while the core remains soft. This last point is what is referred to when we say deep or shallow hardening.

Whether a steel is deep or shallow hardening depends on its chemical content. As a general rule, the more complicated the makeup, the deeper hardening the steel is. Some elements may extend or shorten the time required to harden. This is why one steel, such as W2, more readily takes a hamon, while another, such as 1084, is a little tougher to selectively harden. 

So there you have it, the basics of differentially heat treating. This process can be broken down into two main branches: differential tempering, done usually by heating the spine post hardening, and differential hardening. Differential hardening in turn can be broken down into either selectively heated edge quenchingmono-temperature edge quenching, creating an "autohamon", and clay quenching
The Hamon is the visual effect of a differentially hardened blade, brought out either by polishing, acid etching, or a combination of the two. 
The type of steel will affect whether it is possible to selectively harden a blade, given the same processes. Some steels are shallow hardening, requiring a short window of time to harden, and some are deep hardening, allowing for a longer window of cooling rate to harden.

This is the sixth and second to final article on basic heat treating; next week I'll do a basic run-through of the subjects we've covered as well as some anomalies that didn't fit into the subject titles we've named so far. 

An Intro to Heat Treating, part V

Tempering The Blade

The first thing to remember is the principle I brought up in the very first article of this series:

Steel can be thought of on a sliding scale of Toughness/Softness to Hardness/Brittleness. 

If hardening jumps us to the extreme hard and brittle end of the spectrum, tempering is the process to bring it back down to the ratio we want. That said, tempering is really a very simple process. 

The basic idea is to thoroughly heat the blade at a fairly low temperature, ranging from 350 F - 450 F. The hotter the temperature, the softer the steel. Many folks use a kitchen oven (use a thermometer; the built in one is usually fairly inaccurate), but a toaster oven is plenty ideal.

Now many people use a method called "torch tempering", which technically works okay, but it's inconsistent and risky. Torch tempering is using a torch to heat only the spine of the blade anywhere from a straw yellow to a blue color (called "temper colors"), thus toughening the spine and leaving the edge hard. However the problem with this is that it's incredibly easy to heat the cutting edge, softening it too much and thus ruining it. This would result in having to re-harden the entire thing. The other issue is that the steel isn't completely "soaked" in the heat for a sufficient time, and it can still remain very brittle. 

So, the ideal way to temper is for several hours. Various makers will temper anywhere from 3 to 6 hours, ensuring a thorough heating. A common method is to temper in several cycles: for instance a few hours at 420 F, cool in water, then a few hours at 375 F. 

Now remember, the temperature you use depends on the steel. This can be found by googling different heat treating charts for the steel type. 1095 steel tempered to 400 F will not act the same as 1084 steel tempered to the same temperature.
Not just this, but a sword will require a higher temperature than a paring knife, for example. A sword needs to be much tougher than a paring knife does. 

So, that gives you quite sufficient to perform your own heat treats. The key to getting excellent results (chopping boards of wood while retaining the edge, bending without snapping, etc.) is constant testing, noting, modifying, and retesting.

Walter Sorrels has an excellent video on testing, I highly recommend checking it out. 

Now, as tempering is indeed fairly simple, I'll move on to the subject of warps, and more specifically, removing them. 

Warps and Straightening

Especially with longer blades, it's very easy to get a warp during the quench. This is usually due to uneven cooling in the quench, mostly something like stirring side to side rather than in a cutting motion. In any case, a warp is always possible, even when you think you've done everything to prevent one. 

So, straightening the blade. There are several methods to doing this. A common one, is to flex the blade in the opposite direction of the warp directly out of the quench, while the blade is still steaming or smoking. It's generally able to take a change in direction before it's completely cool and set, but this is incredibly risky. One pound too much force and snap goes your blade. This can be done effectively, however it's best left to those with a lot of experience and have a deep intuition to know the breaking point. 

Another method is to try putting pressure on after tempering. Assuming you've tempered the steel soft enough to take a set rather than spring back in blade, you can set up a jig in a vise using a few pegs to straighten it. Two pegs on one jaw of the vise, on the inside of the warp's curve, and a third on the other jaw, on the outside and center of the warp's curve. When you tighten the vise, this will bend the blade in the opposite direction, hopefully enough to take just enough of a set to straighten it. 

However, if you didn't temper hot enough to take a set, one of two things will happen. First, you'll have to bend it very far to take a set. Second, it may snap before it gets quite that far. Tough though blades may be, they're not meant to be flexed that far. 

The two methods above are known as cold straightening. I've snapped many a blade trying these and it's not pretty, though it does enable you to see the grain structure to check your normalizing job. 

So, since then I've elected to take a much safer, and in my opinion, more effective, of an approach. 

Quite simple, really. I've found that if you straighten the blade while it's tempering, it's in a hot state, and changing, and is more susceptible to taking a set. My preferred method for doing this is get a flat bar of fairly thick steel and gently clamping the blade along it using a c-clamp (inside of the warp's curve against the steel). Sometimes if this still springs back I'll put two pegs between the blade and the bar, each peg at opposite ends of the warp. The C-clamp's pressure in the middle causes a reverse flex, and while not enough to take a set cold, it should do so nicely while tempering. Be careful not to over tighten before it's begun tempering; hardened steel is extremely brittle. I like to sometimes just get the blade "taught", then tighten at several intervals throughout the tempering cycle. 

This technique has produced the highest rate of successful straightening for me. Another method that I've found to be effective without risk of snapping is when cooling in between tempering cycles. When you cool with water between cycles, if you cool in a sink or something and spray the water on the outside of the warp's curve, that will cause that side to contract slightly faster than the opposite side. With experimentation, proper application, and a dash of luck, you can bring the blade back to perfectly straight using just water. This of course depends how serious the warp is. 

So, tempering is the process of heating hardened steel anywhere from 375-450 degrees fahrenheit. This softens and toughens the steel just the right amount, to bring the steel to the desired balance between toughness and hardness. Warps can occur during the quench, and I've found that straightening is best done during tempering, both for effectiveness and to lower the risk of snapping. 

With tempering, we've covered just about everything in basic heat treating. Next week, I'll delve a little bit into selective hardening and hamons. Note however that I'm not an expert on clay heat treating, but I've done my bit and will pass on what I've found most helpful. 

An Intro to Heat Treating, Part IV


After several cycles of normalizing, and after the blade has been ground to rough shape, you still don't have what many knifemakers will call "a knife".

The more philosophical among our craft will call it a "knife shaped object" at this point. 

The quench is where the steel begins to become a proper and capable blade. 

So, let's get down to what happens

If you remember the very first article in this series, in blade terms there are two basic qualities that are on opposite ends of a sliding scale. The first is toughness/softness, and the second is hardness/brittleness. A very tough blade is rather soft and will not hold an edge, but a very hard (edge retaining) blade is rather brittle and will break easily. 

The final heat treating in simple terms is broken down into two basic parts. Hardening and Tempering. Hardening comprises of hardening the blade (who would thunk?) and Tempering softens it sightly, to the exact hard-tough ratio you like. Today we're talking about hardening. 

The smith begins by heating up the blade as evenly as possible in a furnace, kiln, or forge, to reach critical temperature. Professional smiths will use a very controlled heat treating kiln, to reach just the right temperatures and hold those temperatures (soak) for just the right amount of time. These times and temperatures of course vary from steel to steel, but for basic terms in the scope of this article we're just concerned with getting a nice even hardening. 

In something like forge with a single area of heat that might not reach the entire length of the blade, the best method is to slowly draw it back and forth to get an even heat over the entire length. If possible, keep the blade out of the direct flame, this helps to reduce warping, scale buildup, and uneven heat. A common technique is to use a scrap steel pipe as a "mini kiln", which is place under the flame. You can then draw the blade back and forth inside this pipe to get a near perfect, even heat. 

In the absence of a thermometer there's one way to test whether you've gotten up to critical temperature or not. Once the phase change happens (atomic structure goes from BCC to FCC), the steel loses it's magnetism. So, you can touch a magnet to the steel as a test: if the magnet sticks, it's not hot enough. Put the steel back in the forge and heat it up some more, then remove it and test again with the magnet. 

An excellent technique is to always heat treat at night or in relative darkness. That way, after you've heat treated a few blades using a magnet to test, your eye will grow accustomed to the glow and color of the hot steel, and you'll know exactly when you're at the right temperature without having to magnet test. If you heat treat at various times in various lighting, the steel will appear different each time. 

Now that the steel is up to temperature, the next step is the quench

The basic idea behind hardening is this. The steel, at roughly 1600 Fahrenheit, switches atomic structure. If you can cool the steel down to a certain temperature within a certain timeframe, the steel convert to a hard, almost glass-like state. Cooling in plain air is too slow for most steels (there are exceptions!), and so a liquid is the best bet for an even, fast quench. 

That said, let's have a word about different quenchants. Plain water is a common quenchant, traditionally used in Japanese bladesmithing, but it's often too violent of a quench. Both because it's extremely fast of a quench, and because the vapor bubbles create a very unstable environment for the now-glassy steel, and the shock often results in a sharp *PING* noise, indicating a fracture in the blade. 
Because of this, various oils are used. The standard professional oil, Parks 50, is of course ideal, but other oils such as motor (not healthy when burned, so be warned) oil, peanut oil, olive, canola, etc. work very well. Often using a brine helps eliminate the violent bubbles in a water quench. Preferably do some preheating (120 F or thereabouts being ideal) for the quenchant, this lowers the viscosity and makes for a more thorough quench. 

Now sometimes there can be issues with using just oils (too slow), as well as just water (too fast or too violent). This is where you can experiment with interrupted quenches, but that'll be a future article. 

The actual quench must be very quick and very precise. Have the quench tank nice and close to the forge but give yourself plenty of room. 

You can quench either horizontally (edge first) or vertically (tip first), but for our intents and purposes it doesn't really matter all that much. Quenching horizontally, if you can, gives you the ability to edge quench (again, more on that in future articles) and observe different factors that may lead to a failed heat treat. 

Completely submerge the blade in the liquid. For constant cooling, move it back and forth in a cutting motion: don't go side to side, as the blade is still in a plastic state and this can cause warping which is a nightmare to get rid of. 

If you quench in oil, there can be flame, but really you don't actually want this. Any portion of the steel hot enough to cause a flame should be completely submerged. To put it simply, fire needs three things. Oxygen, Fuel, Heat. If the hot steel is completely submerged, there's no oxygen immediately adjacent to the oil that's actually hot enough to catch fire. 

Continue to stir back and forth until the smoke is nearly gone, and the steel will be cool enough. I like to remove it, wipe off with a rag, file test, check for warps, examine for any cracks, then immediately go to tempering. 

The File Test

There are numerous factors that could lead to a failed quench. Whether too cold of steel or oil, uneven heating, too slow of transfer, even things such as atmosphere humidity can affect a quench. So, it's important to check whether the steel hardened up properly. 

The way to do this is with a file. Files are heat treated to a very high hardness; the teeth need to be extremely hard to stay sharp, but a file isn't put under much stress so it can be fairly brittle. Of course, they are still tempered (softened slightly), so a file is the perfect thing to test hardness. 
It's quite simple. 
If the blade hardened successfully, it will be harder than the file. If it failed, it will be softer than the file. So, if the file is harder than the blade, then when it's drawn across the blade's edge, it will cut in and drag. If the file is softer than the blade, then the file will simply skate over the surface as if it were glass. 

Examining for Cracks

Possibly the most heart dropping sound in bladesmithing is the tiny *ping* or *tink* that comes with a fracture forming in the quench. The fractures that come from this rarely form across the entire blade, rather being anywhere from a few millimeters to more than a centimeter in length, perpendicular to the blade,  and starting at the cutting edge. 

Unless you're making yourself a scrap blade, if you find a fracture, it's time to scrap it or see if there's enough steel left to chop it into a smaller blade. Never try selling a blade with a crack in it, even if it's practically invisible. 

Now the blade is usually covered all over with black oxides from the quench and it's often hard to see if what you're looking at is a crack, or something like a scratch filled with oxide. So long as you see a matching mark on the other side, indicating it went all the way through, then you can be sure it's a fracture. Scrap it.

So, you've heated, quenched, tested for hardness, and examined for cracks. The final step before tempering is to check for warping. It is very easy to do this. Hold the blade up so you're looking down the length of the spine, as if you were looking down a rifle or shotgun barrel. This enables you to see very clearly any wobble or warp. If the spine looks straight, flip it over and look down the cutting edge. If they check out, it's on to tempering.

If not, if you see a warp or wobble, then there's an extremely tense straightening process to do simultaneously to tempering. Most blades are ruined during the quench itself, but a close second is when you try to straighten and don't do it just right. 

I'll cover correct warps as well as tempering in next week's article. See you then!