Wednesday, May 12, 2010

Tutorial, a knife start to finsh

A hand made knife, from start to finish

I am often asked “ How long do it take to make a knife?”. The answer is, it depends. That is not really all that helpful, so I decided to document a knife from start to finish and see how long it takes me.

Disclaimer: This is just how I work, and to some extent, just how I worked on this project. One should not infer that every smith works this way, or even that any other smith does. Any errors, misrepresentations, fabrications, or outright lies, are mine alone. In addition, this document is not intended to teach anyone how to make a knife. It is merely an illustration of the process, intended to inform my customers and passers by.

The Beginning

A knife starts as an idea. It can be an historical piece, an idea sparked by another maker, a picture, a drawing, or just the product of my overheated imagination. I am constantly thinking about knives (which is not nearly as creepy as it sounds) and I don’t really count this time as part of the design time in a project.

After the design is complete in my head, I often make drawings. I keep a bunch of sketch books scattered about the house and shop, so I can get something down as an idea occurs to me. A drawing can be a quick thing in pencil, just to get a basic shape, or a much more finished thing, with a materials list and such like. I often will do a single blade shape with four or five handle ideas, and I also do close up drawings of transition points like the heel of the blade.

The other thing that happens is I will get an idea, run right out to the shop with nothing in hand and start hammering steel. Sometimes these projects turn out just what I envisioned, and sometimes the steel takes me off in strange directions, there is no way to know before hand.

So what I am saying is, the time this stage takes….depends. It can take a few minutes, or it take several days, or weeks, or longer. Since I rarely work on just one knife at a time, I can be thinking about one or more pieces, sketching one or more, forging, grinding, polishing and all of the other stuff, all at the same time. As I am writing this, there are about 20-30 pieces scattered about the house and shops ( I am currently in three buildings, a bench in the house, a grinder shop, and the main shop housing the machine shop and the hot shop.).


Assume that I have just one project in train. I take the sketch out to the hot shop (where my forging gear lives, anvils, forges, AbiYoYo my power hammer, and all of the other tools), tape the sketch to a wall and find a piece of steel.

Just a short note, I try to use steel of a size appropriate to the knife I want to make. I could forge down a 2 inch round of 1084 to 1 x ¼, but I won’t without good reason. It’s one of the advantages of a modern smith, I can get steel in the size and shape I want, I don’t have to make it myself. It saves a bunch of time.

For this project I decided that I wanted to make a flattop Bowie/fighter in an early 19th century style. I have done several of these, including this one from my JS suite,

so I didn’t make a sketch.

I took a piece of steel from under the forge (which is where all of the tool steel cutoffs end up) and got started.

The piece is 1084, 1.5” by .250”, about 6.5 inches long.

I start by forging down the tip, this picture is just after one heat. I could just cut off the tip, but that tends to make the point of the knife a bit stubbier, and I wanted the length.

As I forge the tip down, the mass of the metal has to go somewhere, so the piece of steel gets wider. I know that I’m going to want to get rid of that extra thickness at some point, so part of every heat is spent forging the width back down to the .250” (more or less) that I started with.

This is after the 2nd heat. You can see that I’ve gained an inch or better in length already. This is the advantage that a forger has over a grinder, I’m not grinding away steel, (yet). I can start with a fairly small piece of steel and change the dimensions of it with my hammer.


I’ve forged in some taper to the blank, you can see the ripples from the hammer in the surface. I’ve gained some more length and cleaned up any problems. It’s taken me 4 or 5 heats to get to this point.

At this point I decided that I wanted more length, so I took a couple of heats to narrow the blank and draw some length. I could do this by hand, but since I have a powerhammer, I used it. It saves me time and effort. You can see that I’ve gained another inch or better in length.

I decided to start the tang, just because it makes holding the piece a bit simpler. I start with a top and bottom fuller to set the notch, and then one heat on the hammer.

This is the basic preform for most everything I do. A preform is just what a bladesmith calls a piece of steel that has been given a basic knife-like shape. None of the bevels have been put in and often, none of the taper. If I was going to do a dagger or a spear point I’d adjust the final position of the tip, or if I were doing a Seaxe, I’d leave well enough alone.

Once I have the preform made, I start to forge the edge bevel. I work 2 to 3 inches at a time, blending the hammered surfaces together. Forging the edge pulls the tip up toward the spine of the blank. The edge is getting longer and the spine is staying the same length, so the tip rises. In this case I also forged the tip into place, because I wanted a flat back, and not a big curved clipped point like a Southwest Bowie.

I’ve forged the edge bevels back to where I want them, and I’ve set the start of the plunge cuts. One thing you can see in this photo is the blank has a decided curve to the spine. On some other project, that might be fine, but in this case I don’t want it, and I will take it out in the next couple of heats.

And so I have. I’ve also squared up the ricasso. I took a couple of heats to take out the lumps and bumps, finish drawing out the tang, and to make sure that the blank was square and straight.

Here is the final product from the forge.

At this point I took 3 normalizing heats (I bring the whole blade up to about 1600 degrees F, and let it cool to ambient temperature.) On the 3rd heat the blank is buried in vermiculite overnight and allowed to cool slowly. Normalizing relieves stresses caused by forging, refines the grain structure of the steel, and sets the steel up to be hardened. This takes longer to talk about than it does to do. The forging is the most fun of any project (for me, at least) and takes the least time. Not counting any think time, I’ve spent about 40 minutes forging and another 45 minutes or so normalizing.

Total time: About 1.5 hours


Unless you like the look of a forge finished blade, and I don’t right now, everybody has to grind their knives. The next day I pulled the blank out of the vermiculite and ground off the forge scale with an angle grinder. The scale is very hard and will eat up a grinding belt. Once the majority of the scale is gone, I profile the blade and grind to a clean 36 grit surface. There is another 2 to 3 hours spent to get to this point.

Total time: About 4.5 hours.

I spent another 2 hours or so grinding at 60 grit and then at 120 grit and putting in the Spanish Notch. The blank is ready to heat treat.

Total time: About 6.5 hours.

Heat treating or hardening

I do all of my heat treating after dark. Even though my shop faces North, there is still too much light in the shop during the day. The light washes out the color of the steel and make it difficult to judge how hot the steel is. Too hot is as bad or worse than not hot enough. Being in near total darkness makes it much easier to see the color of the blade. Most of the books will tell you that steel needs to be a “Cherry Red” color to harden. I don’t really know what that means, to me, the color is “Tangerine”, whatever THAT means.

However, when the steel is at the right temperature (about 1550 degrees F for the kinds of steel I use) it becomes non-magnetic. This is the transition or Curie point. I keep a magnet stuck near the forge to test the blank with. Before I quench the piece, I do two more normalizing steps on the blank, just like I did before grinding it. This relieves any heat or grinding stresses I might have set up during the grinding process.

On the third heating pass, I carefully bring the blank up to the transition point. This is where being in the dark makes the difference. By taking the time to figure out what color the steel is at the right temperature, it’s much easier to find the color for the hardening step. When I have the blank at a uniform color, and have tested it with the magnet, I soak the blank for 30 seconds more and quench it, usually in oil. After quenching (if everything goes right) the blank is hard as glass. If I were to leave the blade in this state it would be too brittle to use, and it might even crack or break just sitting on the bench, so the next step is tempering.

All in all this takes a couple of hours, partly because the forge I use for this step is different than the one I forge in. Because it is also my welding forge, it has a brick floor and it takes 30 to 40 minutes to come up to an even heat.

Total time: About 8.5 hours


Quenching the blank makes it hard. Tempering trades off some of the hardness for toughness. I do this by heating the blank in an oven for several hours. The temperature the blank is heated to depends on a number of factors, the type of steel, and the intended use are most important to me. I could go into the details of the process, but in the interests of time and space, I’ll leave that research for the folks who really want to know. There are a number of good resources on the Web that talk in great detail about heat treating steel.

Once the piece is heat treated and tempered, it looks like this

You can see by the dark line that I quenched just the edge of the blank (I still call it a blank at this point. It’s not really a knife yet, just some kind of KSO (knife-shaped-object). Some time in the next few hours it will go from blank to blade, it’s still not a knife to me until all of the pieces are assembled). Edge quenching gives the finished blade a hard edge and a soft spine. A blade like this will take a lot of abuse, bending and twisting, without breaking. The black smudge is excess oil from one of the other blades I heat treated at the same time, resting on it in the oven.

I usually temper in 3, four hour blocks of time, so it takes half a day to do. On the other hand, I’m not actually doing anything except waiting. So I’m not going to count this time. If I have other projects in the works, I’d go and work on them, or mow the lawn, or clean the kitchen.

Total time: About 8.5 hrs. 12-14 hours wait time.

More grinding

Once the blade is hardened and tempered, I go back to grinding, in finer and finer grits. I have to be careful. If I over-heat the blank at this stage, I will have ruined the heat treating in that area. As the edge gets thinner I have to slow down my grinder to avoid over-heating, which means that it takes more time to remove the steel I need to take off. I start at 36 grit and work my way down to 220 grit, removing all of the scratches of the previous grit before I move on. 220 grit is where I like to start my hand work. This takes 2 - 4 hours for a moderate sized blade, up to 10 or 12 inches.

Total time: 12.5 hours.

Hand sanding and polishing

This is just what it sounds like, a lot of time, a lot of elbow grease. Depending on how well I have gotten all of the pre-220 grit scratches out, and how complex the blade is ( a double edge blade, or one with a false edge, or one with fullers (blood grooves) can take a lot longer) it can take 15 or 20 minutes per side, or several hours per side to get a clean, 220 grit, hand rubbed surface. Once that clean, 220 grit surface is established, the hand sanding goes quicker, but it can still take a couple of hours to get to about a 400 grit finish. I usually stop at 400 grit and start working on the guard and handle. I need to have the blade close by to test fit the guard and handle, and it’s going to pick up dings while it sits on the bench. This blade took about 4 hours to get to a 400 grit finish.

Total time: About 16.5 hours.

Making the guard, handle, and furniture.

I start with the guard. In this case, after fiddling with drawings for a while, I decided on a double lobe “S” guard in German Silver (a Copper/Nickel alloy). I sawed a piece off of a block and roughed the shape with my milling machine. Then I started to refine the shape with a file. It took an hour or so to this point.

After filing the basic shape I cold worked the arms of the guard. In the lower part of the next picture you can see how the arm has been tapered and elongated, and the end flattened.

The short arm was treated much the same, and then I dished the ends of the arms and cut the slot for the knife tang.

In the next picture you can see the dished finial and the back side of the guard (where the handle will meet the guard), I have relieved the back side of the slot so that as I file the slot to fit the tang, I don’t have to file away so much material. You can also see that the guard is not very symmetrical. It gets better as it goes along.

I decided to do a version of a shell guard. Since I had never done one, I spent half a day finding pictures of historical shell guards. Mine is not a copy, but my interpretation of one. I also decided that edges of the guard needed some file work as well, and that the spine of the blade (where I often do some decorative file work) didn’t need anything.

I then finished all of the surfaces while the guard was still flat, and then carefully bent the arms into the shape I wanted (at all times reminding my self which side had to go which direction. Being dyslexic is a challenge, at times). All in all I spent about 6 hours on the guard, and another 2 on online research for ideas.

Total time: 24.5 hours

Here is a picture of the guard and blade mocked up together. I did this so I could think about handles.

Fitting and gluing

It took me a full day to create the things in the following picture, about 8 hours of fitting and fiddling. There is the paper template at the top and the mock handle in maple. A German Silver frame with the pin holes located and the pommel attached and two bone slabs with pin holes drilled. Everything has to fit perfectly for a handle like this (well, for any handle, but this is fussier than most), if anything moves or is mis-drilled, it’s start over time. The two ironwood slabs were a test to see if I liked the look.

Total time: 32.5 hours

The next day I cut the shell features into the pommel and glued up the bone slabs. Once that had cured overnight, I ground everything to fit. All of that took another 5hours, not including the dry time for the epoxy.

Total time: 37.5 hours

Now we are in the home stretch. I took the blade and finished up the hand work, polishing to 1200 grit, marking the blade, final fit of the guard, and gluing and pinning the handle in place, and, after the epoxy was dry, cleaning the nooks and crannies where epoxy hides. I spent another 4 hours on all of that.

Total build time: About 41.5 hours.

And here is the final product, photo by Mitch Lum.

I don’t want you to think that this knife was finished in a just a week In fact, I did the forging parts and the rough grinding, and then the piece sat around for a couple of weeks while I worked on other things. After I heat treated this piece (and about 6 others) I pretty much worked straight through to finish it. So the actual time spent was several weeks, it’s just that not all of that was spent on this one project.

As I said at the beginning, I am often asked “How long does it take to make a knife”. The real answer is that I don’t have a clue. I usually say something like “Oh, 20-25 hours.” That’s when Marianne smacks me in the head and says that its double that, at least. I’m beginning to think she is right. I am not the fastest maker in the world, but I’m pleased with the way this piece turned out, so I guess that is OK. I hope you have found this little look into my world enjoyable and informative.

Geoff Keyes
5 Elements Forge
August 2008