I started this sword project in my free time several years ago when I was in high school. The first order of operations was tracing an actual katana in illustrator to serve as a template to cut the blade and tang’s general shape. I had to beef up the dimensions a bit for the sake of durability, but I wanted it to be as authentic as possible while still being something that I could make.
Next I moved onto the wood. I was aiming for as traditional an approach as best I could throughout the entire build, and as such I turned back to look for inspiration from real swords. Inspired by both the construction method and heat treatment process, I decided to laminate a piece of white oak between two thinner pieces of red cedar. The idea was that combining a softwood exterior with a hardwood interior would mimic the traditional Japanese style of surrounding a hard, brittle steel in between a soft, malleable one. Second, the beveling of the blade would expose the oak, playing as reference to the hamon of Japanese swords. This pattern is made when a layer of clay is applied to the blade before it gets heat treated. An interesting article on this process can be found here.
The tsuba was a quick part to make, first tracing a somewhat-typical shape onto the piece of oak. Next I cut the pattern out on the band saw and refined the shape with the belt sander. I traced the dimensions of the tang cross section on the guard and used the drill press to eliminate most of the material in that space. I also added a hole next to where the blade will be as seen in some real tsubas. Lastly, I finished shaping the two holes using needle files.
I started the process of making the handle by first splitting the board on the chop saw into two twelve inch long pieces. This is longer than most regular tsukas, but the idea was that I could use two pegs to help with lining up the pieces for gluing together. I then traced the tang of the blade onto the handle material, leaving a space between for the tsuba. Then began the first of several hours spent meticulously chiseling out spaces, starting with one half of the tsuka. I was surprised at how easy it was to get through the hardwood, but it was still a long process, with each side taking around an hour. I was quickly not looking forward to doing the saya.
Once I got both halves to incorporate the tang, I clamped them together with the sword inside. I removed the blade and drilled two holes for dowels, so that once I went onto gluing, they would line up perfectly without needing to risk gluing the tang inside. This worked out really nicely, so I removed the clamps and slightly tapered two dowels. I put one in each hole of one half and spread glue on the other. Then it was just a matter of clamping the doweled-half down onto the glued-half.
Once the glue was dry I ran the tsuka through the table saw until I was happy with the general dimensions, also lopping off the doweled section with the miter saw. Lastly I rounded the corners down on the belt sander.
Lastly I moved on to the saya, and I wanted to try my hand at the traditional method of creating the entire negative space in one piece of wood and then closing it off with a flat piece of wood. Aside from this difference, the process was very similar to making the tsuka. After around five hours of chiseling I felt satisfied with the fit of the blade in its housing. Before gluing the third wall on, I cut the general shape I wanted the scabbard to end up with on the bandsaw; however I could not control the blade’s wandering, so I left more material on that I would have liked. Finally I wiped glue onto the walls of the chiseled and shaped piece and clamped the other half on.
In the morning I first cut off an inch of end-grain to make the bottom edge uniform, but I went too deep and cut into the channel left for the blade. Thankfully I left extra room for the tip and was already planning on adding a small walnut end cap. After gluing a small square of scrap I had over the hole, I began to sand the saya down, rounding the corners and attempting to get a more uniform curve. However, this is when I started running into more issues. Because I hadn’t had success with the bandsaw, I was using the belt grinder to do the vast majority of subtraction. This was extremely slow and difficult to do smoothly on the concave side, but I persevered and got a shape I was relatively okay with. Unfortunately, when I went to put the sword in the saya, I found that it was much tighter than the night before to the point of the blade getting stuck very easily.
Since I could not just open the sheath up and continue chiseling out the channel, I was forced to sand the blade itself. Unfortunately, no matter how much I seemed to take off the blade just didn’t want to go in. Because I put myself in an unfortunate battle with the deadline I moved on to a rough finish, sanding from 80 grit to 220. I finally used linseed oil on all of the wood, resulting in deep browns in the grain of the oak, walnut, and pine while turning the cedar from pink to a vivid scarlet.
There are several things I would like to fix, the saya in particular. I am extremely disappointed that I couldn’t get the blade to fit nicely, and I know I could do a better job getting a more uniform curve. I also want to make the tsuba with a different type of hardwood (perhaps maple). I thought the roughness of the open pores oak has would play a nice homage to the texture of cast iron, which is what most guards were made from. Instead I just felt that it looked unrefined and a little bit boring. In a similar way, in the future I intend on using a different wood for the inner part of the blade. The pores once again let me down, splintering easily at the edge of the sword. Overall though I am happy that this assignment led me to continuing and finishing a project left covered in dust for years.