A class on knife forging was just completed at MASW. The class was taught by a world renowned blade maker, Matthew Parkinson. (Matt is the artist on the right in the website photo.) It is so incredible to attend a class with nine other students taught by someone amazingly skilled in the art. The class ended with my knife not quite completed, my own choice.
I will just relate a brief summary of the class with many photos showing the progress and struggles of my first attempt at making a knife. Many mistakes were made along the way, none due to the teacher, most due to my difficulties hearing over the roaring forges, the making of which was taught the weekend prior to my class. Each student was provided with 16 pieces of steel, eight 15N20 and eight 10-84. The former is an alloy with 3% Ni and the latter an allow with 4-5% Mn. The nickel alloy stays shiny after etching, while the manganese turns black, leading to the damascene patterning.
The pieces of steel were laid on top of one another to make a 2" stack about 3" long and 1" wide. The ends of this stack were welded forming a block and a handle of rebar was welded on. First time I have welded since shop class in high school. I found it difficult to see the fine wire used for welding, but produced serviceable welds with a little help from a fellow student. This block was the beginning of my knife.
The block was heated in a forge powered by propane to almost yellow. From the forge it went to a 12 ton hydraulic press, where it was squished from two directions, causing it to flatten and stretch. This was repeated multiple times. The last pressings were done with spacers to produce a long 3/8" X 1" bar.
The bar, now at 16 evenly spaced layers, was cut into six pieces, each about 2" long. These six pieces were taken to the grinder, a metal cutting 30 grit sanding belt, where the faces were flattened. (My first of many injuries occurred when a block, wet from cooling, slipped out of my hand sending my left forefinger into the rapidly moving belt.) The six flattened pieces were stacked, welded, and a handle added as before. The stack went through the same process at the forge and the press. After this pressing the resulting bar (96 layers of steel) was cut into three pieces each about 3" long. Again the process was repeated giving a bar with 288 layers. Matt said over 600 layers yields more haze than pattern. The last forging did some minimal shaping to produce a rough knife shape.
At this point various modifications can be made to the resulting bar to influence the resulting patterns in the metal, cutting grooves, drilling, etcetera. I stuck with a random pattern. A forging error on my part left me with too little metal to feel comfortable removing any. Consequently, I defaulted to a random damascene pattern.
This bar went back into the heat and then under the hammer. The hammer was used to stretch the metal in the desired direction. With many corrections from Matt I was finally able to move some steel via hammer blows. My slightly undersized bar was not amenable to producing a true chef's knife, so a carving knife became the target.
The roughly hammered knife shape was then ground into final shape, removing both the scale and metal. The back was kept relatively straight, while the cutting edge was curved. The pits near the tang were forging errors of some type and were not removed by further grinding as that would have left the knife too thin.
Sanding by hand at the bench removed the coarse sanding marks from the belt sander as well as the many dings my grinding produced. 220 grit was followed by 400. At a later stage of construction the knife was sanded with 600 grit just before etching. The unconformity in the cutting edge seen in the first photo below was easily sanded out once spotted.
The handle was made next. A block of hard maple was selected. The maple had the shape of the tang traced onto it with the top of the block aligned with the sides of the knife to both sides of the tang. The center of the tang was marked on the end of the block. Then the center of the block in the orthogonal direction was also marked on the block's end. These marks were extended down the sides of the block. The block was set up in the drill press vise at an angle so the tang sketch was vertical. The block was drilled passed the length of the tang by about 1/2" with a 3/8" drill. The hole was opened to a partial slot by filing with needle files. A specially made tool for this purpose, provided by Matt, made this a much faster process.
A bolster was recommended, so a piece of bronze was cut from stock. This was backed by a piece of black plastic to highlight the bronze. A slot was milled in the bronze with an undersized end mill. The slot was then widened with a file to fit the tang tightly. The piece of plastic was drilled and also widened with a file, but did not require the tight fit.
A handle shape was scribed on the Sharpie blued face of the bronze with the aid of a template. The bolster was ground to rough shape and then finished with a file and sandpaper. The plastic and the bronze were then glued in place for shaping on the belt sander. This was a challenge for me, so much of the shaping was done by hand with file and sandpaper. Eventually a shape was reached that looked good. Various mistakes were corrected by Matt to make everything fit, including some tang modification.
The handle was sanded to 400 grit. Black streaks were evident on the handle from the iron dust everywhere. These were easily removed with a dilute solution of oxalic acid, the same acid in spinach that makes your teeth feel funny. After drying the handle was dyed with a leather dye, giving it a deep brown color. Unfortunately, the dyeing left two spots undyed, one seen in the photo below. Glue from reattaching the plastic and bolster (both came off twice during shaping the handle) must have dripped onto the handle and gone unnoticed during subsequent sanding.
For most students the handle was also finished with a coat of resin treated tung oil. I skipped this step as I wanted to eventually hide those two spots. In lieu of a finish I applied paste wax prior to glueing the knife into the handle. This wax will need to be removed with alcohol prior to applying a final finish. The next step was etching the knife. The blade, held by the tang, was set in a large jar of ferric chloride in dilute vinegar. After one minute the blade was pulled out and sanded with 800 grit sandpaper, while wet. The blade was returned for an additional two minutes and sanded again. The figuring was apparent, but faint. The knife was returned to the acid solution for an additional fifteen minutes, with one intermittent sanding session. The final sanding no longer caused the dark black areas to lighten, so was considered complete. Amazing to watch the patterns emerge during this process, turning out much more beautiful than I expected. The knife was neutralized in a bath of aqueous sodium bicarbonate, wiped off, sprayed with Windex, wiped off again and sprayed with WD-40.
Two part epoxy was pumped into the handle through the slot in the bolster with a syringe. The knife after wiping off the oil was inserted, wiggling to ensure no bubbles of air were trapped. The knife and handle were clamped with a block of wood bearing a hole protecting the knife tip. Excess glue was wiped off with a towel wet with acetone. The junction between the bolster and the blade was smeared thickly with paste wax to prevent any further seepage of the glue. The knife was left to dry for twenty minutes.
Matt then taught us his knife sharpening method, beginning with a fine diamond stone, 800 grit and progressing through three levels of Arkansas stones. The diamond stone is kept wet with a dilute soap solution, and a light oil is use to keep the Arkansas stones wet. The knife should be sharpened to a 15-20° angle on both sides. My sharpening did not proceed beyond the diamond stone. I was afraid to get it too sharp as I still need to do a lot of work on it. A good way to wrap the knife during subsequent operations will need to be devised.
The plan for hiding the white spots on the handle is inlay. A number of possibilities were explored, from abstract designs to birds to flowers. A simple acanthus leaf scroll was selected from a drawing seen on the web. This design was printed out and then cut out. It is shown below hiding the white dots on the handle.
The complexity of this inlay gives me some hesitation. It is one thing to inlay metal on a flat surface where depth can easily be controlled, quite another to do the same on a curved surface. Two things will be explored prior to cutting the handle: cutting and engraving the acanthus leaf and inlaying it on a round dowel. The knife is beautiful and I don't want to screw it up.
The first order of business is cutting the brass into the paper pattern seen above. Brass thickness was considered, ruling out 1/16", as it would probably not bend around the curve. Some 0.017" thick brass seemed too thin for engraving, so 1/32" thick brass was selected for the first attempt at cutting, engraving and bending.
A rectangle of brass was cut from stock. China white was smeared on the brass. The design was laid on the brass and with carbon paper behind it, the design was traced onto the brass. The acanthus leaf was cut from the brass rectangle with a jeweler's saw.
The brass was affixed to a wooden block with carpet tape. Engraving was done with the block held in the engraver's vise.
A three inch length of 5/8" oak dowel was cut and used as the bending template. After aligning the engraved brass to cover the two undyed spots, the brass was marked. These marks were used to align the brass with the grain running the length of the dowel. A scrap of leather and a hammer were initially used to begin the bending. A bit of bending could be done by hand. The hammer was not very effective as my hand could not hold the piece in place, while hammering. Pliers and vise grips proved to be more useful, always with the leather between the brass and the jaws. The photo below shows the mostly complete bending. Adjustments are needed in a few spots.
A realization hit me last night. The brass inlay needs to be rounded on a 9/16" circle not 5/8" as that will be the diameter of the pocket eventually cut into the handle. The oak dowel was quickly turned down to 9/16". After a lot of work with the pliers and the hammer I am no closer to getting this bent appropriately. It really needs to be heated to soften the brass. A metal dowel would need to be used as a form. It is not clear what can be used to shield the brass from the pliers and avoid marking the brass. The leather, used so far, will burn. Will heated brass be soft enough to bend with gloved fingers?
Heating brass and quenching in water is the way to soften it, so a scrap from cutting out the acanthus was heated in the garage with a propane torch. The scrap was quenched in water. It was no easier to bend than the acanthus. The brass has not been work hardened, but is just tough to bend. With vise grips and a large pair of pliers the brass scrap could be bent sufficiently to fit a 9/16" steel form.
With a large pair of vise grips, a large pair of pliers, and a small pair of pliers the acanthus leaf was slowly bent around the steel form protected by the leather scrap. It is not perfect but as good as I can get it. Now it is time to practice inlay on a curved surface.
In order to pull this off a fixture of some sort needs to be made for the Dremel, so a consistent depth of cut can be held. Ideally the fixture would be adjustable, but that adds significant complication in the small fixture needed. Depth will be set by cutter placement in the Dremel collet. Before making a new fixture the old Dremel routing fixture will be tried.
A printout of the acanthus leaf was taped to the dowel over carbon paper and the design transferred with careful pencil tracing. The fixture worked reasonably well. The challenge is not the fixture, but the small amount of base sitting on a 5/8" dowel. The photo below shows the result. Some deep dives can be seen where the fixture was tilted up off of the dowel. For the most part it went well, with rotations of the fixture around the dowel working up to about 15° off center. The dowel only had to be rotated twice to cover the design.
A small problem was discovered when attempting to put the brass leaf into the routed depression. The orientation is off. The axis of bend is not the same as what appears to be the center of the leaf, which is how the picture was transferred to the rod. This is purposeful to make sure the leaf masks the two white spots on the knife handle. The leaf orientation will somehow need to be transferred to the paper template prior to transferring the leaf to the dowel.
After thinking about these inlay struggles overnight, I have decided that inlay is not the best path forward for someone with my limited skillset. It is not just getting the orientation and cutout completed but also the cleanup afterward. The handle cannot be sanded. Any sanding removes the thin layer of dyed wood revealing the white maple beneath. Way too many difficulties for me to feel comfortable they all can be successfully surmounted. All means of hiding the two white spots by putting something in their place require sanding, a no-go.
Most web discussions for removing paste wax recommended wiping with mineral spirits on a soft rag. This will be tried on the end of the handle first to determine effectiveness and assess dye removal. The mineral spirits with no dye removal, leaving the expected dull finish. The handle will dry for an hour or two before adding the finish coats.
The Half and Half on hand was selected. Matt recommended tung oil, so that is what I will use. The mixture of tung and citrus oils was wiped on with a rag, applied heavily, though not as heavily as can be accomplished on a flat surface. It was reapplied after 45 minutes followed by two more applications. After drying overnight the handle was coated with Johnson's Paste Wax (no longer available for purchase) and buffed after drying on a soft clean buffing wheel.
Sharpening is the final task to complete on this knife. Pulling out my sharpening supplies, purchased for a carving class, revealed three diamond stones: 400, 1200, and 8000 grit. A 4000 grit stone would be ideal, but there doesn't seem to be a DMT diamond stone between extra fine, 1200, and extra extra fine, 8000. 1200 and 8000 will have to do. Both stones should be used with water or soapy water.
The knife was sharpened first with the wet diamond stones and finished with a leather strop. The edge was not as sharp as I prefer, hinting at my sharpening abilities. I used a set of Arkansas stones to give the knife a keener edge. Not as sharp as Matt's knives, but sharper than anything in our kitchen, where it now resides.