Multibar Mosaic Longsword

[AJ Chalifoux]

This is a project I've been working on for two years more or less nonstop. It was originally meant as an engagement present, though it was just taking too long so I missed the mark by about four months...luckily she understood when I told her I had something in the works and I wasn't going to rush it . I finally gave it to her this past weekend and she was ecstatic to say the least.

 

Since it was such a personal project, I documented the whole thing and I figured many of you would like to see the process as well as the results. There are a lot of pictures, so I'll do this piecemeal.

 

To start, here was the original inspiration, housed in the National Museum of Ireland. Note the double pommel and the grip is two hollow steel collars (originally with a wood core) and a brass spacer between them. There were also jewels set into the pommel, grip spacer, and quatrefoils which indicates this was a sword meant for a very important person. The hilt is significant because it has both Irish and Scottish design elements, and its origin is unclear.

 

From the museum:

“Sword hilt. Brass and iron. Hand and a half grip, cylindrical grip, iron, with brass band. Double pommel, first part iron, circular, domed either side with brass rosette. Second part circular, flat with brass disc on either side. Straight quillons, (one partly missing,) terminating in quatrefoil ornament with brass plate on either side. Found in River Barrow, near Monasterevan. Possibly Irish, 15th century.”

 

 

The intention was not to make an exact copy, but rather take the design elements I liked from this sword and make something new. The original sword likely had a broad, fairly hefty cutting blade, and I wanted something a little slimmer and more suited to fencing than a war sword, and I knew I wanted to incorporate stabilized burl into the lower half of the grip to add contrast to both color and shape.

 

I knew I wanted a multibar blade with a mosaic core in "explosion" pattern, so the first step was the steel. These pieces are 1.5" wide and 12" long. The final stack was over 3" tall and weighed something like 15lbs. I wouldn't have even attempted this without a hydraulic press.

 

 

 

 

The first weld is set and the surface ground to check everything welded properly. It all looks good, so full steam ahead.

 

 

 

 

 

Beginning to draw it out. A billet this size quickly outgrew my forge.

 

 

 

After the billet was drawn out to (if I remember correctly) 1.5" square, I cut it into 3 pieces and stacked it up with the grain rather than against to start the crushed Ws pattern.

 

 

 

This stack is welded and drawn out as before, and I check the pattern on a cutoff end.

 

 

I don't know why this picture is upside down, but you get the idea.

 

 

More drawing out.

 

 

 

Now I cut the bar in half and stack the pieces as before to get more Ws.

 

 

 

The bar is forged on a 45-degree bias, and this is the result. The pattern shifts so that it comes from the corner rather than the bottom edge.

 

 

 

And some more forging out to reduce the bar to 1" square.

 

 

 

Next up is the making of the mosaic.

 

-A.J.

 

Edited February 21, 2022 by AJ Chalifoux
Formatting.

[AJ Chalifoux]

The bar is cut into 4 equal pieces and the ends etched lightly to show the pattern. They are then arranged into a star. There are a lot of different bar orientations possible since either end of each individual bar can be used and any piece can be put in a different position, so I tried to find the orientation with the most symmetrical pattern where the bars still fit tightly together with no big gaps/misalignment.

 

One thing to remember is it's very easy to forge the bar's cross-section into a parallelogram instead of a square, so getting them as square as possible in the forge and arranging the mosaic bar with as few gaps as possible is important. This of course would have been a lot easier with squaring dies, but I didn't have a set of those made yet. I got the bar close enough without squaring dies, so I'll add that to my future projects list.

 

 

The billet is then welded and drawn out.

 

 

After drawing out, it's time to lay out the accordion fold. You can do the fold as either big triangle cutouts or narrow slit cutouts. The narrow slits waste far less material, but I wanted a bolder, more dramatic and drawn-out pattern so I opted for the triangular cutouts.

 

 

Each corner has a hole drilled.

 

 

The web of material is then cut out. The bar is ~3/8" thick at this point. Note the end of the bar is kept square. This section will become the tang.

 

 

The bar is then straightened slowly and carefully at a welding heat. I open up the accordions a bit with a hand hammer first, then put them in the press to get it all flat and even. Then I grind the surface and etch quickly to take a peek at the pattern.

 

 

Now to start on the edge bars. For these, I am going for medium-layer random pattern damascus on either side of a 15N20 core. I don't want my final (sharpened) edge to be too obvious, so I don't want the pattern going all the way to the edge. 30 layers total to start with.

 

 

 

 

Once the bar is welded and drawn out, I cut it into 6 equal bars and tack weld them into two sets of 3 bars each.

 

 

These are then welded into two bars that are very narrow and very thin (1/2" wide and 1/4" thick). This takes a very, very long time as presses don't like to squish material that thin .

 

I then cut both bars in half and use them to sandwich a core layer of 15N20.

 

 

 

Once these are welded and shaped to 1/2" X 3/8" rectangles, it's time to weld them to the core bar. Here is everything tacked together awaiting forging:

 

 

Here it is all welded up. I didn't get a picture of it, but for the tip I cut a V out of the front and forged the two prongs together and welded them.

 

 

After some more drawing out to get the right shape and approximate thickness, this is what the forged blade looks like. I did have one delamination near the middle of the blade while forging it. Luckily, I was able to clean it up with only a slight narrow section. This wasn't really an issue at the end, because it just meant that the profile taper of the final sword has more of a concave curve rather than a straight taper, which is not ahistorical.

 

 

The surface is rough-ground to check for cracks or voids, and I check the pattern while I'm at it. You can see a bit of scale left on the flat from where the delamination occurred during forging. Since the bevels will be ground in, this isn't an issue.

 

 

 

 

Next up will be the grinding of the blade including distal taper, bevels, and fullers.

 

-A.J.

[Joshua States]

What an awesome project! How long is that blade? It looks about 38 inches?

[AJ Chalifoux]

Thanks! The blade is 40" and the overall length of the finished sword is 51-7/8".

[AJ Chalifoux]

Next, I marked the blade at 2" intervals and measure the current thickness. Then I decide on the desired final thickness at each point based on both blade feel and historical examples, and grind accordingly. This allows me to creep up on the desired thickness and check as I go so I don't end up with thick (or thin) spots.

 

 

 

Then I check the pattern in addition to the distal taper. This is the part that went wrong for me. The blade was originally slightly thicker than I wanted, and I noticed the mosaic pattern on one side was less pronounced than on the other. So I ground a bit heavier on that side to try and correct it and get the pattern even. I ended up grinding it a bit thinner than I wanted, and while the mosaic patterns are now fairly even, the outer random pattern on the edges was nearly ground away on the middle section of one side (visible in the 3rd-to-last picture on this post). Moral of the story: etch and double-check the thickness of the outer layers on any san-mai before grinding.

 

 

 

The blade is then normalized and straightened by clamping between two boards.

 

 

Now for the fullers. I got a pneumatic pencil-type die grinder for about $35 and it was one of the best purchases I've ever made. First I mark the fullers, then I use a carbide burr to rough the centerlines. Finally, I grind them wider and even them out using a sanding drum.

 

 

 

 

Now coat the blade in layout fluid and mark the bevels, then rough grind and file them to shape. Grinding the bevels on san-mai means one has to be careful that the edge remains centered for the length of the blade.

 

 

 

 

 

At this point the blade is heat treated. I don't have a kiln wired up yet, and I did not trust my usual methods for a sword such as this, so I sent it out for heat treat.

 

Here is the heat treated blade next to the final design.

 

 

Next up: hilt components.

 

-A.J.

 

[Alan Longmire]

Fairly awe-inspiring project!  Thanks for the detailed WIP. 

[AJ Chalifoux]

The original guard used a thin sheet socket welded to the main body of the guard. I tried to duplicate that, but rather than weld, I opted to try and solder it. This was because 1. I thought I could get a cleaner joint, 2. I wasn't confident enough in my forge welding ability to attempt this, and 3. I certainly wasn't comfortable enough in my regular welding ability to attempt this on thin stock. Here is the guard body shaped and ready for the socket:

 

 

And here is the socket fitted around said guard.

 

 

This was my first idea, and I spent about 6 months (yes, 6 months) trying to get it right. I had to have made the socket at least a dozen times, and tried to do this dozens of different ways, each with their own plusses and minuses. Eventually I got a good, clean joint around the entire perimeter...which immediately popped open the instant I put it in a vise. I realized that without welding it, the socket idea wasn't going to work. At this point I decided to do what I'm most comfortable with and mill it from one big piece of mild steel.

 

I got a plate 1" X 6" X 12" and cut the whole thing out, milled all the sides, and even angled the milling head and milled the angles into the socket body. Then I went to drill a 3/4" hole into the handle (at this point the guard was milled down to 7/8" thick) so it was still hollow, and something in my brain said "yeah, my drill press is accurate enough for this." It was not. The bit walked and poked a hole in the side of the grip that I couldn't weld shut. So I started over yet again, with another plate, and this time I drilled the hole earlier, and used both the milling machine and a 11/16" drill bit and that all worked. It's a bit heftier than it would have been with the socket, but only by maybe an ounce or so.

 

 

 

Once the guard was shaped, I cut, drilled, and filed the brass plates for the quatrefoils. They were then soldered in place.

 

 

 

After some smoothening with the random orbital sander, the guard is almost polished. I hadn't used a random orbital sander before this, but I wouldn't be without one now. It's amazing the finish you can get with them. Several times over the course of this project, I took a piece as-filed and sanded it using a 220-grit disc and it took all the scratches right out. At that point I could easily move onto 400 or 600 grit hand sanding, or use a finer grit disc and bring it right to the buffer.

 

 

I then made the brass spacer to go between the grip halves.

 

 

However, this one was too small, so I made it again a bit bigger (and more true to the drawing).

 

 

Much better. The spacer also has a 3/4" hole in it to fit a wood core to reduce weight, and everything is kept from rotating via pins. To get the holes for the pins half in the wood core and half in the brass, I used an end mill rather than a drill bit, which worked well. Just make sure to go slow or the end mill can also walk.

 

 

Next up is shaping the lower half of the grip.

 

-A.J.

 

[Bjorn Gylfason]

That looks absolutely amazing

[Joshua States]

I think it's Geoff Keyes that has a quote in his signature that goes something like "Just because it worked should not be taken as proof that you are not insane".

My brain hurts just thinking about working a piece of 1 inch plate, let alone doing it twice.......

[AJ Chalifoux]

Thanks for the kind words!

 

For the lower grip, I knew I wanted a lighter wood to fit with the rest of the hilt, but one that had a busier grain to connect the hilt to the blade. I bought a few pieces, and eventually settled on this stabilized box elder burl.

 

 

The piece was then cut to length and the slot was milled and filed, and the whole thing rough-shaped on the belt grinder.

 

 

I then added the layout lines for the spiral flutes.

 

 

I then cut them in using a triangular file and a regular flat file. These needed a good amount of cleanup after this picture, but needle files worked very well for refining all of the roping. I also made the bottom brass spacer in tandem with the grip.

 

 

Now it's time to start the pommel. The main body was drilled, milled, cut, and shaped from one piece of 1/2" flat stock. Both sections of the pommel are hollow (as they likely would have been on the original) so the larger section has a 1-1/4" hole through the middle, and the smaller section has a 1" hole through the middle.

 

 

 

 

The brass medallions are then cut to fit on the smaller section.

 

 

Now it's time to start on the bosses for the larger section. Since I don't have a swage block, I took a large (~2-1/2" diameter) steel rod and ground one end into a dome. Then I took some 3/32" thick steel sheet and forged them around said dome.

 

 

 

 

 

 

I cut the pieces out, fitted them to the pommel, and then cleaned them up on the grinder. I also drilled a small hole in the very tops where I center punched earlier. This is so I can pin the brass rosettes. I thought about just soldering them, but I decided pinning them would still look good and give them a much stronger mechanical hold. 

 

After peening the rosettes in place, The bosses were welded to the main pommel body and the whole thing cleaned up.

 

 

 

 

 

The grip is nearing completion. Here you can see some shims next to the sword which I use to take up any extra space in any of the slotted hilt components so everything is rattle-free. These aren't technically shim stock — they're 1/2" wide feeler gauges which come in rolls in any thickness you could want. They're very handy for tightening up handles.

 

 

The upper grip's wood core halves are made and I drill some holes for alignment pins.

 

 

 

Next up: polishing the blade.

 

-A.J.

 

[AJ Chalifoux]

Just now, Joshua States said:

I think it's Geoff Keyes that has a quote in his signature that goes something like "Just because it worked should not be taken as proof that you are not insane".

My brain hurts just thinking about working a piece of 1 inch plate, let alone doing it twice.......

 The hardest/most tedious part was cutting through it with the angle grinder to get the big chunks taken out. Once that was done, it was kind of fun to think about how to keep it all as rigid as possible in the milling vise for as long as possible (those quatrefoils are only 1/4" thick). Then again, I enjoy machining things so I didn't mind too much.

[Joshua States]

1 minute ago, AJ Chalifoux said:

those quatrefoils are only 1/4" thick

Something about this statement seems weird to me. Maybe it's the use of "only" with "1/4" thick"

Great stuff here.

[AJ Chalifoux]

With the hilt mostly shaped, it's time to start polishing the blade. This was done using, you guessed it, a random orbital sander and fine files. The files were really just to refine the edges to take them down from pre-hardening thickness to pre-sharpening thickness, or from about the thickness of a dime to about the thickness of a stout butter knife.

 

 

At this time I etch my signature and then finish sand and buff the blade.

 

 

 

Quick fit-up. I also make the peen block at this point. The below picture actually shows the blade partially-etched, though that was finished on a different day.

 

 

Now it's time to etch the blade. Multiple soaks for 15 minutes, followed by scrubbing the oxides off.

 

 

 

 

After etching, I coffee darken it. I actually scrubbed and buffed the original darkening off because it just came out too dark. I then redid it with bright, shiny 15N20 and the contrast was much better. The blade is, at this point, very difficult to photograph, however...

 

 

The hilt components and base of the blade are taped to try and keep them clean while everything is epoxied and the pommel peened. After this, the only things left to do were final cleanup and sharpening.

 

 

Next up: the finished sword!

 

-A.J.

 

[AJ Chalifoux]

And here is the finished sword. I can't quite stress enough how the pictures do not do it justice because the hilt is very light and the blade is very dark so no matter what I did, the contrast always looks off. I'm tempted to get it professionally photographed, but for now this is the best I can capture it.

 

 

 

 

 

 

 

 

 

 

 

 

Here are the overall stats for the finished sword:

 

- Overall Length: 51.875" (131.8cm)
- Blade Length: 40.0" (101.6cm)
- Blade Width at Base: 1.389" (3.527cm)
- Blade Thickness at base: 0.260" (0.660cm)
- Blade Width at Tip: 0.924" (2.346cm)
- Blade Thickness at Tip: 0.080" (0.202cm)
- Guard Width: 9.750" (24.765cm)
- Grip Length: 7.750" (19.685cm)
- Guard/Grip Thickness: 0.874" (2.220cm)
- Pommel Length (Bottom Section): 1.852" (4.705cm)
- Pommel Diameter (Bottom Section): 1.940" (4.928cm)
- Pommel Thickness (Bottom Section): 1.433" (3.641cm)
- Pommel Length (Top Section): 1.496" (3.800cm)
- Pommel Diameter (Top Section): 1.512" (3.841cm)
- Pommel Thickness (Top Section): 0.620" (1.576cm)
- Weight: 3lbs 8.2oz (1.593kg)
- Center of Gravity: 3.008" (7.640cm)
- Primary Node (Center of Percussion): 25" (63.5cm) forward from guard
- Secondary Node: 1.630" (4.140cm) back from guard
- Forward Pivot Point: At point
- Aft Pivot Point: 7" (17.8cm) forward of guard

 

Thank you all for watching!

 

-A.J.
 

[Francis Gastellu]

This is incredible! Thank you so much for posting so many details on this build, it is fascinating to follow and I know I'll be coming back to this thread in the future for inspiration.

 

She is absolutely gorgeous.

[Brian Dougherty]

Fantastic build.  I'm about to embark on a project of my own that will likely take me longer than a year, so I've found your WIP to be inspiring.

Your fiancé must be quite the catch.  I hope your passion for each other continues to burn this brightly for many years

 

[Alan Longmire]

I should have pinned this earlier, but it's never too late!  Excellent work!

 

One question:  When you were trying to make the socket via soldering, what solder were you trying?  I've had good results with silver brazing for that sort of thing.  It will not split or break.  It does tend to show once the steel darkens a bit from age, though.

[AJ Chalifoux]

20 minutes ago, Alan Longmire said:

I should have pinned this earlier, but it's never too late!  Excellent work!

 

One question:  When you were trying to make the socket via soldering, what solder were you trying?  I've had good results with silver brazing for that sort of thing.  It will not split or break.  It does tend to show once the steel darkens a bit from age, though.

Thanks for the pin!

 

I did just try regular lead-free solder. I had some higher-temp stuff I also tried briefly, but didn't have too much luck though I admit I didn't play with it for too long. Part of me was worried that the collar would heat up much faster than the guard body and I would end up with a dirty joint if I went too hot too fast.

[Alan Longmire]

Understandable.  Should you want to play with this in the future, get some Harris Safety-Silv 56 and an appropriate flux, like Handi-flux.  You'll need an acetylene or oxy-propane torch to get enough heat on stuff as thick as the guard.  I've never gotten a dirty joint with it, but it's easy to flow too much metal, which results in an annoying cleanup.  And as I said, the joint will always show once things oxidize over time.  Your solid-block version is superior in that respect!  

[AJ Chalifoux]

I'll have to remember that. I'd like to try something similar again one day, and I would much rather silver solder than weld it, especially on thin stock.

[Don Abbott]

That was an incredible journey.

 

Thanks for sharing it.

 

 

[Pieter-Paul Derks]

Incredible work! that turned out a real lovely sword, I love that original as well.

getting that guard out of that poiece of steel is almost like sculpture.

[Doug Lester]

I'm still trying to get my head around what went into making that sword.  The handle is incredible.

 

Doug

[AJ Chalifoux]

Thank you all for the kind words! The handle took a lot of planning, and the guard, pommel, top spacer, and peen block all had to be done at least twice, if not more.

 

It took at least a dozen or so tries to get the guard into a shape I was happy with, and the pommel very nearly posed a similar challenge. Getting a straight slot through that long of a pommel that had a tight fit, no large gaps, and wasn't corkscrewed with the plane of the blade was difficult.

 

[peter fontenla]

Impressive, what a good display of skills, I loved it!!!!! thanks for sharing