September Project: Roman Earrings

This month has a lot going on, so I wanted a quick project. I didn't expect it to be this quick! I finished these in an afternoon. I'll probably make a second pair. I want to try harder to get all the links to close up completely, even if I have to solder them separately. They're amazingly tiny and fiddly, made with 20 ga fine silver. The hooks are the same wire, and the beads are wired on with 24 ga. Next time, I'm going to see if I have any 22 ga. I think it'll be better. Getting the twists around was tricky, and i think that shows. The plate is 26 ga, and I think it's perfect. The originals are at the British Museum, 3rd century Roman in origin.

A few progress photos: 

August Project: Revisiting a Favorite

I wanted to make another version of the Byzantine necklace from the Met collection. I've given away the other two I've made (and in all honesty, I'm making several now so that I can give more away). This may be my favorite piece to gift. Here's the original (photos from the Met website).

I approached different versions (this time around) different ways. I wanted to try 18 gauge wire for the inner "heart" shapes, as opposed to 20, which I've used before. The first one, I made up the hearts one by one, to fit my drawing. This was a mistake. It didn't come out terribly, but it lacked consistency and one of my clasp pieces is decidedly oblong. It's not the absolute worst thing in the world, but it bothers me. So for my second version, I went back to 20 gauge wire (fine silver) and actually measured, going in stages so that I ended up with precisely measured V shapes that fit together nicely. Then I formed the hearts, and I'm much happier with the finished pieces. Finally, not one to give up, I followed the same method with the 18 gauge wire (also fine silver), and achieved a much happier result. 

And here's the 18 gauge version. 

I feel like the 20 gauge is more true to the original proportion, but the 18 works. Here are the two pairs side by side after soldering. 

The next step is to form the beaded wire frame. 

Following that, I added the loops for attachment, the bail, and the hook. This is the hardest part. These attachments definitely aren't as clean as I'd like on all the pieces. See that little bead at the bottom of the second one? Yeah, that was a hiccup. I can't get it off (I tried) so that'll become a hook end. 

And the hooks (what will become hooks). These were easier than the small bails. Still less than perfect, but I'm happy with them. They vary from the original. I tried doing something more like the original, but it didn't work. These little granules offer support and municipal the other side of the piece, so I rolled with it. 

My writeup from the last time really helped me with this step. Melting the piece at this point is heartbreaking, and I've done it before. Holding the piece the the third hand and placing the larger bail on the charcoal block really helps. I also tinned the larger bail this time, for two out of the three. Not a period practice, but in period, they would have had powdered solder, which is definitely a step closer to tinning than using chip solder.

They just need a little polish and they're ready to become necklaces! I'm calling the project here, for this month. The one that I freehanded came out quite a bit smaller than the ones I carefully measured. They were all based on the same sketch, so that's interesting. I may make a smaller sketch at some point, and purposefully make smaller ones, but I think the larger ones are the right size to match the original more closely. What is a happy thing is that while you can still see that one of the smaller ones is decidedly oval, it's not as bad now as it was before the bails went on. 

In the end, I definitely want to make another one of these, probably multiples if I'm being honest. The originals use square section wire, and have a substantially thicker bail. I use to have some half round wire that I used on a previous version, but I apparently ran out. So I used 16 gauge because it's what I had on hand, but it's woefully puny compared to the original. So yeah, I think I have the process down for making the components, I just want to get or make more accurate materials for the next time or three. 

Bonus Project-Anglo Saxon wire work

A friend asked me to make her ten of these little wire loop pendants for a recreation she's making of a grave find from Kent. She's inspired me to attempt my own full recreation of the necklace myself, at some point, when I get into making beads. (It'll happen eventually.)

But for these little wire pendants, she even sent me a video tutorial. I strayed from it a little bit, but I'm very happy with the end results. And I can't wait to see her finished piece.

They are more regular than the originals, and I'm torn about that.  But overall, they're the same shape and style. And it was a fun project for an evening.

Project 3!

Continuing my work with Hercules knots, this is my first interpretation of a common motif in necklaces. Most examples I've found have been from auction sites, though the Walters Museum has at least two in their collection, and the below example is from the British Museum. All the examples that I've seen have had emerald or otherwise green beads. It's sometimes listed as green agate. I really only trust the museums so much in correctly identifying ancient stones. One thing the museums and auction sites do seem to agree on is that most examples date from the 2nd and 3rd centuries CE.

This example below, which may or may not be at the British Museum (I found it on Pinterest, and reverse image search didn't bring up anything definitive. It could be a modern copy, but the bead quality and cut makes me doubt that), is what I based my knots on. Mine are definitely not perfect, but this was meant to be a quick project to get my feet wet again. Success! My feet are wet, and my goal is to remake this with better knots that are a little smaller. I think I also want to shoot more for the shape of the one above, with a completely closed neck. 

One final example from the British Museum, an individual link that's 2.5 cm long. This one has less refinement, and is more like what I accomplished. I need more practice. Though this is the size I'd like to aim for next. My current ones are more like 4 cm, end to end.

My links are 20 gauge fine silver (all examples are gold), soldered into loops that were further formed into shape using a ring mandrel, round nose pliers, and my fingers. Each knot is soldered at 4 points. I chose not to solder the center points. I probably should have, but I melted quite a few of these, and decided to call them done with 4 points of contact connected. I would probably use 18 gauge wire next time. 

Loops! 

Rough forming. 

The shape is kind of like a club, or almost a ping-pong paddle. I want to get a more defined neck with a more circular smaller end next time. 

Finished knots

All the failures. I love saving my failures! These will eventually have the solder cut off, and then probably become granules. The rings may be salvaged for smaller rings or filigree. 

First (and second) Project of July

This project sort of fell into my lap, happily, when a friend asked me to work on a collaboration. Khalja supplied three fine silver coins, struck by Arion Wanderer, and asked me to make earrings and a pendant with them. She knew exactly what she wanted them to look like, and I knew what she wanted. They were a gift for her new Majesty Brynhildr of An Tir, SCA.

The earrings are set in a frame of fine silver wire, with prongs soldered on. This is a modern design that's commonly found using old Roman coins. I used modern silver ear wires, but in hindsight, I wish I'd just made my own. If I'd had hard silver wire, I would have.

The pendant, on the other hand, is a medieval as they come. A very simple design, the bail is a piece of fine silver sheet that I cut and formed over, and then it was riveted to the coin. Examples of riveted coins are all over the Viking era, and they're easy to find. What I haven't researched enough is if they're regional to anywhere specific, or if it was a more generally diffused practice. There are definitely examples from Gotland. Again in hindsight, I wish I'd don't some stamping on the bail before forming it. That's not as common a practice based on examples, but I found at least one. If I get to do this again, I'll do that. 

The little pin rivet was made from 18 gauge fine silver wire. I melted a little head on one end and peened the back side flat. The head flattened a little bit, but that's okay. It still stands up like a little button. This is truly so much fun to do! The biggest trick was getting the head melted on the wire. I started out holding it in pliers, which were too much of a heat sink. Next time I'll just wait until I've melted the head to cut it, so I can just hold it further down, in my hand. That's how I've done this before, for Tilda's pendant.

Back at it

First time back in the shop in two years, and I'm pretty pleased. I spent yesterday sketching projects I want to try this year, and today I finally lit the torch. And I started none of those projects. I want to knock the dust off, literally and figuratively. So basic soldering, shaping, a little pick soldering. The latter isn't, to the best of my knowledge, period, but until I make my own solder powder, it's what I sometimes need to do with chips. 

Making my own solder is in the cards for this year, I just need a graphite rod, some powdered borax, and a crucible. I have the raw materials already. I'm a little nervous about soldering with just an alloy of silver and copper, but I'm trusting the process. Worst case scenario, after I try it a few times, I can melt it back down and add some zinc if I don't like it as is. I'm really game to try it, though. It worked for medieval goldsmiths.

For now, these are imperfect, but I'm not trying for perfection. I'm just working on progress. (These are also desperately in need of cleaning, and that's going to be a job for the flex shaft, which needs an extension cord.)

Starting over

I have barely made anything since the pandemic. I stress wrote a book when I got laid off, then we moved and I've gone from one high intensity job to an even higher intensity job. But I've finally got the inspiration back to get back out in the shop. I suffered a few failures when I tried shifting to larger projects, and that was discouraging. I'm over it! 

My goal is to make one object a month from July 2024 through June 2025. A minimum object is a soldered bronze ring (my very first soldering exercise). At the absolute least, I'll make one of those. But I have tons of materials sitting on my bench just being to be made into something. I bought butane today, and I have a lead on what's wrong with my oxy-propane torch. 

The struggles with that torch have not helped my artistic journey, but I'm convinced it's just regulator settings.

On a happy note, let me share a project that I made for a friend's elevation to the Order of the Laurel back before we moved. This might be the last thing I made before taking my break. I'm not 100% thrilled with how it came out, but it was my first real adventure with pin style rivets, making them and setting them, and that came out well enough. And the Laurel wreath broke, but I "fixed" it by doing the wire overlay, which I like better anyway. The inspiration piece had more depth to the settings, and I'd like to figure out how that was accomplished. Those pie pan shaped bezel settings are still boggling my mind. Anyway, photos! 

A Period Inspired Gift

My wonderful husband turned 41 yesterday. In the SCA, his persona is 10th century Irish. I've spent a LOT of time looking at jewelry and metalwork from Ireland and England, and this piece is definitely not a recreation. It should honestly feature enamel in those three small circles, but I'm not set up to do enamel. And the stamping is largely a nod to my love's own metalwork. I didn't take a ton of process photos, and in all honesty, I made up the design on the fly. The only thing I had set in my mind was the central 10mm stone. 

The outer border is 2x1 half round fine silver and .5mm beaded sterling silver. The filigree is 20 gauge fine silver, and the back plate is 24 gauge fine silver. The stamping was accomplished with a spring punch, and in hindsight, I wish that I'd used a hammer and a steel punch instead. The spring action was very heavy, and it translated through to the back. It also slightly contorted the shape of the entire pendant, so that it is slightly dished in the center. Overall, though, none of this negatively affects the aesthetics. One final note: I used silver solder for the filigree, because it's what I'm familiar with. However, I ended up with some marks that I couldn't easily sand down. So I ended up texturing the entire back plate with a small ball burr in a flex shaft tool. Absolutely, thoroughly modern, but I think it was a happy accident, as it makes the filigree stand out nicely.

A Different Frankish Ring

A friend of mine commissioned a custom ring for his lady. After talking a little bit, I sent him a photo of this one from The Met, and asked if something similar (with his stone preferences, and in silver) would work, and he said yes! So I got to work. We had discussed using a 6mm stone for the center, and two 4mm stones for the sides, so I made those bezels, and then used 24 gauge silver wire to make the twisted wire detail. Four granules done, and I was ready to lay out all the pieces on a back plate. But wait! This setting ended up too large for a ring like the photo. I should have paid more attention to the original dimensions. 6mm is roughly equal to 1/4". Which is the overall width of the original bezel. Erego, the original center stone must have been closer to 4mm, with the side stones at 3 or 3.5mm.

Progress pics:

Since I already had the stones in hand, I sent my friend a progress photo and explained the issue. I gave him a few options, and he chose to just turn the setting so that the lengthwise axis will run parallel to the finger, rather than perpendicular to it. Perfect! That's further from a replica of the original piece, but it's still in keeping with the design aesthetic of medieval jewelry.

I ended up needing to remake the setting because there was a gap between the first and second bezels that bothered me. The final setting still isn't perfect, but my friend okayed it. 

After soldering all the new pieces onto the back plate, and finishing it, I chose to use the same split shank attachment seen in the original Frankish ring I remade. Given the extra size of the setting, the additional surface area was desirable.

Setting the 4mm stones proved a bit of a challenge, especially because their bezels were on the tall side. I could have put a small ring of silver wire underneath each stone, to raise it up a bit, but I chose to just muscle through. Overall, all three of us are very pleased with the outcome! This is definitely a style I want to play with some more. 

Soldering Research Continued, and Some Future Plans

On first read, Theophilus' treatise "On Divers Arts" is on the complex side, and he includes a lot of steps for things that sometimes border on mysticism. The most famous example of nonsense is his assertion that the urine of an old, fasting goat, or alternatively, a small redheaded boy, is most effective for quenching metal in hardening. His information on soldering isn't quite that outlandish, but it is rather complex.

Let's start by trying to demystify his instructions on soldering for silver. He starts with a recipe for solder that's a simple 2:1 silver to copper alloy that stays consistent through Cellini's treatise on goldsmithing, and isn't too far off from modern solder. So nothing strange there.  Then he says to take argol (the sediment found in bottles wherein the best wine has been stored), wrap it in a cloth and burn it, blow away the ashes of the cloth, and then grind the burnt argol in a copper pot, together with salt and water to make a paste. He then instructs the goldsmith to spread this paste on his joints, and sprinkle shavings of the pre-made solder over it.

So "argol" is also called wine stone, or tartar, and is a crude form of potassium bitartrate. A relatively refined version of potassium bitartrate is found in baking: cream of tartar. Armed with this information, I have to wonder if ordinary cream of tartar, mixed with salt and water to make a paste, would act as a functional flux, when used with modern solder. So this is an experiment I intend to undertake in the very near future.

Secondly, let's look at his instructions on "Solder for Gold". Theophilus instructs the goldsmith to make lye using beechwood ash, and then to strain the lye water through the ash again, to thicken it. You boil this and reduce it by 1/3, mixing in with it some unspecified amount of soap and pig fat. Once cooled, he instructs us to strain it and store it in a specially described copper pot. Leaving that aside for the moment, he then tells us to take a flat, thin piece of copper, wet it, and then rub it with salt. Put the copper in a fire until it is red hot and then quench it in clean water. Continue this process, collecting the burnt copper oxides in the clean water until you have a quantity of such. Then pour off the water and allow the resulting powder to dry, and then grind it with an iron powder. Put the powder on coals and burn it again, and then grind it again. (This gets a bit repetitive.) Then add soap to it, mix it carefully, burn it again, and grind it again. At this point, pour the lye into the vessel with your copper powder and let it boil and reduce for "a long time". Once it's boiled and cooled again, pour it back into the copper vessel you were previously storing the lye in. This is its storage container. Theophilus tells us to add several pieces of copper to this container, and to stir them together with the liquid.

At no point in the instruction does he mention making a separate solder material. This resulting acidic soapy copper oxide is the only material he goes on to reference in soldering gold. The very helpful editor provides a note, at least in my Dover edition, explaining that this is a somewhat convoluted version of a process that was used for hundreds of years, and that Cellini describes a similar process, as well. The science is this: copper oxides are mixed with a paste flux, and applied to the metal joints. When heated, the copper converts to metallic copper, and, being in contact with the silver or gold, it immediately creates an alloy at the contact spot that's got a lower melting point. It creates a fusion of the materials without requiring the fine metal to get to its own melting point. Fusing at that point becomes less precise in nature, and does not result in the finely detailed filigree and granulation work that we see in extant examples. Fusing at the lower temperature created by the copper alloy preserves the delicate detail work.

Cellini's recipe for this is to use verdigris, mixed with sal ammoniac and borax. Which supports my assertion that the fundamental parts required to function in this way are copper oxides and flux. So in a slightly farther future experiment, I intend to acquire some copper oxide and sal ammoniac (I am not prepared to make my own lye at the moment), and see what I can accomplish!

Medieval Methods vs. Modern Methods: Soldering

Much like sewing, not a whole lot has changed in the last few millennia. Soldering has been a technique in use since early on in the process of metalworking. Give or take 5,000 years. And metal, like fabric, can be made into hundreds of thousands of different things, but the methods of making those things are fairly limited, and defined by the material's properties. Metal can melt. Metal can bend. Metal can harden and soften. Metal is reactive to some acids. Metal can be mechanically fastened to other materials. And metal can be melted and "fused" onto other metals or itself (welding), or a metal with a lower melting point can be used like glue to fuse two other pieces of metal. Which is soldering.

By its nature, metal has to be clean and free from oxidation in order for it to be properly soldered. Flux is what we use to prevent oxidation, and there are several different chemical forms of flux, with the simplest being a borax cone that gets combined with distilled water in an unglazed ceramic dish. Borax and its chemical properties with respect to metal were known and can be documented in medieval metalsmithing practices. Interestingly enough, while Theophilus of Edessa mentions the use of borax in "On Divers Arts" (12th century), he separately provides a recipe for flux to be used in soldering silver. It involves the resin left at the bottom of wine bottles. He also provides a recipe for soldering gold, using lye, salt, and copper, but mentions that it can be used to solder both gold and silver. Cellini, in the 16th century, simply indicates borax.

Heat must be used to melt, weld, fuse, or solder metal. Modernly, most people use torches, with working pieces placed upon heat-reflective fire safe surfaces. I, and many people, use a simple charcoal block. There are other options available, including ceramic honeycomb, vermiculite, or magnesia blocks. Now, specifically, I work on a dense charcoal block made specifically for this use, that sits on a vermiculite block, that sits on my desk. The vermiculite block is bigger, and acts as a secondary, heat-safe workspace for things to be set upon safely, and to act as a safety if sparks fly. Charcoal doesn't act as a heat sink. It reflects heat back on your piece, and it can, by its own nature, ignite and smolder, so working with it on a secondary heat-safe surface is important. Some people choose to douse or quench their charcoal blocks after use.

That said, charcoal is indispensible to the discussion about medieval metalwork. Charcoal fires can get as hot as 2110'F. In a forge, with oxygen added to it via bellows, a coal fire can get as hot as 3590'F. Charcoal fires are how medieval metalworking was accomplished. For reference, the melting point of silver is 1,763'F, the melting point of gold is 1,948'F, and the melting point of steel is 2500'F.

Theophilus describes building a forge on a bed of sand with charcoal, and then for a soldering process, placing the pieces to be soldered on the charcoal, supporting them with more coal if needs be, and then building up charcoal walls around the piece, so that it can be heated evenly all the way around. Gaps were left so that the piece could be observed through heating. The coal fire was fed oxygen with bellows until it became hot enough for the solder to flow, and then the charcoal was removed and the piece was taken off the heat.

Ultimately, while a few of the tools have gotten better, and the visual aids and artificial lighting we have modernly make fine detailed work easier, the actual processes of metalworking have been virtually unchanged for millennia, which is pretty danged cool, if you ask me.

Byzantine Necklace Take 2: What Worked Better

So I gave the first Byzantine necklace to a charity fundraiser, but since I made it once, I wanted to try making it again, better, to submit to a competition. I know a little bit more now, and wanted to try a few things in a more refined way.

I made the central motifs in a slightly simpler way, but I still joined them together first. I was also more attentive to how the points of contact worked with the outer rings. When joining them to the outer rings, I had eight individual points of contact that needed solder. I used easy solder this time, and that ended up being an okay choice. Tiny bits of solder will melt and ball up just like any other metal. I balled up my tiny bits of solder before placing them at the joints. For some reason, probably because of the beaded wire's grooves, these were much easier to place, and much more stable.

After torching:

The other thing I did differently was when attaching the bail, and unfortunately I don't have a photo of this. But I reversed what was being held in the third hand. I put the bail on the charcoal block and held the filigree motif in the third hand. The third hand acted like a heat sink while the charcoal reflected heat back on the bail, letting it heat up faster and more thoroughly. This made the operation much smoother than the other way around, since the filigree is going to heat to temp much faster than the bail.

Overall, this one was much easier done.