Warranty Void if Removed

My second stone setting project (well, shell I suppose) was a pair of silver cufflinks bezel set with Blue Paua shell from New Zealand. It's also known as Blue Abalone.

The shell is cut and polished into 18x13 oval cabochons, available from Kernowcraft. I buy my jewellery materials from Kernowcraft and Cookson Gold. Kernowcraft seem to stock a slightly wider range of stones, but have more of a crafty vibe than Cookson, not selling the more specialised jewellery tools for silver and goldsmithing.

I designed the pieces in inkscape and laser printed stencils. I stuck them to silver sheet and pierced them out with a jeweller's saw and a medium blade. They consist of five main silver sheet parts - a bezel base and wall, the curved stem, the knuckle of the hinge, and the button. The button, stem and knuckle is cut from 1.5mm sheet, the bezel base from 0.9mm and the bezel wall from 0.4x4 strip.

Here's the remains of the stencil with the parts cut out. Note that those parts did not all come from that sheet, that's just the 1.5mm sheet in green.

The parts were cut oversize and carefully filed down. The button was given a nice, deep curved bevel on one face. The knuckle and stem hinge were very carefully filed to give a perfect fit. Everything was kept square by using the nice square reference surfaces of the parallel pliers. They are a prerequisite for this project, it really wouldn't have been possible without. The hinge part on the stem was filed to a nice semicircle. The knuckle was left square at this stage. The hinge pieces were left a slightly tight fit to allow for polishing later,

 

The stem hinge part was then drilled with a 0.6mm bit. I used a carbide bit held in my fingers, but snapped a couple. I'd recommend HSS instead. Since a 0.6mm hole in a 1.5mm sheet leaves less than 0.5mm of material each side, it's important to get it in the middle. I marked it and "centre punched" it by eye and hand with a needle under the microscope. If you don't have some kind of indexable drilling table, this is about the best bet. It's also very important to get it square. I did it carefully by eye. The holes in the stem were very carefully countersunk to allow the rivet ends to mushroom out. This was done by hand with a conical burr under the microscope.

The knuckle was drilled the same way. The longer hole in the knuckle ended up a tight fit for the 0.6mm wire rivet that would form the pivot, so I had to enlarge it. Using a larger drill bit would have been suicide. In a sticky metal like silver, it would have got stuck and snapped immediately without a very high speed drill. I didn't want to use the PCB drill, as I knew the razor sharp bit would pay no attention to the tiny needle dimple if it was held in a rigid chuck. A broach would be the correct tool, but I don't have any, so I decided to carefully stretch the hole with a needle. This worked very well, surprisingly.

Here's the hinge test assembled around a piece of 0.6mm silver wire. 

I then carefully filed the knuckle to a semicircle to match the stem hinge part. Here it is half filed

The finished knuckle. It really is quite small (2.5x1.5x2). It's probably the most precise work I've done with hand tools alone

 

I've lied slightly, that's actually a test knuckle I made for this test hinge. I made the real knuckles at opposite ends of a single piece of silver about 6mm long, then chopped them off when all the drilling and filing was complete. This was much more sensible as I didn't have to constantly be holding a tiny speck in the parallel pliers. For some reason I forgot to take photos of that stage.

I then marked the bezel base for attachment to the stem. I scribed lines with the help of the step measurement of my caliper

I made a little structure from binding wire to hold the stem to the base for soldering. Note the U shaped bends. They add some flex to allow for the different expansion coefficients of the silver and the steel wire. Without them the wire can sometimes bite into the silver surface, or distort very small parts. 

Soldering the stem. Note silver solder will not fill holes like electronics / plumbing solder. The end of the stem was filed to a right angle so that it sat square to the base. A non-square end held square to the base would have resulted in a poor joint. The little brackets are standoffs to allow rapid heating. Silver solder joints should be made rapidly, otherwise the solder goes all weird and pitted, and in extreme cases seems to actually eat the silver. 

Next the bezel strip was formed and measured around the base. This shot gives a nice view of the countersinks for the rivet.

The bezel strip was formed into a rough circle and the ends were soldered. Flux is being applied here. There is more detail of mixing flux, and some other things in the flickr set. The snapped drill bits are for support. 

The soldered bezel wall was then rounded on the mandrel and test fited on the base. The base was filed until the parts fitted perfectly.

The bezel wall was then soldered to the base with medium solder, so as to not melt the hard solder holding the base to the stem. Here it is with flux and pallions of solder, ready for heat.

The soldered bezel. 

The bezel was then roughly polished with 1200 grit wet & dry and  silicone polishing wheels until the solder join disappeared.

One half of the assembly was then done, leaving the knuckle to be attached to the button. A similar marking out process and holding arrangement was used

The button was hard soldered and pre-polished in the same way as the stem. For some reason, I didn't take any photos. The next step was stone setting. Here it is part way through. Note how opposite sides of the wall are pushed over using the bezel rocker tool. The "long" sides were done first, then the "short" ones, then the "corners" and so on, as far as those terms apply to ellipses. 

The setting was then burnished with a pointed burnishing tool, and the finished setting was pre-polished.

Finally the two halves were riveted together. A pin was cut from 0.6mm silver wire, and annealed to soften it for riveting. The riveting was done against a steel slab. Everything was held perpendicular and the correct height from the slab using blu(well, white)-tack. A 3mm steel rod was used as the drift and it was driven with a tack hammer. This was scary, but it all worked out and the hinge was nice and free.

The final polish was then applied. The finished item: 

I'm very pleased with the results, the hinges represent my most precise work with hand tools so far.