How I made my fiancée's engagement ring

8/5/2013

After nearly seven years as a professional goldsmith and metal arts teacher, it was time to make an engagement ring for my own girlfriend.

I started out with a budget - in dollars and time. Mass produced engagement rings can take under one hour of labor. When I teach engagement ring making classes, a basic solitaire takes 10-12 hours. The most complicated engagement ring a customer has made took 40 hours. I figured I could spend up to 100.

I started with a little research about what other DIY romantics who have crafted their own engagement rings made. I was impressed. People with almost no prior experience made beautiful rings, some creative rings, and the documentation was excellent.

Of course, I have the tools and know-how to make an engagement ring, and had already come up with a few ideas that were exciting to me. I put my thinking cap on for a full 50 hours of research before getting my hands dirty - and stopping occasionally to take a photo. Here is how I made my fiancée's engagement ring:

Part I: Refining

Tree of Life brooches are coming back in style

Here is an example of a tree-of life brooch, which my girlfriend had inherited from her
grandmother. I had invited her to come make some stacking rings in my workshop a
couple years ago, and she brought this, hoping to use it for those rings.

Unfortunately, there are many alloys of gold. This one was made specifically to flow
into the twists and knurls of a 70's era tree of life, and not for milling into a ring.

As I roll it further, the cracks become worse.

Now it's a flower.

Here's our cracked gold ingot again. I will now quarter the gold: mix it with alloy
to bring the percentage of gold below 25%. This is necessary for refining later.

The gold and silver are melted in a crucible.

The ingot is hammered to flatten and strengthen it a bit.

The resulting mostly-silver ingot is soft and ready to be rolled into a thin strip.

Sam Abbay holding very thinly rolled quartered gold

The rolling mill (far left), rolls it thinner and longer, increasing the surface
area around 40x. The greater surface area will speed up the reaction.

The thin strip of rolled gold in a beaker. Nitric acid will not dissolve
unwanted metals if the gold content is too high.

We have a fierce reaction shortly after the nitric acid is added. The nitric will
dissolve silver, copper, and most unwanted metals. It will not dissolve gold.

The result is flakes of high karat gold, with a tiny bit of silver that the nitric
couldn't reach. This is placed in aqua regia, which will dissolve gold but not silver.

The aqua regia contains Grandma's gold. A small amount of silver chloride can
be seen at the bottom of the larger beaker. Please do not try this at home.

The gold is precipitated out with ferrous sulfate.

Pure gold settles to the bottom of the beaker.

The gold precipitate is melted together into nuggets.

After more hours than I would like to admit, three beautiful former-tree-of-life
nuggets of 24k gold are ready to be alloyed into my fiancée's engagement rings.

Part II: Plasma

This is a plasma arc speaker. The purple plasma in the photo is actually playing music.
I know it is hard to imagine if you haven't seen one live, and any video you see would be playing
from your computer speakers. In person, it is amazing. Due to the expense
and inefficiency of the technology, it is an obscure novelty for those into electronics and music.

Precious metal alloys for testing in the speaker

Most plasma speakers arc off of a copper surface. I tested a couple dozen alloys to
create boutique jewelry metals that would be feasible within the system.
Several melted. Others blew out speaker components. A lot of speaker components.

Here is a video of one of those such tests. Mouse over and click play. Much is
lost when the sound is playing from your computer speakers rather than
the plasma arc. Sorry, the video will not play on an iPad or iPhone.

I draw some of grandma's gold engagement ring alloy into wire for testing.

A tiny test ring fits in the speaker...

Now I test a demo engagement ring with a Moissanite stone, which does not crack from the heat. If the
Moissanite can withstand the heat, a diamond should be able to. I am now ready to make a
singing engagement ring.

This would not have been possible without the inspiration and support
from Andrew Zwicker and Princeton Plasma Physics Labs. Thank you!

Part III: Construction

Remember the gold? Time to melt, pour an ingot, and mill it into the parts I will
need to make a couple engagement rings.

Here is a square gold rod, some bezel strip, and a second bezel that I have
already formed.

A special alloy of gold designed to melt at a lower temperature is placed over the seam.

The ring is heated until the solder melts. Capillary action draws the liquid gold
into the seam to create a strong joint.

Gold filings are everywhere as the gold ring is shaped. The ring will now be cut open
and the bezel will be spliced into the shank.

The head is now soldered on to form a complete ring mounting.

Here is the centerpiece - an antique Old European cut diamond. This cut features
fewer, larger facets, a taller crown, smaller table, and a visible culet. The Old European
Cut is less scintillant than a modern brilliant without being subdued.
More info about diamonds here.

Sam Abbay setting a diamond using a microscope

Using a microscope, I carefully fold the gold bezel over the diamonds.

Part IV: Finished!

Old European Cut Diamond Engagement Ring

Stacking engagement rings! One antique Old European cut diamond and one
round brilliant white (with a faint blue tinge) lab-created diamond in yellow gold.

The finished diamond ring as the substrate for the plasma speaker. I recorded a
proposal and played it for my girlfriend, along with one of our favorite songs.

The End!

Part I: Refining

Part II: Plasma

Content on this page requires a newer version of Adobe Flash Player.

Part III: Construction

Part IV: Finished!