Back to Ganoksin | FAQ | Contact

Coefficients of thermal expansion


I’m designing a piece which is basically a cast bezel to be fitted
around the carved stone by shrinking the stone by cooling (i.e. chill
the stone using dry ice to shrink it small enough to fit within the
retaining edge of the bezel, when warmed the stone and bezel should be
a precise fit.)

I was looking for on the amount of expansion/contraction
for various stones in order to determine how much depth to allow (so
as to permit the stone when chilled to pass through the bezel…)

Anyone done anything like this? Anyone know where I can find the
data? I can, (and will, if I must) do it empirically, but I was
hopping to get a rough idea before designing the piece…

Thank you all in advance.
Charles Barnard


This is truly an ambitious project. We deal with the CTE rate of
materials often in my work. I don’t know the CTE of stone nor of the
bezel. My experience is from another trade, wouldn’t it be the bezel
that you would want to expand then shrink to fit stone. I would have
to think the metal expansion would be much greater than even a deep
frozen stone would ever shrink.

Best of luck,
Advantage Tool & Microweld
Wire EDM and tool room support for manufacturing.


Wow Charles, I’ve never heard of anyone doing any such thing, but I’m
mighty interested in the results. I’m sure the question on most of
our minds is: why not the usual setting methods??? I’m thinking not
too many stones will hold up to the temp change, but I’m extremely
curious as to how this plays out. What gemstone material are you
using and what metals? Mike.


Charles - For starters you may want to try THE HANDBOOK OF CHEMISTRY
AND PHYSICS. It has been many years since I’ve owned a copy, but it
is published every year or so (used to be), and has all sorts of
valuable info including things like coefficients of thermal
expansion. Of course you’ll also need a good mineralogy text to find
out the chemical composition of the stone(s) you’re using, so you can
check the coefficient properly. I expect any good public library will
have what you need to start with.

Jim Small
Small Wonders


dear charles barnard - don’t go there!!! or, as the mercator maps
said: ‘here be dragons’. using the method you propose would be a lot
like a high school guy’s worst date nightmaRe: get the temperature
way up & then slap ice on the jewels - total disaster! ive


Howdy Charles, I did do some of this research a few years ago but
don’t recall all my sources specifically. Much data for the garnets,
spinels, corundum and other synthetics and lab-created material can be
found on-line. I found some data at my local library. It was very
frustrating. Be aware that some of the coefficients are 10 to the -6
and others to the -7! Also, tourmaline has one xtal axis that is a
negative COE ! Also, any inclusions will make the project more likely
to fail due to strain. You didn’t mention the gem or the receiving
setting material. Most of the crystalline gems are in the low double
to single digits. Much lower than the metals. heating the metal would
accomplish much more tha chilling a stone. Good luck and let us know
how it works out. Carl 1 Lucky Texan


Are you planning to custom cut the stone to fit the castings or are
you planning on buying standard stones… the reason I ask is that
there is actually quite a bit of variation in faceted stones and I’m
not at all sure that your idea of chilling the stone and inserting
it into the ring is going to work, I think you will have some
problems locating stones that will fit properly.

Daniel Grandi We do casting and
finished products for people in the trade in sterling,
Gold,bronze/Brass and Pewter