Back to Ganoksin | FAQ | Contact

Designing settings with CAD for 3D printing

I am starting to design jewelry with the Rhino4 program and will
have the master models made with a .Stl file on a 3D printer. I would
like to use some small gem stones in the designs ( rounds 2-3mm and
ovals 3.5X4.5mm). I am wondering about the feasability of creating
the settings in the computer model. I would like to use bezel or
possibly prong settings. Does any body know if this is a standard
practice in the industry, or would I be trying to do some thing that
is very difficult. Things I would like to know aRe: can a small
bezel setting be cast or will the cast bezel not bend over well. And
does anyone have any info on, whats the best way to size up the
setting in the computer model to compensate for shrinkage.

Thanks Bruce


Extensive testing with various waxes, growing materials and
investments has shown that the wax or growing material does not
shrink. The investment expands.

Same effect. Depending on your investing procedures and the
materials used you may experience very little to maybe 6%
“shrinkage”. Experimentation will soon give you a direction for your
own procedures; there is no correct universal answer.

In Rhino or Matrix, the 3D Scale tool in the Transform Menu is what
is used. See the “HELP” file if you need more, but it is very

Good luck on the journey!
Wayne Emery


I like this question. The answer as I see it-Maybe. Depends. Do you
want to open up options for later with your one model? It is a common
practice to cast in prongs but-

It depends greatly on what metal you will be casting in. Taking a
very conservative view for a minute-Cast metal is not as good as
fabricated/stamped/drawn wire at holding the stones in place. From
this same viewpoint, anytime we design the prongs or bezel into the
casting we are saving money, time, or difficulty at the expense of a
slightly (at best) or much (at worst) less secure gem. Admittedly,
sometimes the only practical way is in the casting. However, why not
consider two piece casting? Shanks are thick and prongs are thin.
Might be nice to have the option to cast each at its own ideal
temperature. If your labor cost is by weight, you do not add much
cost at all.

When you compromise like casting in prongs and bezels you become
totally dependant on the casting metal remaining flexible enough. As
we all find out, that does not always happen, and some colors and
karats can confound any of us. With cast in settings-You reduce the
scope of the design potential-Designing prongs into a casting
intended as 14k super white, nickel based casting gold, or 18k red
gold you are asking for problems.

If you design to add fabricated or purchased settings later, you can
go far more red in your 18kt, and the settings can be the color you
choose to buy or make yourself. Two tone anyone?

Everyone who sits down to design in software must ponder metal
issues, casting issues, and those long term ideas like whether the
prong color needs to be other than the rest of the piece to suit the
stone. A design made for multiple materials will be different in
important ways from a design aimed at a specific metal. Like platinum
vs. white gold or 95pd.

Cast sterling bezels work well. Cast 18k rose gold bezel likely-will
not. What metals are you intending?

Daniel Ballard
Precious Metals West
National Sales Manager

I have modelled a few cluster claw settings for wax milling with
Rhino using the same size stones. No problems. And many hand carved
bezels which were cast. Since I tend to use heavy bezels and set with
a hammer and chasing tools I’m not too sure about really thin bezels
which can just be burnished.

As to shrinkage I tend to ignore it, I want cut bearings for any
stone I’m setting. If you want to build the bearings… try a sample
test and measure. Too many variables for a text book number.

Demand Designs


Great post.

Yes, there IS really SOME shrinkage inn the metal, but the
controlled tests we ran for 3D Systems and their growing substrates
vs Kerr hard green wax indicated that the investment created more
"shrinkage" problems through expansion than the metals did via
actual shrinkage from the melt. The exceptions, as I recall, were the
Pt alloys, which gave the greatest shrinkage from liquidus to solid

In conjunction with that, I did a survey of 50 accomplished bench
people (and included my own vicarious experience in the repair and
manufacturing end). The question was:

Given the average ring casting, how much actual removal of metal
takes place in YOUR finishing process? No guesses, please, actual

It was a true learning experience for many, including myself. The
average was 0.3 mm, with care. In ALL cases, that exceeded any
differential caused by “shrinkage”.

In the end, because we were in the business of selling CAD and CAM
solutions, including mills and “growing” machinery, our "suggested"
answer was to apply 7-10% upwards 3D scaling to the CAD model, and
experiment from there.

Concerning the creating of prongs and bezels in the CAD model (to be
cast), I have some serious reservations, but I’m sure you do, too. A
discussion for another time, perhaps.

Thanks again!
Wayne Emery