3D Printing and CAD in Custom Silver Jewelry Design

The first time I held a silver ring that started as a CAD file, I was suspicious. It was too perfect. Every curve was mathematically smooth, every measurement exact, every surface uniform in a way that handwork never quite achieves. It felt like cheating. Ten years later, I use CAD in most of my custom work, and I no longer think it’s cheating — I think it’s a tool, like a rolling mill or a jeweler’s saw, that extends what a silversmith can do. But it’s not a replacement for handwork, and anyone who tells you otherwise is selling software.

3D printing silver jewelry and CAD jewelry design have transformed custom jewelry over the past decade. They’ve made certain things possible that weren’t before — complex geometries, precise stone settings, rapid prototyping. They’ve also introduced new failure modes that traditional silversmiths never had to worry about. I’m going to walk through how the digital workflow actually works, where it shines, and where it falls short. If you’re considering a custom silver piece and wondering whether the designer uses CAD, or if you’re a maker thinking about adding digital tools to your bench, this is what you need to know.

The CAD Revolution: Designing on Screen

What CAD Actually Does for Jewelry

CAD — Computer-Aided Design — lets you build a three-dimensional model of a piece of jewelry on a computer before any metal is touched. You define the ring’s dimensions, the stone’s size and position, the prong thickness, the shank profile, every element as a mathematical object. The software calculates volumes, weights, and even estimates the silver needed before you commit to making it.

This changes the design process fundamentally. In traditional handwork, you design by making — you cut, file, solder, and adjust until the piece looks right. If something’s wrong, you either fix it or start over. In CAD, you design by modeling — you build the piece virtually, rotate it, check it from every angle, and adjust dimensions with a few keystrokes. You can show the client a photorealistic render before making anything. You can print a cheap plastic prototype to check proportions. You can make ten variations of a design in the time it would take to fabricate one.

For custom silver jewelry technology, this is a genuine revolution. Clients can see exactly what they’re getting before committing. Designers can iterate without wasting material. And complex designs that would be extremely difficult to fabricate by hand — say, a ring with a lattice structure, or a pendant with interlocking moving parts — become feasible because the CAD software handles the geometry.

The Software Landscape

The jewelry-specific CAD market is dominated by a few programs. Matrix (now Matrix Gold) is the industry standard — it’s built on Rhino 3D and adds jewelry-specific tools like ring builders, prong generators, and stone libraries. It’s powerful but expensive: a license runs $5,000 to $7,000. RhinoGold (now CounterSketch) is a lighter alternative at around $1,500. For budget-conscious makers, plain Rhino 3D ($995) with free plugins can do most of what the jewelry-specific packages do, just with more manual work.

There are also free and low-cost options. Blender is free and can do jewelry modeling, though it’s not designed for it and lacks precision measurement tools. Fusion 360 has a free tier for hobbyists and is excellent for mechanical designs. Tinkercad is browser-based and very basic, but I’ve seen people make simple ring models in it. The tool matters less than the skill of the person using it — a good designer can make beautiful jewelry in Blender, and a bad designer can make junk in Matrix.

The Learning Curve Is Real

Here’s what the software companies don’t tell you: learning CAD for jewelry takes months of dedicated practice, not a weekend workshop. You’re learning to think in three dimensions on a two-dimensional screen, to manipulate surfaces and solids with mathematical precision, and to anticipate how a digital model will translate into physical metal. The interface alone takes weeks to become comfortable with.

I spent about six months learning Rhino before I produced a model I’d actually cast. The first dozen models I made had problems: walls too thin, prongs too weak, undercuts that wouldn’t cast properly, stone seats that didn’t fit. CAD lets you design things that can’t be manufactured — a common beginner mistake is creating geometry that looks great on screen but is impossible to cast or finish. The software doesn’t warn you that your 0.3mm prong will snap when you try to set a stone in it. You learn that from failure.

From Screen to Wax: 3D Printing the Model

Wax Printers: The Industry Standard

Once the CAD model is finished, it gets exported as an STL file and sent to a 3D printer. For jewelry, the printer needs to produce a model that can be burned out in the lost wax casting process. This means the print material must be a wax or a resin that burns out cleanly without leaving ash residue.

The industry standard for high-resolution wax printing is the Solidscape line (now owned by envisionTEC), which uses a inkjet-style process to deposit wax layer by layer. These printers achieve layer resolutions of 6 to 13 microns — about a tenth of a human hair — and produce wax models that are directly castable. A Solidscape printer costs $20,000 to $80,000, which is why most independent jewelers use printing services rather than owning one. You upload your STL, pay $15 to $50 per model, and get a wax print in the mail a few days later.

Resin Printers: The Accessible Alternative

In the last five years, a new category of printers has made digital jewelry design accessible to small studios: castable resin printers. These are SLA (stereolithography) or DLP (digital light processing) printers that cure liquid resin with UV light, layer by layer. Brands like Formlabs, Phrozen, and Anycubic make printers that cost $300 to $4,000 — a fraction of a Solidscape — and can use special “castable” resins designed to burn out in investment.

The resolution of these printers is remarkable. A $400 Phrozen Sonic Mini can print at 22-micron resolution, which is fine enough for most jewelry. The castable resins (from companies like BlueCast, FunToDo, and Liqcreate) burn out cleanly when you follow the right burnout schedule — usually a longer, slower ramp than traditional wax to ensure complete resin removal.

The trade-off is reliability. Wax printers are designed for production and produce consistent results. Resin printers require more tinkering — calibration, resin handling, post-curing, and the occasional failed print. The burnout schedule for castable resin is more finicky than for wax, and if the investment doesn’t fully burn out the resin, you get a cast piece with carbon inclusions and rough surfaces. I’ve had good results with resin printing, but I’ve also had failures that cost me a day of work and a flask of investment.

The Casting Step: Turning Digital to Metal

Once you have a printed wax or resin model, the casting process is the same as traditional lost wax casting. The model gets sprued to a tree, invested in plaster, burned out in a kiln (the print material vaporizes), and silver is cast into the cavity. The result is a silver copy of your digital model.

This is where the digital workflow’s strengths and weaknesses both become apparent. The strength: you get exactly what you designed, with every detail reproduced faithfully. If your CAD model has a 0.5mm engraved line, the cast silver has a 0.5mm engraved line. If your prongs are perfectly positioned around the stone seat, the cast prongs are perfectly positioned.

The weakness: the cast metal has the same properties as any cast metal. It’s softer than fabricated metal, it may have porosity, and it requires the same hand finishing as any cast piece. CAD doesn’t eliminate the bench work — it just changes where the bench work happens. Instead of fabricating from sheet and wire, you’re filing sprues, removing investment, sanding, and polishing. The finishing time is similar to traditional casting.

What CAD Changes About Custom Design

The biggest impact of CAD on custom silver jewelry isn’t the printing or the casting — it’s the design phase. Here’s what changes when you move from hand design to digital design:

• Client involvement: You can show a client a photorealistic render of their ring before making it. They can request changes — wider band, different stone, thicker prongs — and see the result in minutes. This dramatically reduces the “I thought it would look different” problem that plagues custom commissions.
• Precision: A CAD-designed ring can be made to exact finger size, with stone seats cut to the exact dimensions of the specific stone being set. No shimming, no adjusting, no “close enough.”
• Repeatability: If a client loves a design and wants a matching band, you can reprint and recast identical pieces. The CAD file is a permanent record of the design.
• Complexity: Designs with interlocking parts, lattice structures, or precise geometric patterns are feasible in CAD that would be nearly impossible by hand. A ring with a continuous woven pattern? Easy in CAD, brutal by hand.
• Weight estimation: The software calculates the volume of the model and tells you exactly how much silver the piece will use. This makes pricing accurate and prevents material waste.

What CAD Can’t Do (Yet)

For all its power, CAD has real limitations that the marketing materials gloss over. Let me be specific about what digital jewelry design still can’t handle well.

Organic texture is the big one. CAD produces smooth, mathematically defined surfaces. If you want a hammered texture, a hand-carved look, or the organic irregularity of real handwork, CAD can’t deliver it natively. You can apply texture maps in software, but the result often looks artificial — the texture is too uniform, too perfect. A real hammered surface has variation in depth and spacing that reflects the randomness of human hammering. CAD textures are essentially repeating patterns stamped onto a surface, and the eye can tell the difference.

Hand finishing is still required and still matters. A CAD-designed, cast ring comes out of the mold with the same rough surface as any cast piece. The quality of the final piece depends on the finishing, which is handwork. A poorly finished CAD piece looks worse than a well-finished hand-fabricated piece. The digital design doesn’t guarantee a good result — it just gives you a precise starting point.

CAD also can’t replicate the structural properties of fabricated metal. A CAD-designed, cast ring is still cast metal — softer, more porous, and less durable than a ring fabricated from milled stock. If durability is the priority, CAD design combined with casting is still a step down from hand fabrication. You can design a beautiful ring in CAD that will dent more easily than a simpler ring made by hand.

And there’s a subtler limitation: CAD tends to homogenize design. When every designer uses the same software with the same tools, designs start to look similar. The ring builder in Matrix produces rings with a certain aesthetic — clean, symmetrical, geometric. That aesthetic is everywhere now, and it’s starting to feel generic. Handwork, with its inevitable imperfections and idiosyncrasies, produces work that’s recognizably individual. CAD produces work that’s recognizably CAD.

Post-Processing: The Steps Between Print and Cast

Printing the model is only half the battle. What comes off the printer is not ready for casting — it needs post-processing, and this is where many beginners lose time and quality. Let me walk through what actually happens between the printer and the casting flask.

For wax prints from a Solidscape machine, post-processing is minimal. The support material (a different wax that holds the model during printing) gets dissolved in a solvent bath. The model is rinsed, dried, and inspected under magnification for layer artifacts or voids. If the print is clean, it goes straight to spruing. Solidscape prints are the gold standard because they need almost no cleanup.

Resin prints need more work. After printing, the model is removed from the build plate and washed in isopropyl alcohol to remove uncured resin. Then it’s post-cured under UV light — usually 15 to 30 minutes in a curing station — to fully harden the material. At this point, the model has support structures (small pillars that held it up during printing) that need to be carefully cut away with a scalpel or flush cutters. The removal points need to be sanded smooth, because any mark on the model will appear on the cast silver.

Layer lines are the next issue. Even at 25-micron resolution, resin prints have visible layer lines if you look closely. For most designs, these lines don’t matter because they get polished out during finishing. But for designs with large, flat surfaces or sharp edges, the layer lines can be visible in the final casting. Some makers sand their resin models before casting to remove layer lines, but this is fiddly work on small, delicate models and risks breaking them. The alternative is using a finer layer height (10 or 15 microns), which increases print time from hours to days.

Common CAD Mistakes That Ruin Castings

After several years of CAD-to-cast work, I’ve cataloged the mistakes that consistently produce failed castings. If you’re learning CAD for jewelry, these are the pitfalls to watch for.

• Walls too thin: Anything under 0.4mm is risky for casting. The metal may not fill the cavity, or the wall may be so thin after finishing that it bends under pressure. Minimum wall thickness for cast silver should be 0.5mm, and 0.7mm is safer.
• Undercuts that trap investment: Deep undercuts in the design can trap investment material that doesn’t break out cleanly after casting. The result is a casting with investment stuck in crevices that’s nearly impossible to remove without damaging the silver.
• Sharp internal corners: CAD lets you draw sharp internal corners (where two surfaces meet at less than 90 degrees). But sharp corners are stress concentrators in cast metal and can cause cracking. Fillet every internal corner with at least a 0.3mm radius.
• Sprue attachment points on visible surfaces: The sprue leaves a mark where it was attached. If you don’t plan where the sprue attaches, the mark may end up on a visible surface that’s difficult to clean up. Design the model with a designated sprue attachment point on a hidden or easily finished surface.
• Stone seats cut to generic dimensions: CAD software often has standard stone sizes, but actual stones vary. A 6mm round stone might be 5.9mm or 6.1mm. If you cut the seat to exactly 6.0mm in CAD, the stone may not fit. Always leave the seat slightly undersized and enlarge it by hand after casting, or measure the actual stone before finalizing the model.
• Ignoring shrinkage: Silver shrinks about 5% during casting. Most CAD software accounts for this automatically if you set the correct shrinkage factor, but if you forget to set it — or set it wrong — your size 7 ring comes out as a size 6.5. Always verify the shrinkage setting before printing.

These mistakes are all preventable, but they’re the kind of thing you only learn through failure. Every CAD jeweler has a drawer of castings that didn’t work because of a wall that was too thin or a corner that was too sharp. The software doesn’t warn you because it doesn’t know — it’s a modeling tool, not a manufacturing consultant. You have to bring the manufacturing knowledge yourself, and that only comes from casting experience.

The Hybrid Workflow Most Professionals Use

Very few professional jewelers are purely digital or purely traditional. Most use a hybrid approach that leverages the strengths of each. Here’s what I’ve seen work well:

Use CAD for the structural elements that benefit from precision: the stone seat, the prong positions, the shank profile, the overall proportions. Print and cast these elements. Then add handwork: hammer texture on the shank, hand-engraved details, applied wirework, or hand-fabricated decorative elements soldered onto the cast base. The cast piece provides the precision; the handwork provides the character.

Another hybrid approach: design in CAD, print a plastic prototype for the client to approve, then hand-fabricate the final piece using the prototype as a guide. This gives you the client-approval benefit of CAD without the casting step. The CAD model is a design tool, not a manufacturing tool.

Or use CAD for production pieces — designs you’ll make multiple times — and hand fabrication for one-of-a-kind commissions. The CAD pieces benefit from repeatability; the hand-fabricated pieces benefit from individuality. Many studios operate this way, with a production line of cast designs and a custom line of fabricated pieces.

Should You Learn CAD for Jewelry?

If you’re a silversmith wondering whether to invest time in learning CAD, my answer is: yes, but with realistic expectations. CAD won’t replace your hand skills — it will augment them. The best CAD jewelers are the ones who already understand metal, who know how a piece will wear, who can anticipate problems because they’ve made things by hand. A designer who only knows CAD and has never held a torch produces designs that look good on screen and fail in metal.

Start with Rhino. It’s the foundation of most jewelry CAD software, and the skills transfer. Get a student license for $195. Spend three months going through tutorials — there are excellent free ones on YouTube from people like Gary Dawson and Jansma. Model simple things first: a plain band, a bezel setting, a pendant. Print them in plastic on a cheap FDM printer to check proportions. When you’re confident, invest in castable resin printing or use a service.

Budget for failure. Your first castings from CAD models will have problems. Thin walls, blocked details, stones that don’t fit. Each failure teaches you something the software can’t. I keep a shelf of failed CAD castings as a reminder that digital design still has to answer to physical reality.

The Honest Limitations of Digital Jewelry Design

I want to end with a frank assessment of where 3D printing and CAD fall short, because the technology is often oversold.

Cost is a factor that gets ignored. A full CAD-to-cast setup — software, a castable resin printer, a vacuum casting machine, investment, a burnout kiln — represents $8,000 to $15,000 in equipment. That’s before you factor in the learning time. For a small studio, this is a significant investment that may take years to recoup. If you only make a few custom pieces a year, hand fabrication is more economical.

Castable resin printing is still finicky. The resins are expensive ($100 to $300 per liter), they have shelf lives, they require specific post-curing protocols, and the burnout schedules are longer and more demanding than wax. A single failed burnout — where the resin doesn’t fully vaporize and contaminates the investment — wastes a flask of investment ($15), the silver you tried to cast ($20 to $50), and a day of time. Resin printing is accessible but not yet as reliable as wax printing on industrial machines.

The “perfect” look of CAD design can be a drawback for certain aesthetics. If your brand is built on handmade character — hammer marks, slight irregularities, the evidence of human hands — CAD can work against you. Some clients specifically want the imperfections of handwork, and CAD’s precision is the opposite of what they’re looking for. Know your market before you go all-digital.

Finally, CAD can create a false sense of certainty. A beautiful render looks finished, but it’s just an image. The physical piece may reveal problems that the render hid: a prong that’s too thin, a surface that’s too flat, a proportion that feels wrong in the hand. I’ve had clients approve a render and then dislike the physical piece because it felt different than it looked on screen. Metal has weight, warmth, and presence that a digital model can’t convey. Never assume the render is the reality.

CAD and 3D printing are tools, not a philosophy. Used well, they expand what’s possible in custom silver jewelry. Used poorly, they produce precise, soulless pieces that look like they came from a machine — because they did. The best jewelers I know use digital tools where they help and put them down where they don’t. The CAD file is the beginning of the process, not the end. The finished piece is still made by hands, on a bench, with fire and files and patience. That hasn’t changed, and I don’t think it will.