Quick Answer
A dental STL file is millable when three independent conditions hold: the mesh is geometrically valid (watertight, no intersections, one connected shell), the geometry respects the rules of the target material (wall thickness, edge stability, connector dimensions), and the part is reachable by the machine setup (orientation, undercuts, tool access). A file can pass any two of these and still fail the third.
Your CAM software imported the file without complaints. So it will mill, right?
That assumption costs milling centers real money every week. An STL import is not a millability check. It only proves the file could be read. Whether the case survives nesting, toolpath calculation and the machine itself is decided by criteria your importer never looks at.
The good news: you can test all of them before CAM, in a fixed order, in minutes. This guide walks through the seven checks in the sequence that catches the expensive problems first.
What “millable” actually means
Millability is not a property of the file alone. It is a relationship between three things: the geometry, the material, and the machine setup. Change any one of them and the answer can flip. A coping that mills cleanly in PMMA can chip in glass ceramic. A divergence that a 5-axis machine handles is a dead end for a 3-axis strategy.
A useful definition: a dental STL file is millable when the described geometry can be produced on the intended machine, in the intended material, without manual rescue work, and results in a restoration that seats. Every check below tests one part of that sentence.
The 7-step millability check
Step 1: Mesh integrity
Before anything else, confirm the file is a valid closed volume: watertight, no self-intersections, consistent normals, one connected shell per expected part, no floating fragments. A mesh that fails here makes every later check unreliable, because software cannot even agree on what is inside the part.
Common trap: CAM importers repair silently. The file looks fine on screen and fails three steps later, where diagnosis is expensive. Details on the ten most frequent defects: Dental STL File Errors: 10 Problems That Stop a Case Before Milling.
Step 2: Units and dimensions
STL stores no units. Sanity-check the bounding box against the indication: a crown should measure crown-sized millimeters. A part that is 10 times too large or small points to a unit mismatch at export, and no later step will save it.
Step 3: Orientation and insertion direction
Check how the part sits relative to the blank and the machining axes, and whether a valid insertion direction exists. A poor orientation manufactures problems that the design never had: artificial undercuts, wasted blank height, bad tool angles on the critical surfaces.
Step 4: Undercuts and tool access
Now test reachability: can the available tool directions reach every surface that must be machined? Undercuts relative to the milling setup mean unreachable regions; undercuts relative to the insertion path mean a restoration that will not seat even if the machine could cut it. This is the first check that depends on your specific machine strategy, which is why a file can be millable in one center and unmillable in another.
Step 5: Wall thickness
Compare local material thickness against the minimum for the actual blank material. The limits are material- and manufacturer-specific; zirconia, lithium disilicate, PMMA and CoCr each draw the line differently. Check the known hot spots: occlusal fissures, the transition to the margin, connectors.
Step 6: Margin line quality
Inspect the margin for interruptions, ragged segments and extremely thin run-outs. The margin decides fit, and margin defects are the classic case of a problem that originates in the scan or preparation but becomes production’s problem. If it fails here, the case goes back to the lab with a marked-up report, not into the queue.
Step 7: Material and indication rules
Finally, apply the rules that have nothing to do with geometry validity: minimum connector cross-sections for bridges, edge stability for the chosen ceramic, indication-specific plausibility. This is the layer that generic mesh tools do not model at all, and the reason a “repaired” file is not the same as a producible file. Why that distinction matters: Why a Technically Valid STL Can Still Be Unmillable.
Manual check or automated validation?
| Criterion | Manual inspection | Automated validation |
|---|---|---|
| Time per file | 5 to 30 minutes | Seconds |
| Consistency | Depends on person, time of day, workload | Identical rules on every file |
| Material rules | From memory or lookup sheets | Configured per material and machine |
| Documentation | Usually none | Report per file, visual and archivable |
| Scaling | More staff | More volume, same process |
Manual inspection works at low volume with experienced staff. It stops working at exactly the moment file intake grows, which is the moment quality control matters most. An automated check runs all seven steps on every incoming file and returns a documented go/no-go answer before a technician opens the case; repairable mesh defects are fixed automatically, and everything else goes back to the sender with a visual report.
That is what SimQ DENTAL Mill Check does in the browser. The complete list of checks is documented on What We Check.
The pre-CAM checklist
Mesh watertight, no intersections, clean normals. One shell per part, no fragments. Dimensions plausible for the indication. Orientation set, insertion direction valid. All surfaces reachable in the planned setup. Wall thickness above the material minimum in fissures, margins and connectors. Margin line continuous and stable. Material and indication rules confirmed. Only then: nest.
FAQ
How long should a millability check take?
Automated: seconds per file, including the report. Manual: plan 5 to 30 minutes per suspicious file, plus the interruption cost of pulling a technician out of their workflow.
Can my CAM software do this?
Partially. CAM systems check what they need for toolpath calculation, and many repair silently rather than reporting. They generally do not test material-specific wall thickness rules, margin quality or indication plausibility, and they answer at the latest possible moment, after a technician has already invested time in the case.
Is a millability check the same as an STL repair?
No. Repair fixes mesh defects. A millability check answers a bigger question: can this geometry be produced in this material on this machine? Many unmillable files have flawless meshes.
Does a 5-axis machine make undercut checks unnecessary?
No. Five axes extend what is reachable, they do not make everything reachable. Tool length, tilt limits and fixture collisions still define blocked regions, and insertion-path undercuts remain a fit problem regardless of the machine.
What happens when a file fails the check?
The efficient pattern: automatically repairable defects get fixed and documented; everything else goes back to the lab with a visual report that marks the problem region. That turns a 30-minute phone call into a 30-second decision.
Stop wasting hours on unmillable files. Run the full check on a real case now: start the free browser test. No installation, no credit card.