In this guide
Assign compensation to an identified process stage and owner before changing the jewelry model. Compare nominal design dimensions with measurements from the print, casting and finished piece. Applying an unexplained scale factor to every file can change stone seats, band fit and assembly interfaces—and can duplicate compensation already applied downstream.
Begin with the actual manufacturing route. The relevant printer, pattern material, casting process, alloy and finishing work should be recorded. A percentage remembered from another workshop is not evidence for the current job.
Name the dimension and stage
Choose specific features to track: inside diameter, an overall span, a seat reference or another agreed interface. Record how and when each is measured. A polished finished dimension and an as-cast dimension are not interchangeable observations.
| Stage | Record | Owner to confirm |
|---|---|---|
| Nominal design | Intended dimension and source revision | Designer |
| Prepared print | Applied preparation compensation | Print operator |
| Pattern | Observed feature and measurement method | Print reviewer |
| Casting | Alloy, route and observed feature | Foundry |
| Finished piece | Final measurement and finishing history | Bench reviewer |
The Formlabs material guidance includes material-specific post-processing considerations.1 That is a reason to identify the actual material and route, not to infer one universal jewelry shrinkage allowance.
Check for double compensation
Ask the print operator and foundry whether their preparation already includes an adjustment. Record the magnitude, direction and feature scope. If the designer enlarges a model and the foundry applies its usual enlargement again, the intended compensation may be counted twice.
Use a clear instruction such as “nominal source supplied; process allowance to be proposed by the foundry” or “this revision contains the listed adjustment; do not add another without review.” Choose the instruction that reflects the agreement, and retain it with the exact file.
Calculate only under stated assumptions
For a hypothetical uniform linear reduction of 2%, an initial span required to reach 20 mm would be 20 ÷ 0.98, approximately 20.408 mm. This arithmetic illustrates how a stated assumption is applied; it does not establish that the actual process shrinks uniformly or by 2%.
Do not apply that example to a stone or a complete ring indiscriminately. Keep actual gemstone geometry fixed and review the metal relationships with the setter. See ring resizing without scaling stones for the distinction between design change and overall scaling.
Run a measured pilot instead of guessing
Agree the features, stage measurements and acceptance question before producing the pilot. Retain the source, prepared file and observed results. If the correction varies by feature or direction, record that finding and ask the responsible process specialist to propose a bounded change.
Do not turn one sample into an unqualified production capability claim. Use further samples and the workshop’s process evidence where repeatability matters. The allowance register leaves actual allowance and measurement fields blank.
Preserve downstream identity
When an allowance is approved, identify whether it belongs in the nominal CAD, a process derivative or the preparation job. Keep those files named separately. A later client size change should not accidentally inherit the wrong process correction.
Use the Rendora jewelry workflow for traceable source and manufacturing-evidence handoff. The casting checklist helps connect the file revision to the actual workshop route.
Is shrinkage the same as finishing removal?
No. Record process dimensional change and later material removal separately so the proposed correction addresses the right stage.
References
- Formlabs, Using Castable Wax Resin, reviewed October 26, 2026.
About this guide
Rendora publishes practical guides for teams working with configurable 3D assets and traceable production handoffs. Updated October 27, 2026.
