SkillsProto prints SLA, SLS and MJF parts in 2-4 days and cuts aluminium injection tooling in 7-10 days with T1 samples at about two weeks, so the choice is not about capability but about where your design is. Print while the geometry is still moving and the quantity is under about ten. Mold once the geometry is frozen and the quantity is approaching a hundred, because from there the tooling is amortised and the per-part price collapses. Between ten and a hundred, the honest answer depends on how sure you are that nothing will change.

The quantity rule, and the two exceptions that override it
The rule of thumb is simple: printing wins below roughly ten parts, molding wins above roughly a hundred, and the gap between is decided by confidence rather than by arithmetic.
Two exceptions override the rule in both directions.
Print above the crossover when the design is not frozen. An aluminium tool that has to be modified twice costs more than printing fifty parts and molding afterwards. If a stakeholder review, a fit check or a thermal test is still pending, printing is buying information, not parts.
Mold below the crossover when the part must be validated in production material. A printed PA12 part is not an ABS or PC part, and a functional test on a printed proxy can pass while the molded part fails on warp, sink or weld-line strength. When the test is a regulatory or customer validation, the material has to be the production material, and that means a tool.
What printing gives you that a mold cannot
- No tooling and no commitment. Every print is a new file. A geometry change costs a re-slice, not a tool modification.
- 2-4 day turnaround. SLA, SLS and MJF parts ship in 2-4 days, which is the fastest loop available for an electronics enclosure that has to meet a PCB revision.
- Geometry with no draft and no ejection. Undercuts, internal channels, lattice and text on a vertical wall are free in printing and expensive or impossible in a two-plate tool.
- One part or fifty, same price per part. There is no amortisation curve to climb.
What molding gives you that printing cannot
- Production material and production properties. The resin you will ship in, with its real shrink, its real weld lines and its real creep behaviour.
- Surface finish straight from the tool. A polished or textured tool gives a Class A or a molded-in texture that no printed part reproduces, and the finish is repeatable across every part in the batch.
- A unit price that falls with volume. Above the tooling break-even, the per-part cost is resin plus cycle time, and cycle time is seconds.
- Tolerances that hold across a batch. A molded part holds roughly ±0.05 mm on well-chosen features and repeats it part to part; printed parts vary with orientation and build position.
| Attribute | 3D printing (SLA/SLS/MJF) | Rapid injection molding |
|---|---|---|
| Lead time, first part | 2-4 days | 7-10 days to T1, about 2 weeks with samples |
| Tooling cost | None | Aluminium tool, paid once |
| Economic quantity | 1-50 | About 100 and up |
| Design change | Re-slice, same day | Tool modification or new insert |
| Material | Resins and nylons | Production resins, 100+ options |
| Surface finish | Layer or sintered texture, hand finishing | Tool finish, repeatable |
| Tolerance | Orientation dependent | Roughly ±0.05 mm on chosen features |
| Best fit | Fit checks, jigs, low-volume functional parts | Validation builds, pilot runs, bridge to production |
Cost per part, at three quantities
Take a hand-held enclosure half, about 120 x 70 x 25 mm, in a material that exists both as a printable nylon and as a moldable resin. The numbers below are a relative index with an aluminium tool included, not a quotation.
| Quantity | Printed, relative total | Molded, relative total | Cheaper route |
|---|---|---|---|
| 1 | 4 | 100 (tool dominates) | Printed |
| 10 | 40 | 103 | Printed |
| 50 | 200 | 112 | Printed, narrowly |
| 100 | 400 | 121 | Molded |
| 500 | 2,000 | 160 | Molded |
| 2,000 | 8,000 | 320 | Molded |
The shape matters more than the digits: printing is linear in quantity with a high slope, molding is a step plus a shallow slope. The crossover in this example sits near fifty to a hundred parts, and it moves with part size, because a large part consumes more print time and more resin per unit while a mold's cycle time grows more slowly.
The revision trap
The most expensive mistake in this decision is molding a design that then changes. A tool cut for a 2 mm wall, a specific boss height and a particular connector cut-out cannot absorb a PCB that grows by 3 mm, and the modification costs a tool change plus another lead time.
A cheap guard: print the part, assemble it around the real PCB, connector and battery, and only then commit. The print costs a fraction of the tool and answers the questions a CAD review cannot, which are almost always about clearance, cable routing and whether a finger can reach the connector.
Print for fit, mold for validation
The sequence most successful hardware teams run is a two-step one. Print two or three iterations until the mechanical fit stops changing, then cut aluminium tooling and mold the validation batch in production material.
Aluminium tooling at 7-10 days supports this because the decision does not have to be made months ahead, and because an aluminium tool tolerates a limited number of engineering changes better than a hardened steel tool. Steel tooling at 15-25 days is the right call when the design is locked and the volume is real, not for a pilot batch.
The DFM rules change with the process
Printing and molding reward different geometry, and a part optimised for one is usually mediocre in the other.
| Feature | Printing | Molding |
|---|---|---|
| Wall thickness | Uniform is still better, but 1.2 mm ribs print | 1.5-3 mm walls, ribs at 50-60% of wall |
| Draft | Not required | 1-2 degrees on every vertical face |
| Undercuts | Free | Slider or lifter, cost added |
| Radii | Generous radii help, sharp corners print | Sharp internal corners cause stress and warp |
| Holes | Print round, may need reaming | Mold round with a core pin, draft the pin |
| Text and logos | Free on any face | Needs draft, depth and a tool insert |
SkillsProto returns a DFM report with the priced quote in 24 hours for both routes, which means the same model can be quoted as printed and as molded before you commit either way.
Getting both routes priced
Send the STEP file and the quantity to /quote/. SkillsProto quotes printing (2-4 days) and aluminium tooling (7-10 days to T1, about two weeks with samples) from the same model, holds ±0.005 mm on machined inserts and roughly ±0.05 mm on molded features, runs MOQ 1 on printed and machined parts, and delivers DDP in 3-5 days.
Frequently Asked Questions
At what quantity does injection molding become cheaper than 3D printing?
Usually somewhere between fifty and a hundred parts for a small enclosure, once an aluminium tool is included. The exact point depends on part size, because large parts cost more per unit to print while a mold's cycle time barely changes. Below that range, printing is cheaper because there is no tool to amortise.
Can a 3D printed part be used for functional testing?
For fit, form and moderate mechanical testing, yes, especially in SLS or MJF nylon. It cannot substitute for a test in the production material, because printed nylon differs from molded ABS or PC in impact strength, anisotropy and creep. If the test is a customer or regulatory validation, mold the part.
How long does rapid injection molding take from CAD to parts?
Aluminium tooling takes 7-10 days, with T1 samples at about two weeks including the sample run and inspection. Compare that with 2-4 days for a printed part. The tooling route is slower on the first part and far faster on the thousandth.
Is a printed surface finish good enough for a customer sample?
For an internal fit check, yes. For a customer-facing sample, usually not: printed parts show layer lines or a sintered texture, and hand finishing is not repeatable across units. A molded part takes its finish from the tool, so every part in the batch looks the same.
What is bridging, and does it make sense here?
Bridging means using one process to reach the next milestone: print for the fit build, mold in aluminium for the validation batch, then cut steel when the volume arrives. It makes sense when the schedule has more than one gate, because it avoids paying for steel tooling before the design has stopped moving.
Do I need draft angles if I am going to mold later anyway?
Yes, add them now. A part designed without draft can be printed and then has to be remodeled before tooling, which costs a revision cycle at the worst moment. One to two degrees on vertical faces costs nothing in printing and avoids a remodel in molding.
Related capabilities
Related guides
Have parts to quote?
Upload your files — DFM-checked quotation within 24 hours.
