CNC vs Investment Casting for Metal Parts
Short answer: investment casting (lost wax) wins when you need a complex near-net metal shape in production volumes — typically a few hundred to tens of thousands of parts — where the per-piece machining time of cutting from solid would be enormous, and where the casting's tolerance (roughly ±0.1–0.25 mm on small features) is acceptable. CNC machining wins for prototypes, low volume, tight tolerance, high-strength requirements and any geometry that must be proven before tooling. The crossover for many small parts sits between 200 and 1,000 pieces; above that, casting plus light machining usually wins on cost if the alloy and tolerances allow it.
This is one of the oldest manufacturing decisions and one of the most frequently mismade. Engineers pick casting for cost and discover tolerance problems; they pick machining for precision and discover price. The useful way to think about it is not which process is "better" but where each part lands on four axes: volume, tolerance, alloy, and geometry complexity. This guide gives you the crossover math and the honest questions to ask before you commit tooling money.
How the Two Processes Compare
| Factor | CNC machining from bar/billet | Investment casting (+ finish machining) |
|---|---|---|
| Tooling cost | None (programming + fixturing only) | Pattern die $2,000–$20,000+ depending on size/complexity |
| Lead time to first parts | Days (prototype) | 3–8 weeks (pattern, shell, cast, finish) |
| Typical tolerance | ±0.01–0.05 mm machined | Cast: ±0.1–0.25 mm; + machining where needed |
| Volume economics | Best at 1–500 pcs | Best above ~200–1,000 pcs |
| Material utilization | 20–60% (chips are waste) | 80–95% near-net |
| Alloy range | Bar/billet availability limits | Wide: castable grades of steel, stainless, aluminum, bronze, superalloys |
| Strength | Wrought properties (often higher) | Cast properties (fine but different; porosity risk) |
| Design freedom | Limited by tool access | Very high: hollow, internal, thin-walled shapes in one pour |
| Surface finish | Ra 0.4–1.6 µm typical | As-cast Ra 3–12 µm; machined where required |
Where CNC Machining Wins
Cutting from solid is the right answer in four situations. Low volume: below a few hundred pieces, tooling amortization kills casting economics. Tight tolerance: any feature at ±0.05 mm or tighter needs machining regardless, and if most features are machined anyway, starting from bar often wins. High mechanical requirements: wrought bar stock has consistent, directionally tested properties; castings can hide internal porosity that only shows up in fatigue. Speed: a prototype or urgent repair part is machined in days, cast in weeks. And iteration: if the design is still changing, machining costs nothing per revision while every casting pattern change is money.
Where Investment Casting Wins
Casting earns its tooling when three conditions hold together. The shape is complex enough that machining from solid would waste hours of spindle time and most of the material — hollow geometries, undercuts, thin walls, internal passages. Volume is high enough to amortize the die — for typical small hardware, the crossover lands around 200 to 1,000 pieces, and drops fast if the part is stainless or titanium, where bar cost and machining cost both run high. And the alloy is available in castable form — many grades cast well (CF8M/316L equivalent, 17-4PH, aluminum A356, C95500 bronze) while some wrought-only grades do not. Casting plus light finish machining on critical faces is a hybrid used constantly: the casting supplies the shape, the machine supplies the ±0.02 mm datum faces and threads.
Cost Crossover: A Worked Comparison
Take a 316L valve body, roughly 60 mm cube, with internal passages and three threaded ports. Machined from hex bar: bar cost modest, but cycle time 40–60 minutes with deep-hole and port work — at a $60–80/h shop rate plus material, the part lands in the $60–120 range each at low volume, falling slowly as fixturing improves. Investment cast: pattern die $6,000–$12,000, casting unit cost $15–35 at 500–2,000 pcs, plus finish machining of the three ports and faces at $20–40. At 200 pieces the total is roughly a wash; at 1,000 pieces casting saves 30–50% per part; at 10 pieces machining wins by an order of magnitude. The crossover moves left (casting wins earlier) for stainless, titanium and complex thin-wall parts, and moves right for aluminum and simple shapes where machining is fast.
Tolerance Strategy That Keeps You Out of Trouble
Never specify a cast surface at a machined tolerance unless you plan to machine it. Casting houses quote as-cast tolerances around ±0.1–0.25 mm on small features, and linear dimensions across a large casting drift more. The professional approach: mark functional faces, threads, bores and sealing surfaces as machined, leave the rest as-cast, and put a note on the drawing that as-cast surfaces follow ISO 8062 tolerances. That single habit eliminates most casting disappointment. If a supplier quotes ±0.05 mm as-cast, they are either lying or plan to machine it — ask which.
Design for Each Route
For machining: design around standard bar sizes, avoid deep narrow pockets, and let internal radii match standard tools. For casting: design for uniform wall thickness (2–6 mm typical in small parts), generous fillet radii at junctions, avoid isolated heavy sections that shrink-porosity, and remember that holes under ~3 mm are usually drilled afterward rather than cast. A good DFM review before tooling decides which route your part should take in the first place — our engineering team runs that review on your model for free before quoting, and will tell you plainly if casting is the wrong road for your volumes.
Frequently Asked Questions
Q: Is investment casting cheaper than CNC machining?
A: Above the volume crossover — typically 200 to 1,000 pieces for small hardware, less for stainless and titanium — investment casting plus light finish machining is usually 30-50% cheaper per part. Below that volume, CNC machining from bar wins because there is no tooling cost. The crossover depends on part complexity, alloy and tolerance.
Q: What tolerance can investment casting hold?
A: As-cast, small features hold roughly ±0.1-0.25 mm depending on geometry and alloy; larger linear dimensions drift more. Features that need ±0.05 mm or tighter should be machined after casting, which is standard practice for functional faces, threads and bores.
Q: Can I cast the same alloy I currently machine?
A: Often yes, but check. Stainless 316L (CF8M), 17-4PH, aluminum A356/6061 (castable but different grade), and many bronzes cast well. Some wrought-specific grades and tempers do not translate. A foundry can recommend the castable equivalent, but verify mechanical properties against your application before switching.
Q: How long does investment casting tooling take?
A: Pattern die tooling typically takes 3-8 weeks including sampling, depending on part size and complexity. First articles usually need one iteration to dial in shrinkage. Plan casting programs 4-10 weeks ahead of need — that lead time is the main reason prototypes stay on CNC.
Q: When should I use CNC plus casting together?
A: Whenever the part has complex geometry plus critical features. Cast the near-net shape, then machine the datum faces, bores, threads and sealing surfaces. This hybrid gives casting's cost at volume with machining's tolerance exactly where it matters — it is the standard approach for valve bodies, pump housings and structural brackets.
Authored by the BQUQ Engineering Team. BQUQ is an ISO9001-certified source factory in Dongguan, China, running CNC machining, metal stamping, custom springs, heat sink and collet lines under one roof. Send drawings to sc@bquq.com or WhatsApp +86 13713157787 for a quote within 12 working hours. www.bquq.com
