Low-Volume CNC Machining: 1 to 1,000 Parts Without Tooling

Published 2026-11-18 · 9 min read

SkillsProto runs low-volume CNC production from 1 to 1,000 parts at MOQ 1, with ±0.005 mm on critical features, 3-7 day lead times and a 48-hour expedite. The per-part price falls in steps rather than smoothly, because setup, programming and fixturing are one-time costs spread across the batch, and the crossover where aluminium tooling beats machining usually sits between 100 and 500 pieces. It moves with part size, material, and how much of the price is machining time rather than setup.

CNC milling cutter removing metal from a fixtured workpiece
CNC milling of a fixtured workpiece. Cutting strategy, fixturing and tool access drive both tolerance and cost.

What actually changes between 1 part and 1,000

A CNC quote has two halves. There is a fixed half — programming, fixture design, first-article inspection, machine setup — that you pay once per order regardless of quantity. There is a variable half — material, cycle time, tool wear, finishing, inspection sampling — that you pay per part.

At quantity 1, the fixed half is spread over a single part, so it dominates. At quantity 500, it is spread over 500 parts and nearly disappears. This is why the first price break is dramatic and later ones are gentle, and it is why a buyer who doubles the order often sees a 15-25% per-part reduction rather than 50%.

Quantity bandWhat dominates the unit priceTypical per-part trendThe right process
1-5Programming, setup, fixture, first articleHighest; fixed cost spread over very few partsCNC, MOQ 1, no tooling
6-25Setup amortised, machining time now visibleFalls sharply, then flattensCNC
26-100Machining time and materialGentle decline; fixture improvement starts to payCNC, consider a soft fixture
100-500Cycle time and secondary operationsModest decline; crossover zone against toolingCNC or aluminium tooling, compare both
500-1,000Cycle time, finishing consistencyFlat; volume discount is commercial, not physicalCNC, vacuum casting or aluminium tooling
Above 1,000Tooling amortisationTooling wins unless geometry is simpleAluminium or steel tooling

Setup is amortised, machining time is not

The mistake buyers make is assuming the per-part price keeps falling in proportion to quantity. It does not, because once the fixed half is diluted, the remaining price is genuinely per-part work: cutting metal, changing tools, measuring, finishing, packing.

A part with 40 minutes of cycle time costs roughly the same to cut on part 400 as on part 4. What changes is that the 3 hours of programming and the fixture build have already been paid for.

The practical consequence is that volume savings on machined parts come from three places, in this order of leverage: reducing cycle time through design, reducing the number of setups, and reducing the amount of material removed. A quantity discount that is not backed by one of those three is commercial, not structural, and it will not survive a re-quote.

The crossover against molding is a range, not a number

The question "at what quantity does molding beat CNC" has no single answer because the answer is a ratio between tooling cost and the per-part saving.

Take a plastic bracket that machines for a relative 100 per part at quantity 50, and molds for a relative 20 per part once an aluminium tool exists. The tool is a one-time cost. The crossover quantity is the tool cost divided by the per-part saving. Change the part size and the tool cost changes; change the geometry and the machining time changes; the crossover moves with both.

RouteOne-time costPer-part cost at 500Lead timeSensible range
CNC from billetNoneHighest3-7 days1-1,000
Vacuum casting, silicone moldLow mold costMidAround 7 days for 10-500 copies10-500
Aluminium injection toolTool in 7-10 daysLowest of the three for plastic partsT1 samples in about 2 weeks100-10,000
Steel injection toolHighest tool costLowest at high volumeTool in 15-25 days10,000+

Two rules of thumb hold well enough to plan around. Below roughly 100 pieces, machining almost always wins because tooling cannot be amortised. Between 100 and 500, the answer depends on the specific part and both routes should be quoted. Above 500 in a thermoplastic, aluminium tooling is usually ahead unless the geometry is simple enough to machine cheaply.

A third route matters more than most buyers expect: vacuum casting produces 10-500 plastic copies from a silicone mold in about 7 days with no steel or aluminium tool, which sits neatly between machining at 50 and an injection tool at 500.

Where CNC keeps winning well past 1,000

Machining does not stop being competitive at volume when any of these is true:

  • The material is metal and the volume is under a few thousand. Die casting needs a mold in 15-25 days and starts at 500 pieces; below that, machining from billet is often the cheaper total.
  • The geometry is simple and the tolerance is tight. A machined aluminium plate at ±0.005 mm is a routine part; the same part molded needs a secondary machining operation anyway.
  • The design is still moving. Every engineering change on a machined part is a new model. On a molded part it may be a new tool.
  • The program needs parts this week. 3-7 days with a 48-hour expedite on simple geometry beats a two-week tool build plus T1 approval every time.
  • The quantity is genuinely uncertain. Ordering 200 machined parts and re-ordering 200 more costs more per part than 400 at once, but far less than a tool built for a program that gets cancelled.

Five levers that lower per-part cost at volume

These are the changes that produce a real, repeatable reduction rather than a one-off discount.

1. Cut the number of setups. Every additional setup adds a fixture, a re-datum and a scrap risk, and the cost repeats per part wherever the second operation is manual. Consolidating features onto fewer faces is usually the single largest saving at 100+ pieces.

2. Loosen tolerances that do not need to be tight. ±0.005 mm belongs on the features that mate. A blanket tight tolerance across a part means more inspection, slower feeds and more rework, and it repeats on every unit.

3. Reduce removed material. Starting from stock closer to net shape cuts roughing time, which is pure per-part cost. On a 200-piece run this is worth more than most buyers expect.

4. Batch the finishing. Anodising, bead blasting and passivation are batch processes. A run of 200 finished together costs less per part than four runs of 50, and colour consistency improves.

5. Ask for a production-intent quote, not a prototype quote. A prototype quote is built for one part and one deadline. A 500-piece quote should carry a different fixture plan, a different inspection plan and sometimes a different stock size, and it is worth asking for it explicitly rather than extrapolating.

What to send for a volume quote

Send the STEP file, the target quantity, the material and the tolerances that are functional to /quote/. SkillsProto returns DFM notes and a priced quote within 24 hours, machines to ±0.005 mm on critical features with ISO 2768-m elsewhere, holds MOQ 1 across seven processes and delivers DDP in 3-5 days. If the quantity has crossed the point where tooling is the better route, the quote will say so and quote the tool alongside the machined parts rather than letting the order run on the wrong process.

Frequently Asked Questions

At what quantity does CNC machining stop being economical?

It depends on the part, but the useful planning range is: below 100 pieces machining almost always wins, between 100 and 500 both routes should be quoted, and above 500 in a thermoplastic an aluminium injection tool is usually cheaper per part. Metal parts at a few thousand pieces often stay on CNC because die casting needs a mold in 15-25 days and a 500-piece minimum.

Why did my per-part price not halve when I doubled the quantity?

Because only the fixed half of the quote — programming, fixture, setup, first article — gets diluted. Once that is spread thin, the remaining price is machining time and material, which do not fall with quantity. A genuine volume reduction comes from fewer setups, less removed material or a production-intent fixture plan, not from the quantity field alone.

How many parts should I order for a first production run?

Order enough to cover the next build plus a scrapped-part allowance, and no more. Machined parts have no tooling to amortise, so a second order costs the setup again but nothing else, which makes it cheaper to re-order than to hold inventory you may never use. If the design is still moving, keep the batch small deliberately.

Is vacuum casting cheaper than CNC at 100 parts?

Sometimes. Vacuum casting produces 10-500 copies from a silicone mold in about 7 days, so at 50-100 plastic parts it can undercut machining on unit price while avoiding an injection tool. It is the strongest option when the parts need to look and behave like molded plastic but the design is not frozen enough to cut an aluminium tool.

Can I mix routes in one program?

Yes, and it is often the correct answer. Machining the first 20 parts to validate the design, then moving to aluminium tooling at 7-10 days for the next 500, is a standard pattern. The cost of the machined bridge batch is small compared with the cost of cutting a tool for a design that changes after the first build.

What lead time should I plan for a 500-part run?

Plan 3-7 working days for machining, with 48-hour expedite available on simple geometry, plus 3-5 days of DDP delivery to the UK, EU or North America. At 500 parts the constraint is usually material stock and finishing batches rather than machine time, so confirm stock availability at the quoting stage rather than after the PO.

Related capabilities

Related guides

Have parts to quote?

Upload your files — DFM-checked quotation within 24 hours.

احصل على عرض سعر