Consistent Plastic CNC Processing Solutions
Plastic cuts easily but holds size poorly. This page explains why repeatability breaks down on polymers and what process controls fix it. Read it to judge whether a shop can hold your tolerance across a 10,000 part run, not just on the first article.
In this article
- 1
- 2
- 3
- 4
- 5
- 6
Why consistent plastic CNC processing solutions start with heat
Metals and plastics both expand when warm, but they respond to cutting differently. Aluminum carries heat away quickly, so the chip takes most of the thermal load. Plastics conduct heat poorly, so heat stays near the cut edge. POM and ABS soften within a few degrees of their glass transition, and a dull tool pushes that zone deeper into the wall. The wall then relaxes after the tool passes, and the dimension you measured on the machine is no longer the dimension on the bench.
The practical consequence is that plastic needs more cutting speed and less feed per tooth than the same geometry in aluminum. As a rule, run spindle speeds 2–3× higher than you would for 6061 and keep feed per tooth in the 0.05–0.15 mm range for POM and PC. A sharp, polished carbide tool with a high helix clears chips instead of recutting them. Recut chips are the main source of local heating in deep pockets.
Glass-filled and carbon-filled grades behave differently again. The fibers increase stiffness but also abrasion, so tool life drops sharply. Change tools on a count, not on a feel. A worn tool rubs rather than cuts, and rubbing raises the temperature that started the problem. For PA and PEEK with 30% glass fill, plan on tool changes every 2–4 hours of cutting time and inspect the edge under magnification.
Moisture is the other hidden variable. Nylon absorbs water from the air, and a wet blank machines to a different size than a dry one. For close-tolerance PA parts, dry the stock and keep it dry through the run. PEEK and POM absorb far less, so they are safer when the shop has no drying capacity. If your print calls for PA, ask how the blanks are stored before they reach the machine.
Fixturing is where repeatability is won or lost
Plastic is compliant. Clamp it the way you would clamp steel and it will spring back when you release the vise, leaving a part that is round on the machine and oval on the inspection table. The fix is to spread the clamping force over a large area and keep it low. Soft jaws machined to the actual part profile work better than parallels and a standard vise.
Thin walls are the hardest case. A 1 mm wall on a 100 mm long part will deflect under almost any side load. Support it from behind with a matched pocket, or leave a sacrificial web that you cut away in a second operation. Vacuum fixturing is a good fit for flat panels and housings because the load is uniform, but it needs a smooth sealing surface and enough area to hold the part against the cutting force.
For parts that cannot be held rigidly, flip the problem. Take light finishing passes at low radial engagement instead of one heavy pass. Trochoidal paths keep the radial depth of cut small, which cuts the side load and the heat at the same time. The cycle is longer, but the part stays where you put it.
Sequence matters too. Rough, then let the part rest, then finish. Machining releases locked-in stress from extrusion or injection molding, and the part will creep for a few hours after roughing. A rest period of 4–12 hours between rough and finish removes most of that movement on stock shapes. Skipping the rest is the most common reason a first article passes and batch ten does not.
Thermal control and in-process data
A shop that machines plastic all day has a temperature advantage. The spindle, the coolant, and the room are already at steady state, so the first part of the morning is not 3 °C colder than the last part of the afternoon. That matters because a 100 mm POM part grows about 0.1 mm over a 10 °C swing. If your tolerance is ±0.05 mm, room temperature is a process variable, not a comfort setting.
Coolant choice is a trade-off. Flood coolant removes heat well but can be absorbed by some grades, and it needs drying before inspection. Compressed air with a cold-air gun is cleaner and works well for POM, ABS, and PC. For PEEK and other high-temperature grades, air alone may not be enough, and a mist system keeps the edge cool without soaking the part.
Measurement is where many runs quietly fail. Plastic is soft, so a micrometer with normal friction will compress the surface and read undersize. Use a constant-force gauge or a non-contact method, and measure at the same temperature the part was machined at. Write the gauge type and the measurement temperature into the inspection plan so the next operator does it the same way.
In-process checks catch drift before it becomes scrap. Measure the first part, then every tenth part, and log the numbers. A slow trend is a signal that the tool is wearing or the room is warming. A step change usually means a fixture slipped or a new lot of stock entered the cell. Both are cheap to fix early and expensive to fix after the run ships.
When 5-axis helps and when it does not
Five-axis machining earns its place on plastic parts with contoured surfaces or features on several faces. Doing the same part in three setups on a 3-axis machine means three chances to introduce a locating error, and plastic parts are less forgiving of re-clamping than metal. One setup on a 5-axis center removes those errors and usually improves the surface finish on curved walls because the tool stays normal to the surface.
It does not help everywhere. A flat bracket with holes on one face runs faster and cheaper on a 3-axis mill, and the setup is simpler to inspect. Choosing five-axis for a part that does not need it adds cycle time and cost without improving the dimension. The question to ask is whether the part has features that cannot be reached in one orientation, or whether the tolerance stack across setups is tighter than the machine can hold.
Simultaneous 5-axis also lets you use a shorter, stiffer tool. A short tool deflects less, which matters on thin ribs and deep pockets. On a tall part, the difference between a long reach tool in a 3-axis setup and a short tool tilted into the cut can be the difference between holding ±0.05 mm and chasing the dimension all run.
For parts up to 4,000 mm long, a large 5-axis machine with a Ø400 mm rotary table covers most plastic housings, ducts, and panels. Below that, compact centers with 500 × 500 × 450 mm travels handle the majority of brackets, manifolds, and covers. Match the machine to the part envelope rather than to the shop's biggest asset.
What to ask before you release a plastic run
Ask for the process plan, not just the tolerance. A shop that can explain its speeds, feeds, fixture design, and rest periods understands why plastic is different. A shop that quotes the same parameters it uses on aluminum has not thought about the material.
Ask how the first article is validated and what happens when the tenth part drifts. The answer should include a measurement method, a sampling frequency, and a defined action. Vague answers here are the reason plastic runs fail after a good prototype.
Ask about stock storage. Nylon and other hygroscopic grades need controlled humidity, and the difference between a dry blank and a wet one shows up on the print. If the shop cannot describe how it stores and dries material, close-tolerance PA work is a risk.
Finally, ask for the inspection report with the shipment. Dimensional data on the actual parts, at the actual temperature, tells you whether the run was controlled or just lucky. Reports on request are standard; a shop that treats them as routine is easier to audit.
Matching process choices to part requirements
Use this table to check whether a quoted process matches your part, not to pick a machine by habit.
| Part condition | Preferred approach | Watch out for |
|---|---|---|
| Flat bracket, holes on one face | 3-axis mill, soft jaws | Over-clamping thin sections |
| Contoured housing, features on 3 faces | 5-axis, one setup | Tool reach on deep pockets |
| Wall under 1.5 mm | Light finishing passes, support behind wall | Deflection during release |
| Tolerance tighter than ±0.05 mm | Temperature-controlled room, rest between ops | Room drift across the shift |
| Nylon or other hygroscopic grade | Dried stock, sealed storage | Moisture pickup between ops |
| Glass or carbon filled compound | Count-based tool changes | Abrasion and edge wear |
| Large panel or duct, up to 4,000 mm | Large 5-axis with rotary table | Fixture sag over long spans |
| High-temperature grade like PEEK | Mist cooling, sharp polished tools | Heat buildup at the edge |
The honest trade-off
If your plastic part is flat and simple, choose a 3-axis setup with soft jaws and spend the savings on inspection. If it has contoured surfaces, features on several faces, or a tolerance stack that crosses setups, choose a 5-axis process and accept the higher hourly rate. The wrong choice costs more than the right one either way.
Common questions on plastic machining
Can plastic parts really hold ±0.005 mm?
Yes, on stable grades and small features, in a temperature-controlled room, with the right fixture and a rest period between roughing and finishing. POM, PEEK, and filled grades are the realistic candidates.
On large parts or hygroscopic grades like nylon, the practical limit is looser because moisture and thermal movement dominate. We will tell you which limit applies to your geometry before quoting.
Why does my first article pass but later parts drift?
The usual causes are tool wear, room temperature change during the shift, and stress relief in the stock after roughing. None of these show up on part one.
A process plan that includes a rest period, count-based tool changes, and sampling every tenth part catches all three before they turn into scrap.
Is coolant needed for plastic?
Not always. Compressed air with a cold-air gun handles POM, ABS, and PC well and keeps the part dry for inspection.
For PEEK and other high-temperature grades, a mist system keeps the cutting edge cool without soaking the part. Flood coolant works but needs a drying step before measurement.
How do you measure a soft plastic part without compressing it?
Use a constant-force gauge or a non-contact method, and measure at the same temperature the part was machined at. A standard micrometer with normal friction will read undersize on POM and ABS.
The gauge type and measurement temperature belong in the inspection plan so every operator measures the same way.
What is the smallest order you will run?
There is no minimum order quantity. We run from one prototype to 10,000+ part runs, and the process controls above apply to both ends of that range.
For a single prototype we still record the parameters, because that record is what makes the production run repeatable later.
Do you sign an NDA for plastic parts?
Yes. Uploads are secure and confidential, and an NDA is available on request before you share drawings.
That matters most for medical and automotive plastic components, where the geometry itself carries design intent.
Send us the plastic part and the tolerance
We will review the geometry, the material, and the tolerance stack, then quote a process that can repeat it. Quotation and free DFM analysis within 12 hours.
12-hour quote100% inspectionNDA availableNo MOQ
More from GreatLight
We publish setup notes, tooling trials and inspection data from the factory floor.
