---
title: "Choosing a 3D Printing Material: A Selection Guide"
description: "The process picks the material shortlist, not the other way around. Properties by process, what each material survives, and the four questions that decide it."
canonical: https://openspindle.com/blog/3d-printing-material-selection
author: "Tom"
datePublished: 2026-09-05T00:00:00.000Z
dateModified: 2026-09-05T00:00:00.000Z
category: "Materials"
---

# Choosing a 3D Printing Material

Pick the process first, because each process can only run its own family of materials, then pick within that shortlist. FDM gives you engineering thermoplastics but weak layer bonds. Resin gives fine detail and smooth surfaces in materials that are stiff, brittle and degrade under UV. SLS gives tough, effectively isotropic nylon with a matte grainy surface. The four questions that decide it are load, temperature, environment and cosmetics.

## The process constrains the shortlist

This is the step people skip. You cannot print PA12 nylon on a resin machine or a tough photopolymer on a filament machine. Choosing a material before choosing a process usually means choosing again.

So the useful order is: decide what the part has to survive, let that pick the process, then choose within the two or three materials that process actually runs well. A supplier who offers thirty materials still only offers a handful per machine.

The corollary is that a material name alone is not a specification. PETG on one machine at one layer height is a different part from PETG on another, because the layer bond is a process outcome rather than a material property.

## Materials by process, and what each is for

The table covers what suppliers actually stock and run in production, rather than everything that theoretically exists.

| Material | Process | Strengths | Watch out for |
| --- | --- | --- | --- |
| PLA | FDM | Cheapest, dimensionally stable, easy | Softens near 60 C, brittle, not for functional parts |
| PETG | FDM | Tougher than PLA, chemical resistant, food safe grades | Stringing, moderate heat resistance |
| ABS | FDM | Impact resistant, 90 C service, vapor smoothable | Warps on large flat parts, needs an enclosure |
| ASA | FDM | ABS properties plus genuine UV stability | Same warping tendency as ABS |
| Nylon (PA) | FDM | Tough, wear resistant, good living hinges | Absorbs moisture, needs drying, shrinks |
| Polycarbonate | FDM | Highest strength and heat in common FDM | Hard to print, warps, needs high temperatures |
| Standard resin | SLA / DLP | Finest detail, smooth surface, appearance models | Brittle, embrittles further under UV over months |
| Tough / ABS-like resin | SLA / DLP | Better impact resistance than standard resin | Still ages under UV, lower detail than standard |
| High temperature resin | SLA / DLP | Survives well above 100 C | Very brittle, expensive |
| PA12 nylon | SLS | Tough, isotropic, no supports, production ready | Grainy matte surface, higher cost per part |
| PA11 nylon | SLS | More ductile than PA12, better for hinges and clips | Slightly lower stiffness, higher cost |
| Glass filled nylon | SLS | Much stiffer, better heat resistance | More brittle, abrasive, heavier |

## The four questions that decide it

Load comes first. If the part carries a real structural load, FDM is a liability unless you control orientation, because the bond between layers is weaker than the material along them. SLS is the default for functional parts for exactly this reason.

Temperature is second and is where most material choices fail in service. PLA loses stiffness around 60 C, which a part left in a car or near a motor will reach. ABS and ASA get you to roughly 90 C, polycarbonate higher, and high temperature resins higher still.

Environment is third. UV exposure rules out most standard resins over any real timescale, and it is the failure mode people are most surprised by, because the part looked fine for a month. Moisture matters for nylon, which absorbs it and changes dimension. Chemical contact narrows the list quickly.

Cosmetics is last, and it is the only one where resin wins outright. If a customer will hold the part, SLA off the machine looks closer to injection molded than anything else on the list.

## Where material choice meets orientation

On FDM the material specification is incomplete without an orientation, and this is the single most common cause of a printed part failing unexpectedly.

A nylon FDM part printed with the load across the layers can fail at a fraction of the strength the datasheet implies, because the datasheet reports the material and the part reports the weld between beads. The material did not lie; the geometry was oriented wrong.

SLS and resin are far less sensitive, SLS because the powder fuses in all directions and resin because each layer bonds chemically into the one below. If you cannot control or communicate orientation, that alone is an argument for moving off FDM.

Say which direction the load comes from when you send the file. It costs a sentence and it changes what the supplier does.

## What this costs you

Material is rarely the dominant cost on a printed part. Machine time and post-processing usually are, and material choice drives both.

A resin part needs washing, support removal and post-curing. An SLS part needs depowdering and often bead blasting and dyeing. An FDM part in ABS may need vapor smoothing to look acceptable. Those steps are labor, and on cosmetic parts the finishing frequently exceeds the printing.

The genuinely expensive mistake is specifying a premium material for a property the part never uses. Glass filled nylon on a bracket that carries no load, or high temperature resin on a part that lives at room temperature, buys nothing and costs on every unit.

The cheap win runs the other way: on a functional part at any volume, moving from FDM to SLS often costs less per part than the finishing labor FDM would need to reach the same standard.

| Requirement | Reach for | Rather than |
| --- | --- | --- |
| Cheap form and fit check | PLA on FDM | Anything engineered |
| Functional part under load | PA12 on SLS | FDM in any material |
| Customer facing appearance model | Standard resin on SLA | FDM with heavy finishing |
| Outdoor or UV exposure | ASA on FDM, or SLS nylon | Any standard resin |
| Living hinge or repeated flexing | PA11 on SLS | PLA or standard resin |
| Above about 90 C | Polycarbonate, or high temperature resin | PLA, PETG, standard resin |
| Stiff bracket, low deflection | Glass filled nylon on SLS | Unfilled nylon |
| Chemical or fuel contact | Nylon, or a chemically rated resin | PLA, ABS |

## Datasheet numbers and what they leave out

Printed material datasheets report properties from test specimens printed under ideal conditions in the strongest orientation. Your part is neither.

Treat published tensile strength as an upper bound rather than a design value, particularly on FDM. A common working approach is to design against the across-layer figure where the supplier publishes one, and to add margin where they do not.

Heat deflection temperature deserves particular care, because it is quoted at a specified load and a part under a different load behaves differently. A material listed at 90 C is not a material that holds its shape at 89 C under any circumstances.

Where a printed part is genuinely load bearing and failure matters, the honest answer is to test the actual part in the actual orientation rather than to trust a table.

## When printing is the wrong answer entirely

Material selection sometimes reveals that no printed material fits, and it is worth recognizing that early.

If the part needs a tolerance tighter than roughly plus or minus 0.1 mm, needs a genuine engineering metal, needs certified material traceability, or needs a surface finish better than a printed process gives without heavy manual work, machining is the answer and no material choice will close the gap.

If you are ordering the same part repeatedly in the hundreds, the material question is often really a process question, and injection molding gives both better material properties and a lower unit cost once the tool is paid for.

Printing earns its place on complexity, lead time and the absence of tooling. When none of those three is the reason you are printing, quote the alternatives.

## What to send, and what to ask

Send the model with three facts the geometry cannot carry: the temperature range the part will see, whether it takes load and from what direction, and whether the surface is cosmetic. Say whether it lives outdoors.

Give your quantity now and your expected annual quantity, because the honest material recommendation at ten parts differs from the one at a thousand.

Then ask three questions. Which materials do you actually run in production on this process, as opposed to offer? What orientation will my part be built in? And what post-processing is included in this price, quoted separately from the printing?

## Related reading

- [FDM vs SLA vs SLS](https://openspindle.com/blog/fdm-vs-sla-vs-sls.md)
- [3D printing capabilities](https://openspindle.com/capabilities/3d-printing.md)
- [3D printing tolerance chart](https://openspindle.com/reference/3d-printing-tolerance-chart.md)
- [Plastic properties comparison](https://openspindle.com/reference/plastic-properties-comparison.md)

## Frequently asked questions

### What is the strongest 3D printing material?

For most functional parts, glass filled PA12 nylon on SLS, which combines stiffness and toughness with effectively isotropic behavior. Among FDM materials polycarbonate is the strongest but is difficult to print and still weak across layers. Published tensile figures come from ideal specimens in the best orientation, so treat them as an upper bound rather than a design value.

### Which 3D printing material is best for outdoor use?

ASA on FDM or nylon on SLS. Both hold up under UV. Avoid standard photopolymer resins entirely for outdoor parts: they embrittle under UV over weeks to months, and the part often looks fine right up until it cracks. PLA is also a poor outdoor choice because it softens in a hot car or in direct sun.

### Can 3D printed parts be used in production, or only prototypes?

They are used in production regularly, most often in SLS nylon, where isotropic strength and no support scarring make parts consistent enough to ship. The economics decide it: printing wins on complexity, lead time and no tooling. Once you reorder the same part in the hundreds, quote it against machining and injection molding, because the tool usually pays for itself.

### Why did my printed part fail below its published strength?

Usually orientation on an FDM part. The datasheet reports strength along the beads, and a load applied across the layers is resisted only by the weld between them, which can be around half as strong and fails brittly. Tell the supplier which direction the load comes from, or move the part to SLS where the material fuses in all directions.

### Does material choice affect the cost of a printed part much?

Less than people expect. Machine time and post-processing usually dominate, and material choice drives both: resin parts need washing, support removal and post-curing, SLS parts need depowdering and often blasting and dyeing. Specifying a premium material for a property the part never uses adds cost on every unit and buys nothing.

### Is nylon or resin better for a functional part?

Nylon on SLS, in most cases. It is tough, effectively isotropic and UV stable, which is why it is the usual choice for end use printed parts. Resin gives a far better surface and finer detail, so it wins for appearance models, master patterns and fine featured parts, but standard resins are brittle and age poorly under UV.