How to 3D Print Reliable Functional Parts With PETG

PETG has a reputation problem in production shops. Half the operators I talk to run it for everything structural. The other half switched back to PLA after a batch of brackets came off the plate looking fine and split along a layer line three weeks later.

Both experiences are real. PETG is a good functional material and a fussy one, and the gap between those two outcomes is almost never the spool brand printed on the label. It is moisture, wall count, and layer temperature, in roughly that order.

This is what we have learned running the material ourselves and from print farms who buy it by the pallet.

What "functional" actually means before you pick a material

Functional is a vague word. Before choosing PETG over anything else, write down which of these the part has to survive:

• A sustained load, or an occasional impact, or both
• The highest temperature it will ever see, including a closed car in summer
• Chemicals it touches: coolant, IPA, cutting fluid, cleaning agents, skin oils
• UV exposure
• How tight the fit has to be, and whether it mates with a metal part

Most brackets, jigs, enclosures and fluid-adjacent parts land inside PETG's envelope. Anything that spends its life above roughly 70 °C, or takes repeated hard impacts, does not, and no amount of print tuning will move it there.

PETG against PLA and ABS

The honest version of the comparison, with the caveats that matter:

Property	PLA	PETG	ABS / ASA
Tensile strength	High on paper	Lower than PLA, tougher in practice	Lowest of the three
Behaviour at failure	Snaps	Bends, then tears	Bends, then cracks
Heat tolerance	Poor, softens near 55–60 °C	Usable to about 70 °C	Good, 90 °C and up
Chemical resistance	Poor	Good against water, dilute acids, most alcohols	Poor against solvents, dissolves in acetone
Warping	Minimal	Low	High, wants an enclosure
Moisture sensitivity	Moderate	High	Moderate
Fumes	Low	Low	Needs ventilation

PLA's tensile numbers beat PETG in almost every published test, which confuses people. Those numbers come from a tensile bar pulled slowly in one axis. Your bracket does not live in a tensile tester. It gets dropped, over-torqued, left in a warm room, and loaded at an angle across layer lines. PETG stretches before it lets go, and that margin is worth more than a headline MPa figure.

ABS and ASA still win on heat, and ASA wins outdoors. PETG holds up better in sunlight than PLA but it is not a UV material, and unpigmented PETG yellows and goes brittle outside over a season or two.

Drying is not optional

If you take one thing from this: wet PETG is the single largest cause of weak functional parts, and it is invisible until you break one.

PETG absorbs moisture from the air faster than PLA. When that water hits the melt zone it flashes to steam inside the nozzle. You get the obvious symptoms first, which are popping sounds, stringing, and a rough surface finish. The one you cannot see is what matters: the steam disrupts the weld between the extruding bead and the layer below. Parts come out dimensionally correct and mechanically weak. They pass a visual inspection, ship, and fail at the customer.

Typical drying for PETG is 65 °C for four to six hours, and longer if the spool has been open on a shelf for weeks. Check your supplier's stated schedule rather than assuming, since additive packages differ between grades.

Some practical notes from farm operators:

• Dry before the first print of a new spool, not just when prints look bad. A sealed spool is dry at packing, not necessarily at unboxing.
• Print out of a dryer or a dry box if your shop humidity runs above 40 percent. Drying once and leaving the spool on a rack undoes the work in a day or two.
• A hot plate under a cardboard box is not a dryer. You need airflow and a stable temperature, and cardboard spool sides will absorb water and give it back to the filament.
• If a print sounds like frying bacon, stop it. You are not going to save that part.

Do not exceed about 70 °C when drying PETG on a plastic spool. The spool deforms before the filament does, and a warped spool binds the feed halfway through the next long job.

Temperature, and why hotter usually means stronger

PETG generally runs in the 230–250 °C range at the nozzle with a bed around 70–85 °C, though every hotend and every grade sits somewhere different inside that band. Run a temperature tower for your specific combination. It takes twenty minutes and answers the question properly.

The thing people get wrong is optimising for surface finish. Lower temperature and more cooling gives crisper corners and fewer strings, and it also gives noticeably worse layer adhesion. For a display piece that trade is fine. For a load-bearing bracket it is backwards.

For functional parts, bias hot. Take the top of your working temperature range, run part cooling at 20 to 40 percent instead of full blast, and accept slightly softer overhangs. The part will be stronger in the Z axis, which is the axis it is going to fail in.

Watch flow rate while you do this. Pushing PETG hot and fast at the same time can outrun the hotend's ability to melt it, and under-extrusion looks a lot like weak layer bonding in a broken part.

Walls do the work, not infill

The instinct when a part breaks is to raise infill. It is usually the wrong lever.

In a printed part under bending or torsion, most of the stress sits in the outer shell. Going from three walls at 20 percent infill to three walls at 60 percent adds material, print time and internal stress, and buys less strength than going from three walls to five at the same infill.

For functional PETG, a reasonable default is four to six perimeters, 20 to 30 percent gyroid or cubic infill, and five top and bottom layers. Then think about the part rather than the slider:

• Orient so the main load runs along layers, not across them. This is worth more than every setting in this article combined.
• Add fillets at internal corners. Sharp internal corners are where printed parts crack, the same as machined ones.
• For threaded fixings, use heat-set inserts or captive nuts. PETG threads cut directly into the plastic will strip under real torque.
• Keep wall thickness a whole multiple of your extrusion width so the slicer does not leave a gap down the middle of a thin wall.

Where filament consistency shows up

Settings can only compensate for so much. At some point the filament itself is the variable.

Your slicer calculates extrusion volumetrically. It assumes the filament entering the extruder is exactly the diameter you told it, usually 1.75 mm. If the actual diameter drifts, the volume of plastic coming out drifts with it, and nothing in the machine corrects for that.

Do the arithmetic on a 0.1 mm swing, which is not unusual on cheap filament. Diameter at 1.85 mm instead of 1.75 mm is about 12 percent more cross-sectional area, so roughly 12 percent more material through the nozzle for the same commanded movement. On a 0.45 mm wall you are now laying down a 0.50 mm bead. Over four perimeters, that is a wall that is wider than designed, and a bore that is tighter than designed, on a part where the fit was the point.

The other direction is worse. A thin section under-extrudes, walls come out narrow, and the bond between beads gets thinner right where you needed the shell to carry the load. That is a strength problem disguised as a cosmetic one.

This is why we publish a dimensional tolerance of ±0.02 mm on our filament rather than a general claim about quality. It is a number you can check yourself with a caliper and a few random measurements along a spool, and you should check it, on ours and on whatever you currently run. We also supply an SDS, a RoHS statement and a REACH statement per material for shops that need documentation on file.

Tight tolerance does not make a badly oriented part strong. It removes one variable, so that when you tune a profile on Monday it still holds on Friday, and on the next spool.

Speed, and the annealing question

Fast PETG is possible on a machine with a high flow hotend and a stiff frame. On a stock bowden printer it mostly produces parts that measure fine and delaminate under load, because the melt never fully caught up. If throughput matters, buy flow rate rather than turning the speed slider up on hardware that cannot feed it.

The compromise most farms settle on is normal speed for the infill and slower perimeters, since the perimeters are carrying the load anyway. Dropping outer wall speed to around half of infill speed costs a few minutes per part and gives a visibly better bead.

Annealing comes up often because it works so well on PLA. It does much less for PETG. PETG is largely amorphous, so there is no crystalline structure to develop with a heat soak, and you mostly get dimensional change without a meaningful strength gain. If a part needs more than PETG gives after tuning, changing the design or the material is a better use of the afternoon than baking it and re-measuring.

Common PETG failures and what causes them

Part splits cleanly along a layer line

Layer adhesion. Nozzle too cold, cooling too aggressive, wet filament, or all three. Raise temperature, cut fan, dry the spool.

Stringing and blobs everywhere

Usually moisture first, retraction second. Dry the spool before touching retraction settings, otherwise you will tune a profile around a problem that goes away on its own.

First layer will not stick, or sticks too well

PETG bonds aggressively to smooth PEI and to glass, sometimes hard enough to tear a chunk out of the sheet. A textured sheet, a thin layer of glue stick as a release agent, or a slightly larger first layer gap all help. Raising Z a touch on the first layer is the standard fix and it does not hurt anything structural.

Part fits on the bench, binds in service

Thermal expansion, creep, or a diameter shift mid-spool. Check the fit again after the part has been at working temperature for an hour under load.

Part survives assembly, fails weeks later

Creep. PETG deforms slowly under sustained load, especially warm. If a part is permanently clamped or hangs a weight, design in more material than the stress calculation says, or move to a material with a higher Tg.

Test the part, not the print

A functional part is qualified when you have broken one on purpose. Print two extra, load them the way the real part gets loaded, and take one to failure. Look at the fracture surface. If it is a clean flat break along a layer, you have an adhesion problem and the fix is in this article. If the plastic stretched and tore, the material was working properly and you need more material, better orientation, or a different polymer.

Do it once per profile change and once per new spool batch on parts that matter. It takes an hour and it is the only way to know whether Tuesday's brackets are as good as last month's.

Quick reference

• Dry at 65 °C for four to six hours, and print out of a dry box in humid shops
• Nozzle 230–250 °C, bed 70–85 °C, confirmed with a temperature tower
• Part cooling 20 to 40 percent, not full
• Four to six perimeters, 20 to 30 percent infill, five top and bottom layers
• Orient the part so load runs along layers
• Heat-set inserts for anything that gets torqued
• Check filament diameter with a caliper before blaming the printer

Frequently asked

Is PETG stronger than PLA?

Not in a tensile test. PETG is tougher, which means it absorbs more energy before failing and bends rather than snapping. For functional parts that get handled, dropped or over-tightened, that behaviour matters more than peak tensile strength.

Do I have to dry PETG every time?

Dry any spool that has been open to room air for more than a few days, and dry a new spool before a job you cannot afford to reprint. In a shop above 40 percent humidity, printing from a dry box is simpler than repeated drying cycles.

What temperature can PETG parts handle?

Around 70 °C before it softens, lower if the part is under load at the same time. Above that, look at ABS, ASA or a nylon.

Can I use PETG outdoors?

For a season, yes, with pigment in it. For long term outdoor use, ASA is the better choice. PETG loses impact strength under prolonged UV.

Why do my PETG parts fit correctly but feel weak?

Almost always moisture, cold printing, or too much part cooling. Dimensional accuracy and layer adhesion are separate problems, and a part can be perfect on the caliper and poorly welded inside.

See VANYO PETG, or ask us for a sample spool if you are qualifying a second supplier.