Maximum Bridging Distance Without Supports in FDM 3D Printing: Limits, Tests, and Settings That Actually Work

The honest answer most slicer guides skip

The longest bridge a typical FDM 3D printer can produce without supports is somewhere between 30 and 80 millimetres. The wide spread is not because nobody has measured it. It is because the maximum bridging distance is not a property of the printer alone — it depends on the filament you are using that day, the cooling fan you have installed, the part cooling duct around your hot end, the nozzle temperature you happened to land on, and the slicer’s bridge-specific overrides. Treat any single number you read online as a starting hypothesis, not a hardware spec.

That said, useful bands exist. A well-tuned Bambu Lab P1S or X1C running PLA at 210°C with the chamber lid cracked open can routinely bridge 60-80mm without support material if the slicer drops to 25mm/s for the bridge layer and runs the part fan at 100%. A stock Ender 3 V2 running the same PLA might reliably bridge 30-40mm before the strands start sagging visibly into the cavity below. Same physics, different cooling envelope and different mechanical rigidity in the printhead. The number you can hit is the number your specific machine produces consistently across three test prints in a row.

This article walks through the variables that actually move the bridging limit, a concrete test procedure to find your printer’s real number, and the slicer settings that have the biggest effect on whether a long bridge prints clean or sags into a fuzz of stringy filament threads.

Why bridging works at all (and where it stops working)

An unsupported bridge is a horizontal extrusion drawn across an air gap, where the molten plastic must cool from melt temperature down past the glass transition temperature before gravity pulls it more than a fraction of a millimetre. PLA’s glass transition is around 60°C and its melt extrudes at 200-215°C, leaving a roughly 140°C gap that the part fan must close before the strand drops. The longer the bridge, the more time each portion of the strand spends in that vulnerable warm-but-not-yet-solid window, and the more sag accumulates as gravity pulls each unsupported segment downward.

Three things determine whether a strand survives the journey. First, how fast the printer moves: slower bridges give the fan more time to cool each segment. Second, how much air the fan can move across the strand at the moment it leaves the nozzle: better-engineered cooling ducts (Bambu, Voron, Prusa MK4) cool a strand faster than most stock budget printers. Third, the polymer’s behaviour in the cooling window: PLA solidifies fast and clean, PETG remains rubbery for longer and sags more, ABS sags badly without a heated chamber, and TPU bridges hardly at all because it never properly stiffens.

Below those three primary factors, secondary ones matter at the margins: the slicer’s bridge flow rate (over-extruding sags more), the bridge layer line width (thinner strands cool faster), the part geometry (a single straight bridge is easier than a bridge inside a complex cavity that the fan cannot reach), and humidity in the spool (wet PETG can balloon as moisture flashes off in the melt). If you control the primary three, the secondaries usually take care of themselves with default slicer settings.

The variables that actually set the real limit

Bridging distance scales roughly linearly with cooling effectiveness up to a printer-specific ceiling. Doubling part-fan airflow does not necessarily double the maximum bridge — at some point the strand has cooled enough during nozzle travel that more fan does not help. But within the range most stock printers operate, fan upgrades and cooling-duct redesigns are the single biggest lever. A Voron with a stealthburner or a Prusa MK4 with the redesigned hot-end shroud both bridge measurably further than the same hardware would with a stock Ender 3 fan layout, even when running identical filament and slicer settings.

Print speed during the bridge is the next biggest factor. Slowing from 60mm/s to 25mm/s during a bridge gives each strand 2.4× the cooling time before the next strand lands beside it. PrusaSlicer, OrcaSlicer, and Cura all expose a “bridge speed” override that lets you slow only the bridge moves while keeping the rest of the print fast. Use it. The default values (50% of normal speed in OrcaSlicer, 25mm/s in PrusaSlicer) are usually fine; values lower than 20mm/s start to hurt as the molten strand cools too far before the next layer bonds to it.

The third factor is filament choice for the bridge. PLA bridges best, PETG bridges acceptably with reduced flow, ABS bridges poorly without an enclosure, and TPU effectively cannot bridge unsupported beyond about 5mm. If the part requires long bridges and is otherwise material-agnostic, default to PLA. If the part needs PETG for mechanical reasons, accept that bridges over 30mm will likely sag and either add support material in those specific spots or redesign the part to avoid them.

How to test your printer’s real bridging max

Download the Thingiverse “bridge torture test” model (the staircase-style ramp of progressively longer bridges from 10mm to 100mm in 10mm increments) and print it on the filament and slicer profile you intend to use for the actual project. Use your normal print settings except for the bridge override: set bridge speed to 25mm/s, bridge fan to 100%, and bridge flow ratio to 0.95 (slightly under-extruded so strands have a touch more room to cool).

Print the test horizontally so the bridges are oriented across the bed rather than stacked. After printing, examine each bridge from below. The longest bridge with no visible sag and no broken strands is your real maximum. Subtract 20% to give yourself a margin for filament-batch variation and call that your design ceiling. If your test maxes out at 60mm with no sag, design parts to bridge no more than 48mm without supports.

Run the test once per printer per filament brand. PLA from Polymaker bridges differently than PLA from Hatchbox, even at identical printer settings. Document the result in a small spreadsheet alongside the slicer profile, so the next time you start a project on that printer you do not need to retest from scratch.

Slicer settings that move the limit further

In OrcaSlicer or PrusaSlicer, the bridge-specific settings live under “Speed” and “Quality”: Bridge speed (25mm/s default), Bridge flow ratio (0.95 default in Orca, 1.0 in Prusa — drop to 0.85-0.95 if your bridges sag), Bridge fan speed (100% always), and the often-overlooked Thick bridges toggle in PrusaSlicer (turn it off; it produces more reliable thin bridges on most printers).

For Cura users the equivalent settings live under “Experimental” — turn on “Bridge Settings” to expose them. Cura’s defaults are conservative; you can usually pull bridge speed down further than the default 50% and see improvement on long spans. Watch for the bridge-skin density setting: lower densities (around 80%) print thinner strands that cool faster, which is what you want for the longest bridges.

One slicer trick that works on stubborn long bridges: increase the layer height for the bridge layer specifically. A 0.2mm bridge layer cools faster than a 0.3mm bridge layer because there is less material per millimetre of strand. Most slicers do not expose this directly, but if you are designing the part you can make the bridge a 0.2mm-tall surface and let the slicer’s variable layer height feature do the rest.

Material-by-material expectations

PLA: 60-80mm typical max on a well-tuned modern printer, 30-50mm on a stock budget printer. The forgiving polymer for bridging, period.

PETG: 30-50mm typical max with bridge flow dropped to 0.85 and bridge speed at 25mm/s. PETG’s longer cooling window means it sags faster than PLA on equivalent spans. Plan for supports above 40mm.

ABS: 25-40mm in an enclosed printer with chamber temperature around 50-60°C. Lower outside an enclosure because the strand cools too slowly in the open environment. ABS bridging is unreliable enough that designing around it is usually the right call.

ASA: similar to ABS, slightly worse because ASA tolerates less aggressive part fan during bridging without warping the surrounding wall structure.

TPU: do not attempt unsupported bridges over 5mm. The polymer never stiffens enough during the cooling window to support its own weight. Use supports always, or reorient the part.

PA (Nylon): 20-30mm at best, and usually with significant surface artifacts. Nylon bridging is difficult enough that most experienced nylon users design supports into the part geometry rather than relying on slicer-generated supports.

When the right answer is “stop trying to bridge that”

Some bridges should not be printed unsupported even when your printer can technically pull it off. A 70mm bridge across a part’s structural cavity that bears load below it is structurally weaker than the same span printed over solid infill or with proper supports — the bridge layer’s interlayer adhesion is always slightly worse than ordinary layers. For decorative parts the visual sag is acceptable; for functional parts it is not.

The other case where unsupported bridging is the wrong choice is high-precision dimensional work. Even a clean unsupported bridge has a small downward bow at its centre because the molten strand stretches before fully solidifying. For miniature work, mechanical fits, or any part where ±0.2mm dimensional accuracy matters across the bridged dimension, use supports and accept the post-processing time.

The last case: if the bridge can be reoriented to print vertically as a wall instead of horizontally as a span, that is almost always the better choice. A 100mm-tall wall is trivial to print on any FDM machine. A 100mm horizontal bridge is at the edge of what is even possible. Spend ten minutes redesigning the part to bridge a smaller distance — or to not bridge at all — before you spend an hour fighting the slicer.

The takeaway numbers

Test your printer once. Document the result. Design under 80% of that number. PLA bridges further than PETG, which bridges further than ABS, which bridges further than TPU. Bridge speed 25mm/s, bridge fan 100%, bridge flow 0.85-0.95 — these three settings move more bridging quality than any other slicer change you can make. Above 60mm on PLA, or above 40mm on PETG, you are at the edge of what is reliable across multiple prints. At those distances, supports cost a few minutes of post-processing time and save hours of failed-print frustration.