Layer adhesion is a thermal problem before it is anything else. Every new layer has to briefly re-melt the surface of the layer beneath it so the polymer chains diffuse across the boundary and entangle. If the fresh extrudate cools below the material’s glass transition temperature before that diffusion happens, the bond is weak — full stop. That single mechanism explains why prints that look flawless snap cleanly along the layer lines under load: the geometry is fine, but the layers never actually welded.
So to fix poor layer adhesion strength you are really tuning one thing from several angles: how much heat is available at the layer interface, and how long it stays there before the plastic freezes. Below are the levers in rough order of impact, with real numbers and the trade-offs a practitioner needs.
Nozzle temperature: the biggest single lever
Raising nozzle temperature usually has the largest effect on interlayer strength, because it directly increases the thermal energy available to soften the layer below and drive chain entanglement. The practical move: add 10 °C and reprint. If you run PLA at 200 °C, try 210 °C. If ABS is at 240 °C, try 250 °C. Most published “maximum” temperatures are conservative, so printing at the upper end of a filament’s range is generally safe and improves bonding.
Rough starting points, not gospel:
| Material | Typical nozzle range | If layers separate |
|---|---|---|
| PLA | 200–215 °C | +5–10 °C at a time from ~210 °C |
| PETG | 230–250 °C | +5–10 °C, watch for stringing |
| ABS/ASA | 220–250 °C | Push toward 250 °C with an enclosure |
There is an upper bound to respect. Too hot and you get stringing and mushy definition, and with some composites you introduce voids — one study on PLA/carbon-fibre found bubble-like structures forming at higher nozzle temperatures and kept its prints below 220 °C for that reason. So temperature must stay inside the material’s usable window: hot enough to weld, not so hot that it foams or degrades.
The part-cooling fan: the number one cause of weak Z-strength
If temperature is the biggest positive lever, the part-cooling fan is the biggest negative one. The exact cooling that stops overhangs sagging also freezes each new layer before it can bond. Blast room-temperature air across material that is contracting as it cools from 250 °C and you build in layer-line failures that never show on the surface.
Match the fan to the material — this is where most weak parts come from, usually because someone copied a PLA profile onto ABS:
| Material | Fan speed (after first layers) | Notes |
|---|---|---|
| PLA | 80–100% (drop to 60–70% if delaminating) | Tolerates and benefits from aggressive cooling |
| PETG | 30–50% | Too much cooling causes delamination and cloudiness |
| ABS/ASA | 0–20% (fan off, 15% max) | Needs an enclosure, not a fan |
For the first 3–5 layers of any material, reduce or disable the fan entirely. Those layers set the foundation and want maximum interlayer bonding — plus better first-layer adhesion in general. If your first layer is the problem area, our guide to first layer adhesion problems covers the bed side of that equation.
Print speed and dwell time
Speed and temperature are coupled. Faster printing means each blob of plastic spends less time in the hot zone and arrives at the bed slightly cooler, and the nozzle lingers less over each point — less heat delivered, less dwell time for diffusion. Push speed up without raising temperature and adhesion drops.
Two fixes: either reduce print speed by around 25% so the hot nozzle spends more time over each area, or — if you want the speed — raise nozzle temperature by 5–10 °C for every significant speed increase to compensate. When the nozzle moves slowly it holds heat over the interface long enough for diffusion to happen before the temperature falls below Tg.
Line width: the most underrated strength setting
Layer height gets all the attention; line width does more for strength. Increasing extrusion width beyond nozzle diameter — from 0.4 mm to 0.45–0.5 mm on a 0.4 mm nozzle — improves adhesion three ways: more contact area between adjacent lines on the same layer, more squish of each line onto the layer below, and a denser weld overall. Set extrusion width to roughly 110–125% of nozzle diameter and re-test.
A wider nozzle achieves the same thing more aggressively: more material through the tip means fatter layers with more surface area for the next layer to grip. If you print functional parts and don’t need fine detail, a 0.6 mm nozzle at a wider line width is a genuine strength upgrade.
Does layer height affect strength? It’s genuinely contested
Be honest about this one, because the literature disagrees. Some tests show thicker layers print slightly stronger — each layer’s molten core has more thermal mass and holds heat longer, so it bonds better. Others report thinner layers (0.15 mm on a 0.4 mm nozzle) improve adhesion. One controlled study on PPE found layer height had no effect on strength while line width increased it; another found layer height had a bigger influence on tensile strength than annealing. The takeaway: line width increasing strength is well supported; layer-height-versus-strength depends on material and test method, so treat it as something to test on your own parts rather than a rule.
As a working default, keep layer height between 25% and 75% of nozzle diameter (0.1–0.3 mm on a 0.4 mm nozzle), with 40–60% (roughly 0.16–0.24 mm) the sensible band for consistency.
Flow rate: under-extrusion delaminates directly
If each layer isn’t pressing firmly into the one below, it can’t bond — under-extrusion produces weak, brittle parts that delaminate under almost no load. Flow rate (extrusion multiplier) usually lives between 90 and 110%, and it’s a precision adjustment: changes of ±2–3% make visible differences to wall thickness and bonding.
Two rules before you touch it:
- Calibrate E-steps first. If your extruder isn’t pushing the length of filament it thinks it is, every downstream number is wrong.
- Use the wall-thickness method: New Flow Rate = (Desired Wall Thickness ÷ Average Measured Wall Thickness) × Current Flow Rate. Print a single-wall cube, measure with callipers, and correct.
For the full procedure with target values, see our walkthrough on flow rate calibration in OrcaSlicer.
Wet filament: the cause people forget
Moisture is an overlooked adhesion killer. When wet filament hits the melt zone at 200–280 °C, absorbed water flashes to steam inside the molten plastic. That disrupts extrusion — inconsistent line widths and gaps — and leaves micro-bubbles that weaken the part. One vendor guide puts the strength loss at 20–50%; treat that as indicative rather than a controlled measurement, but the mechanism is real and the symptoms (bubbling, popping, hairy stringing, patchy layers) are unmistakable.
Dry it properly, and never exceed the material’s Tg or it will soften and tangle on the spool:
| Material | Drying temp | Time | Caveat |
|---|---|---|---|
| PLA | 45–55 °C | 6–8 h | Softens above 55 °C; deforms at 64 °C |
| PETG | 65 °C | 6 h | — |
| ABS | 70 °C | 4 h | — |
| TPU | 55 °C | 8 h | — |
| Nylon | 80 °C (up to 95 °C) | 12–16 h | Needs a dedicated dryer; most dehydrators can’t reach it |
Note the PLA discrepancy between vendors: Prusa suggests 45 °C for 6 h in a filament dryer, Bambu 55 °C for 8 h in a blast oven — both fine, both under the softening threshold. Store dried spools in airtight containers with desiccant or they’ll reabsorb moisture within days.
Enclosure and ambient temperature for high-shrinkage materials
ABS and nylon shrink hard as they cool from extrusion temperature. On an open-frame printer the bottom layers stay warm on the heated bed while the upper layers drop to room temperature — the resulting thermal gradient pulls the print into internal stress, curls the corners and cracks the layers apart. An enclosure keeps the whole part at a uniform elevated temperature (around 30 °C ambient or higher) so it cools slowly and evenly, which is as much an adhesion fix as a warping fix. The same physics is covered in our piece on how to fix warping in 3D prints.
A working order of operations
- Dry the filament if there’s any doubt — no setting fixes wet plastic.
- Set the fan correctly for the material (off for the first few layers, low for ABS/PETG).
- Raise nozzle temperature 10 °C and reprint.
- Widen line width to 110–125% of nozzle diameter.
- Reduce speed 25% or add 5–10 °C to keep up.
- Verify flow rate (after E-steps) with the wall-thickness method.
- Enclose the printer for ABS/ASA/nylon.
If you’d rather not run the whole gauntlet, upload a photo of the failed part to our Diagnose tool — it identifies the defect from the image and returns slicer-specific settings, including downloadable .ini patches for PrusaSlicer and OrcaSlicer.
FAQ
Why does my print look perfect but snap along the layer lines?
Almost always over-cooling. The part-cooling fan solidifies each layer before it can thermally bond with the one below, so the surface finish is clean but the Z-axis bond is weak. Reduce fan speed (especially for ABS/PETG), disable it for the first few layers, and raise nozzle temperature by 10 °C.
Does higher nozzle temperature always improve layer adhesion?
Up to a point. More heat means more polymer diffusion and stronger bonds, so pushing toward the top of a filament’s range helps. But too hot causes stringing, poor definition, and — with some composites like PLA/carbon-fibre — internal voids that reduce strength. Stay inside the material’s usable window.
Is layer height or line width more important for strength?
Line width. Increasing it to 110–125% of nozzle diameter reliably boosts strength by adding contact area and squish. The effect of layer height on strength is genuinely contested in the research and depends on material and test method, so treat it as something to test rather than assume.
How do I know if my filament is too wet to print strong parts?
Listen and look: popping or hissing at the nozzle, inconsistent extrusion, bubbling, excessive fine stringing and patchy layers all point to moisture. Dry it at the material-specific temperature and time above, keeping under the glass transition temperature, then store it with desiccant.