3D Printing: Warping & Bed Adhesion – Tips

Lisa Ernst · 20.11.2025 · Technology · 11 min

You start a long print in the evening, everything looks good – and in the morning, instead of a clean housing, you find a half-detached, crooked part on the print bed. The corners are lifted, the underside resembles a small waterslide, and in the worst case, the nozzle has already chewed through your workpiece. We at 33d.ch see exactly these scenes repeatedly in the workshop – with affordable entry-level devices as well as professional machines.

In almost all these cases, two causes are common: Warping (i.e., lifting edges) and too weak bed adhesion. Both ensure that projects run for hours, only to fail shortly before completion – and yes, that's incredibly annoying, especially when a customer is waiting for the prototype.

Therefore, we have bundled our experiences: how warping occurs physically, which settings typically work for PLA, PETG, and ASA, which print aids really help, and where myths circulate that bring more frustration than progress.

Quelle: Own Representation

Basics & Causes

What Actually Happens During Warping

In FDM 3D printing, a thermoplastic (e.g., PLA, PETG, or ASA) is melted and laid down layer by layer onto a print bed in thin strands. As it cools, the plastic shrinks slightly. If this happens at different rates in the layers, the upper areas pull on the lower ones – the corners bend upwards, creating the typical warping.

Technically, we speak of internal stresses due to temperature differences that are greater than the adhesion between the part and the print bed. (Source). Large, flat parts and materials with higher shrinkage like ABS, ASA, or Nylon are particularly susceptible, while PLA and many PETG types are considered 'low-warp'. (Source).

The First Layer as Foundation

At least as often, the print fails because the first layer never adheres properly. The filament is only loosely deposited, sticks to the nozzle, is dragged along, or shifts with the slightest bump. (Source). Bed leveling, Z-offset, bed temperature, contamination (finger grease!), and the choice of surface (glass, PEI, textured sheet) play a major role here. (Source).

In practice, we can often 'read' the subsequent print course from the first layer: if it's uneven, pressed too hard, or riddled with holes, the risk of warping increases significantly. (Source). At 33d.ch, it took us a while to get into the habit of really watching the first two to three minutes – today, we'd rather stop a print early than fish a banana-shaped part out of the printer eight hours later.

Quelle: makeuseof.com

The photo shows a typical warping scenario: the corners lift because the lower layers cool faster than the rest and contract. we often take such pictures in the workshop as a reference before changing settings.

Typical signs include small air gaps between the part and the bed or audible 'clicking' as the nozzle passes over lifted edges. In literature, warping is described as the visible bending upwards of edges or entire surfaces, usually starting at the corners of the first layers. (Source, Source, Source). The larger and flatter the part, the stronger these effects – especially at high print and bed temperatures.

A heated bed only solves the problem if the first layer actually lies down cleanly. If the bed is not leveled correctly or the Z-offset is too high, the first layer is too far away (poor contact). If it's too low, the material is squashed, creating an 'elephant's foot', and the filament can bulge sideways. (Source). Additionally, the displayed bed temperature is often 5–10 °C higher than the actual glass or sheet temperature. (Source). And if the bed is also covered with fingerprints or dust, the filament has little chance of really gripping. (Source).

Environment & Drafts

Drafts act like a warping turbo: if a corner of a large part cools down faster than the rest, it shrinks more and lifts visibly. (Source). Especially with ABS or ASA, even slight drafts from windows or ventilation shafts are enough to cause cracks and layer splits. (Source). For PLA, many manufacturers recommend a stable room temperature of around 20–25 °C; large drops downwards – such as a window briefly opened in winter – significantly increase the risk of warping. (Source). An enclosure or at least a draft-protected location reduces these temperature gradients and often makes the results noticeably more reproducible. (Source).

Quick Cause Checklist

If a print is lifting on your end, we usually go through these points in the workshop in this exact order:

• Clean the bed (isopropyl alcohol or dish soap) and visually inspect for scratches or residues.
• Check leveling and Z-offset (paper test or mesh leveling).
• Observe the first layer: does it close cleanly, without gaps and without ridges?
• Exclude drafts (window, air conditioning, fan from an adjacent printer).
• Check bed temperature within the recommended range and adjust in 5-°C increments if necessary.

Material-Specific Settings

Material choice and settings significantly determine how susceptible a print is to warping. PLA is considered relatively uncomplicated and often only needs 50–60 °C bed temperature to adhere well. (Source). PETG sticks much better, tends to string, and can still warp on large parts – especially if the bed is too cold or the cooling is too aggressive. (Source). ASA is mechanically robust and heat-resistant, but requires a significantly hotter bed (approx. 90–120 °C) and benefits greatly from an enclosure to control warping. (Source).

Some practical guides and manufacturer specifications suggest as rough starting points: PLA usually 50–60 °C, PETG 70–85 °C, and ABS 90–110 °C. (Source). For PETG, 70–90 °C is often recommended with a sweet spot around 80 °C, especially on glass or textured sheet. (Source, Source). For ASA, many recommendations are around 90–120 °C bed, sometimes slightly reduced for special formulations, but practically always with a well-preheated plate. (Source).

We consistently start with these ranges for new filaments and then adjust in 5-°C increments. We document each change with a photo of the first layer – this way, you can immediately see which combination results in the calmest edges.

For PLA and most PETG applications, a draft-free room is sufficient; many users print PETG successfully on open devices at 70–85 °C bed temperature. (Source). An enclosure becomes interesting when you print large PETG parts or ASA/ABS components where temperature differences quickly lead to stresses. (Source). For ASA, manufacturers and the community clearly recommend a closed chamber or at least an improvised windbreak to keep the ambient temperature stable and prevent layer cracks. (Source, Source).

How We Test a New Filament at 33d.ch

1. Choose a small calibration part (e.g., a 60×60 mm plate with rounded corners).
2. Set the bed temperature to the manufacturer's recommendation, disable the fan for the first 3–5 layers.
3. Enable brim (5–10 lines), reduce print speed for layer 1 to 20–30 mm/s.
4. Start printing, observe and document the first layer (photo, brief note on temperature and fan).
5. Change only one parameter per run – otherwise, you won't know what really helped afterward.

This may sound like some effort, but it saves an enormous amount of time in the medium term, as you can later rely on your own material profiles.

Print Aids & Slicer Settings

If the basics are right, but a part still acts up, print aids come into play. A brim is a wide skirt of one or more lines that lies directly on the outer edges of your part, increasing the contact area. This stabilizes narrow or tall parts in particular and reduces warping without parking the entire part on a thick base. (Source). In contrast, a raft is a multi-layer plate under the entire model – great for very warp-prone materials like ABS or ASA, but material-intensive and more difficult to remove cleanly. (Source).

In practice, a wide brim is sufficient in many cases for PLA and PETG, while a raft shows its strength for large, demanding ASA or ABS parts. We have often experienced a part dramatically lifting at 80% print height without a brim – the same job with an 8-line brim, however, remained completely flat. A before/after photo of an ASA part without and with a generous brim shows this difference extremely clearly.

Quelle: tronxy.com

Here you can see the difference between a part directly on the bed and a part with a raft or brim. Such comparison prints help enormously to develop a feel for when the extra material consumption is really worth it.

Glass with Pritt/UHU stick, PEI coatings, textured sheet, or special adhesion sprays are all field-tested, but work with varying success depending on the material. (Source). For ASA, manufacturers recommend, for example, glass, Kapton or blue tape in combination with hairspray or 3D printing adhesives to improve adhesion while maintaining a detachable connection. (Source). It is important to regularly clean the bed with isopropyl alcohol or dish soap to remove grease and dust. (Source). If a specific combination – such as PETG directly on smooth PEI – sticks too strongly, a thin layer of glue stick can also act as a release agent and make the parts easier to remove later. (Source).

Besides brim and raft, a slower first layer, slightly increased flow rate, and reduced or deactivated part cooling in the first layers are particularly helpful. (Source). Many practical guides recommend a speed of 20–30 mm/s and 105–110% extrusion for layer 1, allowing the filament time to settle. (Source). For materials with high shrinkage like ASA, a fan setting of 0–20% or completely without a fan is worthwhile – but with a hot bed and ideally an enclosure. (Source). For PETG, moderate fan speeds of about 30–50% after the first layers work well to reduce stringing without unnecessarily increasing warping. (Source).

Slicer Checklist for a Stable First Layer

• First Layer Height Choose slightly higher (e.g., 0.20–0.28 mm) to compensate for minor unevenness.
• First Layer Speed Limit to 20–30 mm/s.
• First Layer Flow Slightly increase (105–110%), but only if the layer doesn't already appear too squashed.
• Brim Enable by default for large, flat, or cornered parts.
• Part Cooling Leave off or heavily reduced for the first 3–5 layers.

At 33d.ch, we've gotten into the habit of going through these points as a quick pre-flight check before long prints – this significantly reduces failed attempts.

Myths & Misunderstandings

Myth 1: 'The higher the bed temperature, the less warping – so just turn it up to max.'

Classification: Mostly False. A bed that is too cold often leads to poor adhesion and thus warping, but a bed that is too hot causes other problems like elephant's foot, overly soft first layers, and extremely difficult-to-remove parts. (Source). Manufacturer recommendations for PLA, PETG, and ABS are typically 50–60 °C, 70–85 °C, and 90–110 °C – far from 'everything at 110 °C'. (Source, Source). Our approach: start within the recommended range, then adjust in small increments, and compare the results with photos of the first layer – anything else is more like guesswork.

Myth 2: 'PETG never warps, warping is only a problem with ABS/ASA.'

Classification: False. While PETG has lower shrinkage than classic ABS, it can warp significantly on large prints and at low bed temperatures – especially at the corners. (Source). Practical guides therefore consistently recommend a heated bed of 70–90 °C and often reduced cooling in the first layers to prevent warping. (Source, Source). In community forums, you find both extremes – from 'PETG never warps' to 'I can't get anything printed flat'. (Source). The difference almost always lies in the settings and the environment, not the label on the spool.

Myth 3: 'A raft fundamentally prevents warping better than a brim.'

Classification: It Depends. A raft offers maximum contact with the bed and masks unevenness, making it fundamentally very good against warping – especially with ABS/ASA or very large parts. (Source). At the same time, community experiences show that the model can still warp above the raft if the environment, temperatures, or cooling are not suitable. (Source). A brim is sufficient in many cases, uses less material, is easier to remove, and changes the underside less. (Source). At 33d.ch, we only resort to rafts when brims and clean settings are no longer sufficient.

Myth 4: 'More fan cooling always helps because the part hardens faster.'

Classification: Mostly False. Strong fan cooling is important for PLA to achieve clean details and overhangs, but it can exacerbate warping, especially in the first layers, because the plastic cools down too quickly and the corners detach more from the bed. (Source). Many recommendations suggest leaving the fan off completely for the first 3–5 layers and then gradually increasing it. (Source). For ABS and ASA, manufacturers often advise largely avoiding part cooling, as it can promote warping and layer splits. (Source).

Myth 5: 'Warping is simply a sign of a cheap printer, not of settings.'

Classification: False. Of course, high-quality printers have advantages in bed flatness, temperature stability, and auto-leveling, but warping arises primarily from physics, not from logos on the housing. Even expensive devices show significant warping with incorrect temperatures, too much draft, or unsuitable materials. (Source). At the same time, many problems on cheaper printers can be greatly reduced if leveling, bed temperature, adhesive, brim, and environment are properly adjusted. (Source). A good before/after photo of a cheap printer before and after calibration is often the best counter-argument to this myth.

For those who prefer to learn about the topic via video, many good step-by-step guides are available. Recommended video: How to Fix 3D Print Warping (in English) – here the most important causes and countermeasures are shown very clearly.

Practical Application in Everyday Life

For everyday use in the workshop or hobby room, a clear procedure helps more than the umpteenth list of 'secret tips'. For new setups, we generally proceed as follows: First, thoroughly clean the bed, then check the leveling (paper test or mesh leveling), and consciously observe the first layer slowly. (Source). Afterward, we start with a reasonable bed temperature in the recommended range and optimize in 5-°C increments, taking photos of the first layers each time. (Source, Source). As soon as larger or very flat parts are to be printed, we add a brim and – for ASA/ABS – an enclosure. (Source).

For evaluating information, we don't rely on a single source. We broadly divide them into three groups: Manufacturer data sheets and technical blogs (basic values for material and temperatures), independent tests like the comparison of PLA, PETG, and ASA by CNC Kitchen (mechanical properties and realistic printing conditions), and community threads for specific problem cases, where we primarily pay attention to recurring patterns and meaningful photos. (Source, Source). This way, we avoid a single experience report leading our setup into a dead end.

Another video worth watching is First Layer & Bed Adhesion Guide (in English). Such videos are a good complement to your own tests – but they don't replace them completely, as each printer and each filament reacts slightly differently.

Despite many articles and blog posts, there are still some gaps. For example, while tables of recommended bed temperatures exist, there are few freely accessible measurement series that systematically investigate actual surface temperatures at different points on the bed – deviations of 5–10 °C between the sensor and the actual glass surface can occur. (Source). On the topic of enclosures, we also rely more on empirical values than hard limits for optimal chamber temperatures depending on the material; often mentioned are 25–40 °C, but this is hardly backed by extensive studies so far. (Source).

Exciting, but still poorly documented are also 'low-warp' or modified materials like ABS+, ASA variants, or filled PETG types. Manufacturers emphasize reduced shrinkage here, but independent comparative data is rarely publicly available. (Source). Our recommendation: Use manufacturer specifications as a starting point, but always combine them with your own small tests – especially with new filaments and unusual combinations of bed surface and adhesive.

In summary, warping and poor bed adhesion are not a mysterious curse, but the result of temperature differences, material properties, and insufficient contact between the first layer and the print bed. If you combine realistic bed temperatures, a stable environment, clean leveling, smartly used brims or rafts, and suitable adhesives, you can get PLA, PETG, and ASA under control in most cases – and your own before/after photos will confirm this.

Mini-Conclusion: 5 Things You Can Try Immediately

• Clean Bed & Check Leveling – in our experience, this alone solves a large part of all adhesion problems.
• Slow Down the First Layer – 20–30 mm/s and a little more flow give the filament time to bond with the bed.
• Brim as Standard for Problem Parts – anything large, flat or cornered automatically gets a brim.
• EliminateDrafts – Move the printer away from windows and vents or use a simple enclosure.
• Document Changes – Photos, brief notes, and your own profiles save you a lot of searching on your next project.

Fits Well: Other Topics for More Stable Prints

• Understanding 3D Printing Tolerances
• Calibrating the First Layer Correctly
• Storing Filament Properly
• Using ASA for Outdoor Applications in 3D Printing
• Checklist for 3D Printed Prototypes in SMEs