Why two parts printed in one build unit may behave differently
Customers will often understandably follow a simple logic of parts printed in one build unit (one MJF build chamber load), all behaving exactly the same way. For Multi Jet Fusion technology, this expectation is understandable: MJF truly ensures high consistency and stable batch quality. But in practice there is an important nuance, which is that the same print cycle does not mean the same mechanical behavior for different designs.
In other words, MJF prints precisely, but how a part will work under load is largely determined by geometry. Therefore, unpleasant surprises in a batch are more often related to how the model was prepared for production, rather than “print instability”.
What a build unit is and why it is not a guarantee of identical behavior
A build unit in MJF is one build cycle in the working chamber where a batch of parts is printed simultaneously. The process conditions for the batch are shared, and this is a major advantage of the technology. But inside one build unit there may be parts with different shapes, wall thicknesses, internal cavities, and different stiffness. One chamber may also contain parts from several different orders.
In addition, it’s important to understand that even inside one build unit the temperature distribution and the packing density of parts in the chamber are never identical – and this additionally affects the result.
Even if the material is the same, two parts will react differently to load, thermal expansion from heating, assembly, and post-processing. This is a normal engineering situation. For example, a thin flat lid and a compact housing made from the same HP PA 12 material will behave differently simply because they follow different structural logic.
The main reason for differences is the geometry of the part
The behavior of a part is most strongly influenced by wall thickness, section transitions, the length of thin areas, internal ribs, and the presence of cavities. Externally, products may look similar and have the same material properties, but they will perform differently. This is clearly seen in a simple example: a flat ruler and a box profile made from the same material have different stiffness.
Therefore, when preparing a model it’s important to ask not only “can this be printed?”, but also “how will the part behave after printing?”. Minimum allowable parameters are not suitable for all tasks. If the part will experience load, work in a snap-fit, or in an interference-fit assembly, the geometry should be designed with a safety margin to avoid 3D printing defects.
Why parts from the same build unit behave differently: typical reasons
Different wall thickness. Thin walls deform more easily, especially if they are long sections or sharp areas. Formally, the model may meet the minimum requirements, but in real use such areas become weak points resulting in potential print quality variation. For functional parts, to avoid dimensional deviation it’s better to identify zones where infill density and additional stiffness is required in advance.
Large flat elements. Wide flat surfaces without stiffening ribs tend to deviate in shape more often than compact volumetric parts. If one batch contains a reinforced housing and a lid without reinforcements, their behavior during assembly and under load will be different. This issue is usually solved with ribs, smooth transitions, and more uniform wall thickness.
Cavities and powder removal. Hollow parts help reduce weight and cost, but only if the cavities are designed correctly. 3D printing layer adhesion problems can sometimes occur if the model does not include powder removal holes or the channels are too narrow, due to material remaining inside. This affects the mass of the part, sometimes cooling rate, and the operation of moving areas. In addition, unnecessary internal partitions complicate cleaning.
Clearances in assemblies and moving joints. In MJF it is very important to define the working clearance in hinges, snap-fits, and mating features in advance. If the clearance is too small, the elements may work stiffly or not move as intended at all. If it is too large, play appears. In practice, differences in “behavior” often turn out to be a design problem in the assembly, as opposed to a printing problem.
Fine relief, thin protrusions, and sharp elements. Even if they print successfully, such elements may behave differently after cleaning and post-processing. For example, small text, a thin logo, or sharp edges may partially smooth out, and on one part this may be more noticeable than on another – simply due to differences in geometry and the placement of elements.
Why not only the model but also file preparation matters
Another source of differences is the quality of the STL file. The CAD model may be correct, but defects can appear during export to STL: open surfaces, intersections, extra elements, or overly coarse curve approximation. As a result, the geometry in printing may not match what was expected.
So, before sending a model to production, it’s important to check not only dimensions but also the file itself: shell integrity, correct export, and absence of mesh errors. For the customer this is often an invisible step, but it’s exactly what helps avoid rework and unstable results in a batch.
Post-processing also affects the final result
Even within one batch, parts may feel different in hand or assemble differently after post-processing. The reason is that different areas of a part react differently to cleaning and finishing. For example, any surface roughness on fine relief and sharp edges will smooth out faster than on massive surfaces.
If the part includes critical mating features, snap-fits, markings, or areas of contact with the user, it’s better to indicate this in advance. Then the post-processing requirements can be agreed before the batch is launched, rather than after the result is received.
How to prepare a model so there are no surprises in the batch
Before sending a model to print, it’s useful to go through a short engineering check.
First, the geometry should be evaluated: whether there are thin walls, long flat areas, weak ribs, or sharp thickness transitions. Then cavities and powder removal holes should be checked, especially if the part is hollow.
It’s also important to check clearances in moving and assembly joints. If the part is supposed to snap, rotate, or fit into a mating component, the working clearances should be defined deliberately rather than “at the minimum”. It’s also worth evaluating reliefs, text, and logos, as to whether they are large enough to remain readable after processing.
Finally, the STL file should be checked before sending it: integrity, absence of mesh defects, and adequate detail. This step takes a little time but often saves the entire batch.
When differences are normal and when the model needs revision
If different parts with different designs are printed in one build unit, their different behavior and tensile strength variation is normal. A thin lid, a rigid bracket, and a hollow housing are not expected to react to load in the same way. But if identical parts in one batch assemble differently, show systematic deformation, or fail to meet critical dimensions, this is a reason to review the model and the requirements.
MJF is a technology with high repeatability, but to deliver results and avoid any 3D printing part inconsistency, it’s important to agree in advance not only on the material but also on the engineering task: where additive manufacturing tolerances are required, where stiffness and layer height are important, where surface quality is critical, and where simplifications are acceptable.
How Makerly helps achieve the expected result
Preparation before printing plays a key role in the future behavior of the part. The Makerly team offers customers the opportunity to review recommendations for preparing models for printing, helps check them, identify weak points, refine clearances, evaluate cavities, and select the appropriate material for the task. If necessary, it’s possible to start with a test batch and only then launch the series.
This approach to repeatability in 3D printinghelps avoid unpleasant surprises in the build unit and ensure predictable results in real operation.
