Tolerances, fits, and accuracy in MJF: what can and can’t be designed “to zero”
MJF is valued not for “perfect dimensions,” but for predictability. When using the technology, what matters is not how clean the nominal value looks in CAD, but how the part behaves in an assembly: that it fits into place the same way, works the same way, and repeats consistently without any warping from batch to batch.
This approach reduces the amount of rework, speeds up product launch, and helps keep project timelines and costs under control.
Below, we’ll look at MJF tolerances and accuracy, including which are realistically achievable, where it’s possible to design “to zero” and expect assembly without post-processing, and where it’s better to build in a technological clearance or plan a simple additional step from the start.
This article also includes use cases for those seeking information specifically relating to PA12 accuracy and shrinkage for their own projects.
Why MJF is considered a predictable technology
In engineering practice, accuracy is important, not on its own, but with repeatability. If 50 parts assemble the same way, without fitting or errors, you get a controlled process rather than something of a lottery.
MJF is well suited to these types of tasks. Parts are formed in a powder bed without supports, so orientation has less influence on geometry, and surfaces don’t need to be restored after support removal.
The material is also sufficiently homogeneous, with parts behaving more consistently in snap-fits, thin elastic elements, and housings where repeatable geometry matters. So, while there are certain MJF dimensional accuracy limits, the customer gets a reproducible result across the batch, rather than just one successful part.
What “designing to zero” means in MJF
Designing “to zero” usually means assembly without additional machining. Zero clearance MJF design works if the interface does not require a defined interference fit, minimal play, or sealing achieved solely through precise geometry.
If, however, we’re talking about interference fits, bearing seats, or assemblies where sealing is critical, trying to assemble “to zero” often leads to unstable results. In such cases, it’s more reliable to choose a different strategy in advance: allow for a clearance, add a seal, use a bushing, or plan for local calibration. This reduces risk and makes the fit controllable.
Real tolerances and how to work with them
Additive manufacturing tolerances in MJF depend on the size of the feature, and the larger the dimensions, the wider the possible variation. This doesn’t mean that large parts are “inaccurate,” but that the design should ensure the assembly doesn’t rely on the tolerance limit, i.e add stops and datum features, move critical fits to smaller dimensions, make the geometry adjustable, or introduce clearance where it’s safe.
When considering tolerance stack-up, it’s convenient to use the following reference values:
Feature size | Tolerance guideline |
1–3 mm | ±0.14 mm |
3–10 mm | ±0.20 mm |
10–30 mm | ±0.27 mm |
30–50 mm | ±0.34 mm |
50–80 mm | ±0.42 mm |
80–120 mm | ±0.54 mm |
Key takeaway: tolerance increases with size, and this directly affects the choice of fit.
Clearances between parts in assemblies
When printing assemblies, a clearance between parts must be planned in advance. If this is not done, the polymer powder will fill the gaps and the surfaces may fuse together. As a result, instead of a separable assembly, you will get a single solid object.
For the HP Jet Fusion 5210, the practically applicable minimum for clearance gaps between individual parts is 0.4 mm, taking into account a tolerance of ±0.2 mm on each side. For moving joints, at least 0.5 mm is usually specified.
In thick-walled elements, mobility worsens due to the increased contact area, so the clearance is increased: for wall thicknesses over 50 mm, an additional +0.5 mm is added for every extra 10 mm.
This approach to MJF moving parts design helps maintain stable operation of the assembly after printing and cleaning, and the force required for movement varies less from part to part.
Where MJF usually assembles without post-processing
Without post-processing, housing elements, covers, and enclosures most often assemble successfully when they are located by stops and shoulders rather than by large flat surfaces.
Snap-fits, press fits, clips, and retainers work well, where the result is determined by geometry and the elasticity of the material. Guides, holders, and internal elements behave predictably if the design includes chamfers and clear, well-defined datum features.
In assemblies like these which often involve part mating, the customer gets exactly what MJF is chosen for: printed — assembled — works.
Where “zero” becomes risky
Bearing seats, interference fits, friction units, and sealed connections without gaskets are sensitive to small dimensional deviations in shape and size. Here, MJF is also applicable, but it requires a different approach, using clearances, design compensations, standard inserts, or local post-processing.
This is not a limitation of the technology, but a matter of taking correct decisions that follow established design guidelines for MJF.
Case studies
A private customer needed a set of 10 complex-shaped parts for airsoft replicas. Other printing methods were not suitable, either because the parts did not provide the required combination of strength and stable geometry, or manual finishing took too much time. Furthermore, the batch size did not justify producing tooling for injection molding.
At Makerly, the parts were printed from PA12 S using MJF precision engineering on the HP Jet Fusion 5210. The customer received a set that fitted into the assembly immediately and met the operational requirements: the geometry was consistent across the entire batch, and the need for labor-intensive fitting was eliminated, which was the key requirement in this task.
HatchTech: a pilot batch as a tolerance check before injection molding
HatchTech was preparing to launch serial production of incubator components and requested a pilot batch of 50 parts. At this stage, it was important for the company not just to quickly see the shape, but to verify fits and repeatability in order to confidently move to the mold and avoid introducing corrections into already expensive tooling.
The parts were printed on the HP Jet Fusion 5210 from PA 12 and TPU. Within five working days, the customer received a set of samples and was able to confirm that the geometry and fits behave consistently across the batch. In this task, MJF acted as a tool for tolerance validation, with the company testing the assembly on real parts and preparing more quickly for serial injection molding.
MJF is a technology where the goal is not dimensional perfection, but predictable part behavior in assembly. Where an assembly does not require interference fits or precision geometry, it’s reasonable to expect it to be done without post-processing.
In sensitive areas, it’s important to plan a clearance or design compensation in advance. This approach to 3D printing fits and tolerances using MJF speeds up product launch, reduces the number of iterations, and delivers stable quality from batch to batch.
