How Do You Clear Powder From a Printed Lattice Without Weak Spots?

Lattice powder removal is one of the more underestimated problems in powder-bed fusion manufacturing. Engineers who have designed lattices for FDM or resin printing are sometimes caught off guard the first time they run an SLS or MJF build: the lattice looks correct in the CAD model, prints without incident, and then proves impossible to fully clear. This article covers why the problem happens, how to design around it, and how to verify the result once the part is out of the machine.

Why trapped powder degrades a part

In selective laser sintering (SLS) and Multi Jet Fusion (MJF), the part builds layer by layer inside a bed of loose powder. The powder bed supports the part during printing, so no separate support structures are needed, which is one of the key advantages of these processes. The trade-off is that every internal cavity, channel, and enclosed pocket fills with powder during the build. After printing, that powder must come out. Unfused powder is not mechanically bonded to the part walls, but it does add mass and it occupies volume. If it stays inside a lattice cell, several things can go wrong: the part is heavier than designed, which skews structural calculations; loose particles can migrate to bearing surfaces or joints in service; and in cyclic loading, shifting powder can initiate fatigue cracks from the inside. More subtly, trapped powder prevents operators from visually or dimensionally verifying the internal geometry, so defects inside channels go undetected. Research on powder-bed printed lattices confirms that trapped powder can compromise the part's final mechanical properties, a risk that grows with cell complexity and density.

The design solution: escape paths come first

The most reliable way to handle lattice powder removal is to design the escape paths before the first print. The principle is simple: every enclosed space needs at least one opening large enough for powder to drain or be blown out. In practice, this means adding drain holes at the lowest points of internal channels relative to the build orientation you intend to use. A single 3 mm to 5 mm hole at the lowest point of a closed channel gives powder a gravity-assisted exit during de-powdering. Two holes, one at the top and one at the bottom of a channel, allow pressurised air to enter at the top and push powder out at the bottom, which is more effective than relying on gravity alone.

The positioning of drain holes is a function of build orientation, so it helps to fix the orientation early in the design process rather than adjusting it after the de-powdering approach has been planned. Mark the intended orientation clearly in the design file and review it with the build engineer before finalising hole placement. A drain hole that is perfectly positioned for one orientation becomes useless if the part is reoriented to minimise support material or improve surface finish on a different face.

How cell type affects powder removal

Not all lattice cells trap powder equally. Open strut-based cells, such as BCC (body-centred cubic) or octet trusses, have large, clear openings between struts and relatively few enclosed surfaces. Powder in these cells has many paths to exit and is generally straightforward to clear. Surface-based cells, particularly dense gyroids and other TPMS geometries, have continuous curved walls that divide the interior into interconnected but narrow channels. The channels are connected, not closed, but their tortuous paths and small opening diameters make powder removal significantly harder. Dense gyroids with high wall fractions, above about 40% relative density, present the greatest risk because the channel openings narrow to the point where powder bridges across them and resists flow.

If a TPMS cell is required for its mechanical properties, the practical response is to choose a lower relative density, which widens the channel openings, and to add drain holes at the minimum spacing that the structural analysis allows. If the application can tolerate a strut-based cell, the open geometry largely removes the powder removal problem. The choice of cell type is not made in isolation from the manufacturing process; powder removal is a legitimate input to cell selection.

De-powdering methods and their limits

Even with well-designed escape paths, some residual powder will remain after the part comes out of the machine. Standard de-powdering works in stages. Bulk powder is brushed or shaken off the exterior first, then the part goes into a de-powdering station where mechanical vibration loosens clumped powder inside channels. Compressed air blown through drain holes dislodges the loosened material. Bead blasting, directing a stream of fine media at the part, removes powder clinging to external surfaces and channel walls near openings. For particularly complex internal geometries, ultrasonic agitation in a liquid bath can reach cavities that compressed air cannot, though this is less common in production environments and requires drying the part afterwards.

The honest limit of post-process de-powdering is that it cannot reach truly closed pockets. Vibration and compressed air work by propagating energy through connected spaces; an isolated cavity with no opening receives no energy and loses no powder. This is why designing out closed pockets is not optional. If a cavity cannot be reached by the de-powdering tools, it must be opened by design or eliminated from the geometry.

Inspection and verification

After de-powdering, confirmation that the part is clean matters as much as the de-powdering itself. The two most practical methods are weight and CT scanning. A clean part should weigh close to its theoretical mass, calculated from the volume of the printed material. A part that is significantly heavier than the model predicts almost certainly retains trapped powder. Weighing before and after de-powdering gives a direct measure of how much powder was cleared and whether more work is needed.

X-ray computed tomography (CT scanning) gives a direct visual confirmation of internal cleanliness and also reveals any internal print defects, delaminations, or dimensional errors in the lattice geometry. CT scanning adds cost and time, but for structural or safety-critical parts it is the only method that confirms both cleanliness and internal quality simultaneously. For production runs, a CT scan on the first article of a new design, combined with weight checks on subsequent parts, is a practical middle ground.

A workflow checklist

Bringing these steps together: fix the build orientation early and commit to it; add drain holes at the geometric low points of every enclosed channel, sized at 3 mm or larger where structurally tolerable; choose a cell type and density that keeps channel openings wide enough to drain; after printing, run vibration, compressed air, and bead blasting in sequence; weigh the part and compare to the theoretical mass; scan the first article by CT to confirm internal cleanliness and geometry. Documenting this process in the design record closes the loop between intent and verification, which matters for any part where trapped powder would be a safety or performance issue.

Key takeaways

• Trapped powder weakens the part and adds uncontrolled mass; the fix is in the design, not just the post-processing.
• Add drain holes at the lowest points of every internal channel, sized generously and positioned for the intended build orientation.
• Open strut cells clear more easily than dense TPMS cells; if a TPMS is required, lower density to widen channel openings.
• Verify with weight checks and CT scanning on first articles; weight alone flags remaining powder quickly on production runs.

Related reading

• How to Design Lattice Structures for 3D Printing
• How Gyroid and TPMS Lattices Behave Under Load
• How Do Engineers Confirm a Lattice Will Hold?

Frequently asked questions

Does trapped powder affect the mechanical properties of a printed lattice?

Yes. Trapped powder adds uncontrolled mass, can shift under cyclic loading and initiate fatigue cracks, and prevents internal inspection. Research on powder-bed printed lattices confirms it can compromise the part's final mechanical properties, which is why design-stage escape paths matter more than post-process cleaning alone.

What size should drain holes be in a printed lattice?

A practical minimum is 3 mm in diameter. Smaller holes can bridge with powder and block completely. Where the geometry allows and the structural analysis is not compromised, 5 mm openings give more reliable drainage and are easier to inspect visually.

Can compressed air alone clear all powder from a dense gyroid lattice?

Rarely, without prior vibration. Dense gyroid channels are narrow and tortuous; powder bridges across them and resists air pressure alone. Mechanical vibration loosens bridged material first, making compressed air far more effective. In difficult cases, ultrasonic agitation in a liquid bath reaches cavities that air cannot.

Is powder removal a problem in resin or FDM lattice printing?

Resin (SLA/DLP) printing has an analogous problem with uncured resin trapped in closed channels; drain holes and part orientation help in the same way. FDM lattices rarely have fully enclosed internal channels because the open strut geometry leaves gaps by default, though support material removal can be an issue in complex internal features.

How do you verify a lattice part is free of trapped powder?

Weigh the part and compare to its theoretical mass from the CAD model. A significant mass excess signals remaining powder. For structural or safety-critical parts, CT scanning gives definitive confirmation of internal cleanliness and also reveals any dimensional errors or internal defects in the lattice walls.