China's TPM3D is combining SLS printing, post-build foaming and zoned lattice structures for footwear soles. At Formnext Asia Shenzhen 2026 the company presented a workflow in which a powder-bed part expands to as much as 2.5 times its printed volume. The stated target is not merely a prototype but a light, resilient and customisable sole produced at industrial scale.

Reported figures include 0.3 g/cm³ material density, apparent density as low as 0.1, 22–26% compressive deformation, energy return above 70%, and no collapse or fracture after 200,000 flex cycles. These numbers are promising, but they describe different properties and tests. They do not replace wear, grip and ageing tests on a complete shoe.

Three connected process stages

TPM3D divides the workflow into lattice design, material processing and post-processing. A digital sole can use a denser heel, medium or low density in the arch and a directional lattice in the forefoot. That allows local stiffness, cushioning and deformation to vary within one material without a separate insert.

The SLS powder bed supports the object, enabling complex internal lattices and tightly nested parts. A controlled post-process then activates foaming. TPM3D does not publish the complete formulation and every parameter, so the result cannot simply be transferred to another material or machine.

Interpreting the density numbers

The reported 0.3 g/cm³ material density and apparent density as low as 0.1 are not necessarily contradictory. The first may describe the foamed solid material; the second can include air inside the whole lattice volume. A fair comparison needs the measurement method, porosity, expansion ratio and lattice fill fraction.

A 2.5-fold volume increase is also a design challenge. The printed preform cannot be assumed to be a uniformly scaled 40% version of the final CAD because expansion may be directional and local wall thickness and heat distribution can change it. Final-size control requires geometry- and lot-specific compensation, measurement and statistical process control.

Energy return and compression

Energy return above 70% indicates that the coupon returns a substantial share of absorbed mechanical energy during unloading. It is not by itself a comfort or running-performance metric. Shoe feel depends on the time response of cushioning, hysteresis, sole geometry, upper, runner mass and movement pattern.

The meaning of 22–26% compressive deformation requires load, duration, temperature and recovery time. Permanent set, deformation under instantaneous load and a standard compression-set result are different quantities and should not be interchanged from a short corporate release.

What do 200,000 bends prove?

TPM3D says a third-party dynamic flex test found no collapse or fracture after 200,000 cycles. That is more useful than a demonstration video, but without the full report the flex angle, frequency, temperature, load, preparation and acceptance criteria remain unknown. It does not automatically equal 200,000 real steps, which add sweat, torsion, impact, dirt and changing temperature.

Commercial adoption should also examine abrasion, wet and dry grip, hydrolysis, ultraviolet ageing, washing, thermal cycling, odour, skin contact and attachment to the upper. Sports footwear needs biomechanical and user testing; protective footwear introduces separate standards.

Up to 190 soles in one build

TPM3D quotes up to 190 soles in a P550DL build. That is a nesting-, size-, orientation- and part-dependent maximum, not a guaranteed count for every design. Powder-bed production avoids dedicated supports and fills the Z volume, but total cycle time includes the full bed, cooling, unpacking, powder recovery and foaming post-process.

Meaningful productivity therefore requires good parts per hour, first-pass yield, labour, energy and powder refresh ratio. The figure of 190 must be paired with full-cycle duration and post-process throughput. If foaming or dimensional finishing is the bottleneck, printer nesting capacity will not equal factory output.

Customisation or mass production?

A digital lattice can change local stiffness and size without tooling. That may be valuable in orthopaedic or premium sports applications if foot scans and pressure measurements are converted into validated design rules. Personalisation also requires data governance, sizing models, part identity and quality assurance; a unique STL alone does not guarantee better fit or injury prevention.

At larger volume, SLS may avoid injection-mould tooling while foaming adds a new cell and control burden. Economics depend on SKU count, size range, demand volatility, powder reuse and scrap. TPM3D does not publish a complete part cost, so the method cannot yet be described as generally cheaper.

Verified fact, maker claim and editorial conclusion

It is verified that TPM3D presented the combined SLS and foaming workflow and its P550DL platform at Formnext Asia Shenzhen 2026. Density, energy return, compression, cycle-test and 190-part capacity figures are company-reported or company-cited results; we found no complete independent report.

LAYERTIMES concludes that the technical value lies in connecting three levels: zoned digital lattices, powder-bed batch production and mass-reducing post-foaming. Commercial success will be decided not by one energy-return percentage but by dimensionally stable expansion, whole-shoe durability, powder and waste management, and good parts per hour.

A commercial assessment would require whole-shoe durability, dimensional repeatability after foaming and good parts per hour to be measured under one independent protocol.