New Marketplace Add-in: Inkbit's SSP 3D Nesting for Powder-Based AM Processes
Efficiently nesting part geometries into build jobs for powder-based additive manufacturing is one of the substantial challenges AM engineers are facing today. Poor nesting directly impacts build times, machine utilization and recycling of unused powder.
With the release of Inkbit's Scalable Spectral Packing (SSP) add-in to the Synera marketplace, it’s now easier and quicker to obtain highly performant nesting configurations within automated build preparation workflows.
Developed through collaboration between Inkbit and MIT, SSP solves the critical bottleneck of packaging large numbers of objects of distinct shapes. The add-in brings significant cost savings and efficiency improvements through optimized 3D nesting - achieving 25-30% higher packing density compared to alternative solutions while arranging parts up to 6 times faster. The Inkbit SSP addin in Synera can handle over thousands of parts in a single build and automatically prevents part interlocks - a common headache for anyone working with complex build arrangements.
As Javier Ramos, Chief Technology Officer at Inkbit explains: "When we developed SSP, we focused on creating the most efficient and highest performance 3D packing technology possible. Partnering with Synera allows us to deliver this capability exactly where engineers need it - directly in their automated build prep workflows."
That means fewer builds, reduced material waste, and lower per-part costs for your additive manufacturing operations.
The Challenge of 3D Nesting in Additive Manufacturing
Traditional nesting solutions typically face three major limitations:
Performance degradation with scale: A common approach in nesting algorithms is to use geometric queries between pairs of objects, resulting in quadratic complexity as part count increases – quadrupling runtime with every doubling of part count. This is why many tools struggle with more than a few dozen complex parts. The problem becomes exponentially more challenging when handling hundreds or thousands of parts - exactly the scenario where optimal nesting delivers the greatest economic benefits.
Suboptimal packing density: To overcome performance issues, many nesting solutions employ simplified geometry representations and heuristic-based approaches that prioritize speed over optimal placement. Others use patterning approaches that arrange parts in regular grids or arrays, which simplifies algorithm complexity but sacrifices density by failing to utilize irregular spaces effectively. These compromises save computation time but lead to substantial unused space. Many solutions also apply excessive spacing between parts to avoid potential issues, further reducing achievable density.
Interlocking issues: Tight packaging can mean that parts become entangled physically and are hard to separate after the build is done, which poses significant post-processing challenges and potential part damage. Solving this problem requires considering not just where parts fit, but also how they can be physically separated afterward - a computationally intensive problem that many solutions simply avoid.
How SSP solves common nesting issues
Inkbit’s SSP algorithm solves these challenges with a number of technical innovations:
FFT-Based Collision Detection
Rather than calculating collisions directly between parts, SSP transforms the problem to the spectral domain using Fast Fourier Transform. Here's how it works:
- Placement order: Objects are sorted according to size. The largest object is placed into the available build volume first.
- Voxelization: Both existing objects in the build volume and the new part to be placed are represented as 3D voxel grids (3D pixel arrays), where 1 indicates an occupied voxel and 0 represents empty space.
- Convolution via FFT: Instead of checking each potential position individually on a geometry level (which would be extremely slow), SSP transforms both voxel grids to the frequency domain using FFT.
- Fast Computation: In the frequency domain, checking all possible placements becomes a simple multiplication operation rather than millions of individual comparisons.
- Inverse Transform: The result is transformed back into the voxel space, producing a collision map showing every possible placement position and whether it causes collisions.
This approach processes millions of potential placement positions simultaneously in milliseconds, rather than checking them one by one. For a typical 240 x 100 x 100 mm build volume (24 million voxels at 1 mm voxel size), SSP computes a collision free part position in approximately 3ms - hundreds of times faster than traditional methods.
For a more detailed algorithm description, refer to Inkbit’s paper online.
Interlocking Prevention System
SSP employs a sophisticated multi-step disassembly algorithm that ensures every part can be physically removed after printing:
- The system sequentially identifies the existance of collision-free paths for each part from its placement to outside the build volume
- If parts cannot be extracted without collisions, they are removed from the packaging
- A flood-fill algorithm is used to re-insert the removed parts into positions that don’t pose create collisions
Proximity-Based Optimization
The possible part positions are related to the voxel grid spacing. Users might want to pick a coarse voxel grid to speed up computation, parts might not be placed in their densest configuration. To ensure the densest packing possible, an additional step is performed after the initial voxel-based placement and interlocking prevention, taking into account a user-defined part distance. With this additional step, packing densities of a 1mm voxel grid, can be achieved with a 2 mm grid while having a much lower computational cost.
Performance Benchmarks
In direct comparisons with other leading nesting solutions, Inkbit’s SSP continuously achieves higher packing densities, most of the time at much lower computation times.
Comparison of Inkbit’s SPP algorithm with current competitors. While drastically reducing compute times in most cases, Inkbit achieves the highest interlocking-free packaging densities. Source: Cui et al. (2023)
When to Use Inkbit SSP vs. Other Nesting Approaches
While SSP is a great algorithm, depending on the build job and AM method, it’s not always the best choice. Here’s when to use it – and when not to:
Use Inkbit SSP for:
- High-volume production with many distinct parts where maximum density improves revenue
- Complex geometries with potential interlocking issues
- Plastic powder-based technologies that don’t require support structures
Consider other, simpler nesting options available in Synera for:
- Simple, predominantly flat parts
- Small batches of similar parts
- Metal AM where support structures are required which would render a dense packaging hard to depowder and disassemble
Integration with Synera Workflows
The add-in integrates directly with Synera as a node in any workflow, with configurable parameters to tweak run time and minimum spacing between parts. It supports per-part orientation and position constraints, for example to keep test specimen right where they belong. In addition, slice area uniformity can be adjusted versus build job height, balancing quality and realibility with build time.
About Inkbit
Inkbit is an additive manufacturing company located in Medford, MA. The Inkbit Vista 3D printing system is designed for mass production of end-use parts. Inkbit have incorporated a novel technology called Vision-Controlled Jetting (VCJ) that delivers high resolution print capability enabling users to print parts with dimensional accuracy and precision at high volume.
Website Inkbit
