Building a laser powder bed fusion machine has never been mainly a mechanics problem. Recoaters, build platforms and inert-gas enclosures are hard, but they are physical problems that a capable lab or machine shop can work through. The harder part is the control layer: the software and electronics that turn a sliced part into synchronized galvo motion, laser firing and machine logic. Danish startup POWDERBED wants to sell that layer as a product. As VoxelMatters reported on November 15, the company will show Core 1, an off-the-shelf scanner and laser controller for LPBF systems, at Formnext 2026 in Frankfurt. The net price is EUR 9,900.
POWDERBED is the brand of Aagaard AM ApS, founded by Sebastian Aagaard, PhD. According to the company's website, Aagaard previously built an open-architecture LPBF system at the Technical University of Denmark (DTU). That history explains the product. Core 1 is the controller a research group would otherwise have to design and integrate on its own before printing anything.
What Is Actually in the Box
Core 1 is a single-laser desktop controller. VoxelMatters reports that the hardware is built around a SCANLAB RTC6 Ethernet scan controller, a common choice for galvanometer scanner control in industrial laser systems. It supports both the SL2-100 and XY2-100 scanner protocols. That matters to builders because it decides which galvo heads can be connected without custom interface work.
On the software side, Core 1 runs Autodesk's Machine Control Framework (AMCF), which is open source under the permissive BSD-3 license. The unit also includes configurable I/O, support for a local equipment network and simulation tools. The company site says the simulation lets builders test machine logic before a laser is ever powered. Anyone who has watched a scanner do something unexpected during first commissioning will see why that is useful.
The product page has one unusual detail: Core 1 ships with project rules for an LLM coding assistant. In practice, POWDERBED expects customers to extend and change the machine software, and says project rules for your LLM are included, so an AI coding assistant can help extend the machine software.
Core 1 does not include the rest of the machine. VoxelMatters says customers supply their own mechanics, optics and laser. Core 1 handles scanning and control. The powder handling, beam delivery and laser source are still the builder's job.
Source Code, Not Just Software
The licensing model may matter more than the hardware list. Each unit ships with full source code, and the company site says commercial extensions built on it can stay private. That combination is what BSD-3 allows: you can read and change everything, and nothing forces you to publish your own additions. For a startup building a commercial LPBF machine on Core 1, that means its proprietary process logic or user interface does not have to become public.
Per VoxelMatters, each unit also comes with:
- Adjustable build styles, meaning the scan parameters and strategies that control how the laser moves through each layer
- Autodesk Fusion CAD/CAM integration
- Scanner and laser commissioning
- An acceptance test
- Remote onboarding
Build styles are where much of LPBF's know-how lives. In a closed commercial machine, these parameters are often locked or only partly exposed. Making them adjustable, with source code behind them, is aimed squarely at researchers who want to study process parameters rather than just use a vendor's defaults.
POWDERBED also offers optional annual support and OEM design-in packages, as well as process development services, according to VoxelMatters.
Why the Control Layer Is the Bottleneck
VoxelMatters notes that the large metal AM OEMs, including EOS and Nikon SLM Solutions, built this control layer in-house over many years. That is the gap POWDERBED is targeting. An established OEM can spread years of controls engineering across a large installed base. A university lab building one experimental machine, or a startup trying to bring a new LPBF design to market, cannot easily do the same.
VoxelMatters says universities building experimental printers and startups developing new LPBF machines usually have to write that layer from scratch, which can take months or years before the first meaningful print. That is engineering time that does not go into the research or the product. Core 1 offers an alternative: a commissioned, tested controller with full source code that can be modified.
For context, EUR 9,900 is not hobby money, but it is small next to the cost of a metal powder bed machine. It is also likely small next to the engineering hours needed to integrate a scan controller, write machine logic and validate the system from scratch. The intended customers, research labs and machine builders as the company site names them, are the ones who would otherwise pay those costs.
What It Means for Makers
To be clear, Core 1 does not make metal printing a weekend project. A builder still has to source a laser, scanner optics and a mechanically sound build chamber, and deal with the safety requirements of high-power lasers and metal powders. None of that is trivial, and Core 1 does not claim to solve it.
What it does change is who can realistically try. The open-hardware movement made FDM printers buildable once motion control, firmware and slicing became commodity, well-documented parts. Metal powder bed fusion has never had an equivalent for its control layer. A controller with full source code, a permissive license, simulation for offline testing and integration with a mainstream CAD/CAM tool is a step in that direction, even if the step is priced for institutions rather than individuals.
For advanced makers in university labs, makerspaces with serious budgets or small hardware companies, the useful question is no longer whether they can write an LPBF controller. It is whether they can build the machine around one. That is a more manageable problem.
The open-source part matters too. If Core 1 gets adopted, BSD-3 machine software running on standard scanner protocols could give experimental LPBF builders a shared base. Improvements could then be compared and reused instead of being rebuilt in each lab.
Where to See It
POWDERBED will show Core 1 at Formnext 2026 in Frankfurt, November 17 to 20, on the Sensortherm stand in Hall 12.0, Stand F85. The company lists [email protected] for inquiries.