3D Printing is a household term at this point. If you don’t own a 3D printer yourself, you probably know someone who does. Today, the differences in mechanics between a consumer fused deposition modeling (FDM) printer and a ceramic digital light processing (DLP) printer are quite large, but the theory underpinning their operation is shared. These AM technologies can trace their lineage to a radical concept that was realized by Chuck Hull in the early 80s: that you could build a functional part from nothing but a digital file, a curing light source, and a vat full of resin.
A Brief History of 3D Printing
In 1983, Chuck Hull invented the first vat photopolymerization system and an accompanying native file format (STL) that is still used today. Stereolithography (SLA) uses a UV laser to cure liquid photopolymer resin layerwise into a 3D object. This technology proved to be useful for prototyping and product development, so with a clear business case, Hull co-founded 3D Systems and filed the first patent for SLA in 1986. Thus, 3D printing was born as a commercial technology.
Other 3D printing processes were developed in the following decades. DLP, another vat photopolymerization technology, was introduced to the market in 1987 by Texas Instruments. DLP uses a UV projector to cure a photopolymer resin into a 3D object, although compatible slurries – suspensions of ceramic particles in a photopolymer resin – were not developed for DLP until the latter half of the 2000s. In 1989, Scott and Lisa Crump filed their patent for FDM technology and co-founded Stratasys. FDM printers extrude molten thermoplastic filaments into a 3D object, after which it cools and solidifies into a functional part. Similar processes were patented and used by a small number of companies, mainly in the aerospace, automobile, and medical industries, to manufacture products that were being sold at upward of hundreds of thousands of dollars.
For the better part of two decades, 3D printing was solely an industrial tool. Unconstrained by the limitations of subtractive manufacturing, it was used by large companies to enable rapid prototyping. In a surprising turn of events in 2005, a University of Bath engineer named Adrian Bowyer launched an open-source initiative called the RepRap Project to design a self-replicating 3D printer. This opened the doors for the creation of several new 3D printers. The floodgates for consumer 3D printers burst open when Stratasys’ FDM patent expired in 2009, allowing a deluge of new printers to enter the market and dropping the average cost for a consumer FDM printer from 10,000 to less than $1,000. By the 2010s, consumer FDM printers – under brands like MakerBot, Prusa, and Creality – were showing up on desktops in libraries, classrooms, and bedrooms. Meanwhile, industrial AM grew larger build volumes and encompassed more complicated materials, like technical ceramics, which were largely excluded from the additive advances of the 80s and 90s.
Like Your Printer, but Larger and for Ceramics
Providing a ceramic part design meant for a subtractive manufacturing process to an AM service bureau like Adva Cera is a bit like writing a screenplay and expecting it to work as a novel. If you have ever owned a desktop 3D printer, you already understand how a CAD model should look going into a print job. You have made decisions about layer height, orientation, and other process parameters. You have watched a part warp off the bed because the first layer did not adhere properly, and you troubleshot the run appropriately. Those informal experiences can act as guidance for designs intended for ceramic DLP printing and can make working with an AM service bureau more efficient. Less time needs to be spent explaining to a customer why supports are required on overhanging surfaces, or why a feature needs to be reoriented, or why tolerances need to be reconsidered. The design arrives closer to manufacturability, and the lead time will be reduced.
Despite differences in feedstock material, the same layer-by-layer logic applies. As in a consumer FDM printer, the orientation of your part in a ceramic DLP printer will affect its mechanical properties, surface finish, and any support requirements. Any support structures that are printed will need to be removed after printing. In FDM, support removal simply follows the process of snapping, peeling, or cutting the supports away and cleaning up the affected surface. Often, FDM printers will allow a layer gap between support and part. Advancements in support parameters turn support removal into a simple peel-away process, increasing accessibility to hobbyists. In ceramic AM, supports are cured through interlayer polymerization, creating a solid link between support and part. While thin connecting “teeth” are typically used to aid support removal, as-printed parts are fragile to the touch and supports must be delicately removed to prevent surface damage. It is advantageous to make design decisions that minimize supports or that split a complex part into simpler components to be assembled.
In terms of printing, thermoplastics are way more forgiving than slurries. Parts produced via FDM are capable of plastic deformation and will survive minor mishaps during handling. Ceramic parts are brittle and soft prior to thermal processing, meaning that an (often unintended) applied force can easily damage them. The green state of a ceramic part is especially fragile and can be damaged during cleaning and handling if the handler is not careful.
The most glaring distinction between consumer FDM printing and ceramic DLP printing is that ceramic parts must undergo two sequential thermal processing steps called debinding and sintering. Debinding removes all organic material from the green part and sintering allows the suspended ceramic particles to densify into a solid, functional part. The loss of organic material and ceramic densification leads to part shrinkage, which is a new consideration when coming from a strictly FDM background. Shrinkage rates are material-dependent, and the magnitude of shrinkage tends to be greater in the z-direction, but the good thing is that this behavior is predictable and can be compensated for during the printing process.
Another important distinction between consumer FDM and ceramic DLP printers is their achievable resolutions. Your average consumer FDM printer can achieve lateral resolutions of roughly 200-400 microns and no less, limited by the physical diameter of its extrusion nozzle. Ceramic DLP printers, on account of their intricate optics system, can achieve lateral resolutions down to 31 microns (depending on the printer and material) with layer thicknesses as low as 10 microns (again, depending on the printer and material), enabling the manufacture of complex structures that are characteristic of ceramic DLP prints. For engineers accustomed to subtractive ceramic processes, or even to FDM printing, that resolution capability often comes as a genuine surprise.
Additive Design Fundamentals – Wall Thickness
Ceramic parts have minimum wall thickness requirements to withstand stresses induced by escaping gas during thermal processing. Wall thickness requirements are also material-dependent and can be found in technical data sheets provided Adva Cera. If any of these requirements are not met, there will be a high likelihood of cracking or total failure of the part.
A common parameter associated with FDM printing is infill, or the density of solid sections of a part. In ceramic DLP, hollow sections must be manually created using CAD prior to printing, and engineers must include holes to allow cleaning of these internal spaces. This practice can be utilized for lightweighting or to accommodate strict design requirements, but it typically comes with an increase in cost. As a general principle, walls should be kept below 4-6mm, depending on the material. Walls within this range or above should prompt a conversation before a file is submitted. Uniform wall thickness is also strongly preferred over variable thickness, as anisotropic shrinkage during sintering can create stress concentrations along uneven walls and cause warping or cracking. If your geometry requires uneven wall thicknesses, a gradual taper or rounding between corners can mitigate these stresses more effectively than a sharp step.
Additive Design Fundamentals – Build Orientation
Surface finish, dimensional accuracy, and residual stress distribution are all part properties that vary depending on orientation. A feature that prints cleanly when vertical may require support structures or produce artifacts when tilted, so critical surfaces may need to be oriented to minimize supports entirely. Expected shrinkage during sintering further complicates build orientation decisions, as improper orientation can lead to warping or out-of-tolerance dimensions in the fired part that didn’t exist in the green part. Adva Cera’s engineering team will optimally position parts to accommodate critical surfaces, so it is advantageous to bring this information to a design consultation.
Additive Design Fundamentals – Internal Geometries
The ability to achieve parts with complex internal geometries, like through-holes and channels, is ceramic AM’s bread and butter and the main reason why customers forego traditional ceramic manufacturing means. These types of features are useful in applications that require fluid flow, heat exchange, or lightweighting.
Minimum feature size applies to both external and internal geometries. Small holes and narrow channels may be difficult to clean and have a propensity to constrict due to excess curing during printing. Likewise, enclosed volumes require an outlet for cleaning solvents and gases released during sintering. If the outlets are not sufficient in a given design, off-gassing during sintering will lead to pressure build-up within the part, and it could potentially crack or explode. When internal geometries are functionally critical, confirm your feature tolerances prior to finalizing your design.
Additive Design Fundamentals – Shrinkage Compensation
As stated previously, sintering a green part will cause it to shrink anisotropically. By analyzing process data, Adva Cera can predict how a part will shrink depending on variables like the sintering temperature, the direction of shrinkage, and the type of feature being sintered, and they can apply a compensation factor during printing to offset it and achieve intended dimensions. While this is a parameter that should not need to be considered in a customer’s design stage, it is an important detail in understanding the full process.
That being said, shrinkage compensation isn’t an exact science. Complex geometries, varying cross-sections, and features that sinter at different rates can produce local dimensional variations that may need to be addressed through trial-and-error. This is why critical dimensions, or features where your part must meet a tight tolerance to function as intended, must be explicitly identified in the design file submission.
FDM Prototyping to Ceramic DLP Production
The iterative mindset that characterizes good design practice for FDM is just as valuable in ceramic DLP printing, but the economics of technical ceramic production make the front end of that iteration loop more important.
A failed or out-of-tolerance ceramic part represents a significant investment of time, material, and kiln energy. A failed FDM print is negligible by comparison. Assuming an FDM printer is available, this asymmetry argues strongly for using cost-effective technologies like these for prototyping as an initial step in the design process before submitting your design files to Adva Cera. This step will not validate the material properties of your design, but it will allow you to address geometric problems (e.g., print orientation, supports, and optimal internal structures for part cleaning) and reduce costs by reducing the number of ceramic iterations that will need to be made. When you are ready to move to ceramic DLP, Adva Cera can prototype and validate your part as composed of ceramic before contracting a full production order.
