Texas Instruments Aims to Evolve SLS with New DLP Chip

Stereolithography introduced to the world the possibility of turning digital CAD models into physical reality by directing a laser at a vat of photopolymer resin, point by point and layer by layer until that object was realized. The introduction of digital light processing (DLP) technology brought a new level of speed to the world of 3D printing through the ability to cure large swathes of material, rather than point by point.

Now, similar DLP technology could potentially be brought to selective laser sintering (SLS) and other additive manufacturing (AM) technologies thanks to a development from Texas Instruments (TI). Engineering.com spoke to Raecine Meza, business manager for Industrial DLP Products at TI, to learn about the company’s new near infrared (NIR) digital micromirror device (DMD) and its potential impact on 3D printing.

From DLP to SLS

Typically, with 3D printers using DLP technology, a light source, often an LED screen, is cast onto a DMD, made up of hundreds of thousands to millions of microscale mirrors that then project the light onto a vat of photopolymer resin. With each layer, a new pattern of light representing an individual 2D slice of a 3D object hardens the liquid into a solid shape. The process continues with each new layer until the object is fully formed and is then removed from the vat. 

Until now, the DMD for industrial systems was capable of withstanding the heat from the light source, associated with energy in the ultraviolet range. TI’s new DMD, the DLP650LNIR, can withstand even greater heat, associated with a NIR range of between 950 to 1150 nanometers. This makes it possible to project a NIR laser with up to 160W of power at the DMD, and bounce it through an optical set up and onto a print bed to melt plastic powder. 

An example of the new DMD in an SLS setup. (Image courtesy of Texas Instruments.)

In other words, some of the capabilities that DLP technology brought to the vat photopolymerization market can now be introduced to the plastic powder bed fusion market. Rather than fuse plastic powder point by point, as is the case with SLS, it’s possible to melt or sinter that powder layer by layer or section by section, depending on the actual build area and machine set up.

“What we’ve done is we’ve really married the device with a very nice thermal package so that we can be very efficient at dissipating the heat that might be generated as we apply those wavelengths to the array,” Meza said. “Then we’ve also increased the window transmission for the wavelengths of interest.”

While Meza did not have any specific numbers related to print speed, she did believe that using the new DMD would increase the speed at which plastic parts could be produced, when compared to traditional SLS. Moreover, Meza pointed out that the DMD enables much greater thermal stability than was previously available, opening up finer, more complex print details printed at constant speeds. 

A typical NIR setup using the new DMD chipset. (Image courtesy of Texas Instruments.)

Most SLS printers rely on mechanical systems for fusing powder, such as galvanometer mirrors. Even high-speed sintering (HSS) and Multi Jet Fusion (MJF) use mechanical inkjet heads. In either case, the mirrors or heads may require recurring maintenance or calibration. Because DMDs are digital, this sort of calibration or maintenance can be reduced or eliminated with this method. The system is also programmable, introducing more flexibility to the printing process. 

“Controlling thermal drift or variation across a system is a very common engineering challenge,” Meza explained. “The ability to have fast, real-time programmability for where heat is delivered has several implications to it.”

 
The DMD chipset and controller are available as single evaluator modules or can be purchased in bulk. (Image courtesy of Texas Instruments.)

The DMD chipset can be purchased on its own, bundled with electronics such as a TI evaluation module, or as part of an optical light engine for a given application.

Beyond SLS 3D Printing

Our interest lies mainly with 3D printing, but Meza described two other applications for the new chipset that could have significance for AM. One application is meant for industrial printing or marking in which packaging can be customized in real time with complex information. Intricate logos or marketing offers could be produced on a per-package basis. More elaborate serial numbers could also be added to packages for real-time supply chain tracking.

Two applications beyond 3D printing in which the chipset could be used. (Image courtesy of Texas Instruments.)

“[The chipset enables] very fast switching times. In a small window of time, we can deliver varying amount of light, which, in this case, translates to thermal energy, to the photosensitive or thermal sensitive substrate so that you can get things like a grayscale print,” Meza said.“This device can do pattern rates of 12.5 kHz. You can imagine that within a very fast time you can do multiple pattern in the same area and still keep things moving at a pretty good rate.”

This expands to full-color, soft package printing, in which colored inks are applied to packages using a thermal process. Similar to the grayscale printing described above, full-color packaging could become customized as well.

The exact applications of these examples to the world of 3D printing are best left up to the AM systems engineers, but one could imagine an inline printing process, in which multiple print beds move in front of the optical system to have a layer fused before a new print bed arrives. This serial set up would be similar to what TNO has done with its Print Valley system (see below). 

Beyond Plastic 3D Printing

Meza described other ways that the chipset might be used to push the capabilities of the chipset.

“Even though we’re doing a 10x step-up in our power handling, it’s still not thousands of watts. So, what you’ll see is that there are actually a couple of things in our favor. One is that continued advances in the material science part,” Meza said. “People are expanding the ways they can optimize lower optical power to these types of materials through modified chemistries. There’s also a lot of innovation that can happen on the system level for the optics. There are definitely folks out there researching different ways to drive up the overall power densities that get applied to the actual print surface using various magnification schemes or even scrolling optical heads and finding ways to get that optical throughput amped up by system-level solutions.”

In the first case, we can imagine additives in thermoplastic that allow the material to sinter at lower temperatures. This might be as simple as using darker powders that enable greater absorption of the NIR energy. This technique has been used in HSS, MJF and low-cost SLS systems that reduce the price of the technology by relying on cheaper, less powerful lasers. 

Meza explained that the device has actually been qualified to handle 500W of power in a smaller block of array. So, in the case of unique optical setups, users could focus the laser on a single part of the DMD, as opposed to spreading it out across the full array. This would make it possible to direct even more power to the print bed. This could be enhanced through the use of magnifying lenses and other tools to deliver even more power to an even smaller part of the print surface and then moving the head around.

Though it wouldn’t be possible with the DMD used in a standard setup, implementing these additional magnification techniques could enable the printing of metal powders. Meza was not able to discuss any customers that are currently using the DMD for plastic or metal 3D printing, but did indicate that there are customers using the DMD for AM.

To learn more about the DLP650LNIR, visit the product page.