1 of 3
Al Siblani with some of EnvisionTECs patents in the area of 3D printing
2 of 3
The EnvisionTEC Xede 3SP 3D printer
3 of 3
EnvisionTEC's Eglass material for the 3SP process
EnvisionTEC is a well established company that is perhaps best known as a leader in what is set to become the next 3D printing battleground — high-precision. From the FormLabs Form1 to the Old World Labs OWL Nano, new manufacturers are popping up that eschew the now saturated FDM-style desktop machine and instead go after the market that EnvisionTEC has led — desktop-friendly high-resolution polymer 3D printers.In this arena stereolithography (SL) and digital light projection (DLP) are the kings – and EnvisionTEC’s lineup includes a bit of both.
I was able to get the lowdown on the company, its philosophy and its products directly from Al Siblani, the founder, chairman and CEO of EnvisionTEC, while at the same time seeing first-hand the company’s production facilities. Al has been in the industry longer than most, starting with the now-defunct Helisys in the early 1990s, selling and installing the first ‘laminated object modelling’ (LOM) machines to the big Michigan-based automakers General Motors, Chrysler and Ford. A number of the Helisys team made the move with Al when he set up EnvisionTEC, meaning the core team has been working together for over two decades.
The first thing I noted upon arriving with Al at the Gardena facility is that’s its no longer big enough. Crates had been wheeled outside in the morning to make room for the production lines inside — a benefit of production happening in California rather than Michigan or Gladbeck in Germany! Al explained: “We have roughly three times the orders this January compared to January 2013, so we need more capacity. We have another unit opposite this [current unit] that comes online at the start of February. We are currently in the process of doubling our production space in our facilities in California and in Gladbeck, Germany.”
Since the early days of the Perfactory the EnvisionTEC line up has changed significantly as new sectors of industry chase the speed and precision offered by the company’s DLP and now 3SP-based systems.
Value vs Cost
Of all the industry leaders I have spoken to, Al Siblani is perhaps the clearest about sticking to the company’s philosophy, even in the face of increasing ‘competition’. I put competition in inverted commas because whether or not other companies in the space are considered as such depends on your point of view. Certainly from within EnvisionTEC the view is that they are differentiated from the competition — they operate alongside other companies operating in a similar space, but companies that have a different mission.
Al explained: “I am not interested in getting involved in a ‘race to the bottom’ by competing solely on price. We have made some significant changes to our technologies that have allowed us to reduce the price of the machines without compromising our core values — if we tried to compete with the low-end machines we would have to make low-end machines, and in the long-term that won’t serve us or our customers.”
Looking around the production facility it is easy to see the emphasis placed on quality in every aspect of the development, manufacture and QC. Following the line around from a bare case to a complete product, the high-quality components become evident — including CNC machined parts made in-house just metres from where they will be tested and deployed.
“We deliver machines that create highly accurate parts, down to 15 microns, and to achieve that for build after build, year after year, the machine itself has to be built incredibly accurately,” explained Al. “Where the less expensive machines use plastic components, EnvisionTEC machines use our own metal parts that are designed to perform for many years facilitating the highest of accuracies. Likewise, our DLP technology is more expensive that that used in the cheaper machines — because to get the real, repeatable performance our customers rely on for their businesses requires the more expensive DLP system. Could we make a desktop printer that was $1000? Of course, we believe we know more about DLP than any other company, but cheap and cheerful is not what our customers want, and it’s not what we deliver.”
Taking a look at the projection system in their DLP systems alone shows how the company is not willing to compromise on quality. DLP projectors can work with square- or diamond-shaped pixels. Square pixels produce the crisp, fine edges required for very high-resolution 3D printing, where the diamond pixels are subject to re-sampling of the image, meaning that where just 1 square pixel may be ‘on’, when resampled at least 4 of the diamond-shaped pixels will be ‘on’ to produce the same image. On large ‘broadcast’ projection this creates a softer, more pleasing image. On a 3D printer, it creates a softer, less desirable definition.
The difference is best exemplified by imaging a straight line made up of pixels. The top edge of a straight line made with squares will be flat, but the top edge of a straight line of diamonds will be ‘ridged’, with peaks and troughs created by the points of the diamond.
Key to ensuring that all of the effort and expenditure of the production process actually makes a difference for the customer, a huge amount of QA and QC testing is performed on every single printer before it leaves the building. Some 280 individual pieces (albeit small pieces) are built on each machine before it ships, with a three-day programme of testing and re-testing undertaken. Rather than running purely test parts, each printer manufactures industry-specific parts — for example dental or jewellery — that are then accurately measured, ensuring that the shipped printer is ready for real world operation.
Even with the higher-cost professional DLP chips and projectors, DLP-based printing systems all face limitations in terms of the size of parts that they can be used to produce. To create parts larger than about 200 mm x 250mm, the projector must be moved further away from the build plate to cast a larger image. However, the number of pixels (whether square or diamond-shaped) remains the same regardless of the image size — thus each pixel becomes larger and resolution is lost.
“We have extensive experience designing DLP printers,” explained Al, who is still deeply involved in the R&D team. “We have tried machines with multiple projectors, moving projectors, a huge number of optics and so on, and there is really no effective way to make DLP suitable for printing of large parts quickly and repeatably enough for customers. We have patents on image stitching for example, needed where more than one projector is used to cover a larger printing are. In theory it is possible to use sophisticated algorithms to ensure that there is no evidence of the place that the two images meet on the part, but in the real world it is very, very difficult. When your machines are being used in production settings, such as within dental workflows, it’s vital that sources of error and complexity are eliminated.”
In order to move into new areas where larger-scale parts are needed — such as the MCAD space — the company developed a new type of process that retains the high-resolution and speed of the smaller-scale systems, but adds a theoretically infinitely scalable build size. The solution is 3SP (standing for Scan, Spin and Selectively Photocure) technology, a laser-based resin curing system rather than a DLP system. The system is actually remarkably simple and surprisingly quick thanks to a combined curing and recoating step performed on both the forwards and backwards sweep across the X axis.
The lasers in the 3SP systems are contained in compact cartridges, reminiscent of a VHS videotape, that themselves are moved in the X direction across the surface of the print bed. The Y axis is covered by the scanning of the laser, which is fired at a mirror revolving at over 20,000 rpm. The laser can be turned on and off in just seven nanoseconds, which relates to accuracy in the Y axis of 4-5 microns, while being up to five times faster than a comparable SL machine.
One of the issues with a laser system controlled by mirrors and housed way above the build platform is that for larger parts, the spot size and shape of the laser changes from the centre of the build platform to the corners, something that has to be computer corrected but can still affect the fidelity of the final models. By correcting the beam shape and size with bespoke fixed optics, the 3SP systems can produce sharp details across the bed area without the need for correction algorithms.
“We started development work on the 3SP technology a few years ago with the aim of developing a technology that fit the EnvisionTEC ethos while allowing us to expand into new sectors. It is completely scaleable and could potentially make huge parts without sacrificing the level of detail or speed by combining multiple ‘print heads’ together, as one would with a conventional inkjet printer.”
If evidence was needed to justify the development of the 3SP technology, Al explained that the 3Dent — a dental specific 3SP-based system — would have to be some $55k more expensive if it were to retain the same specifications but run DLP rather than 3SP. In the last three months a large dental lab group purchased 10 3Dent machines based on the 3SP technology.
Again the quality control procedures were underway for the 3SP range and include measurements of the wavelength, spot size, consistency and more.
The final string in the EnvisionTEC bow is something quite different again — the 3D-Bioplotter. The 3D-Bioplotter is a syringe-based 3D printer that can process more individual materials than any other biofabrication printer — from gelatine to titanium. Using syringes actuated by either air or physical pressure, viscous pastes, gels and liquids are deposited (plotting material) into a matrix (plotting medium) or directly onto a platform/object placed on the printer. The unique properties of the process allow Computer Aided Tissue Engineering and controlled drug release systems that require very well defined internal and external structures coupled with the ability to utilise up to 5 materials at a time on the 3D-Bioplotter.