Haute Fabrication
Texas-based Haute Fabrication is launching a fully automated contract additive manufacturing service powered by self-learning metal laser sintering technology.
The company’s Hybrid Direct Laser Sintering (HDLS) process has been in development for more than seven years and is set to become operational for the first time this summer. Seeking investment via the StartEngine platform, the company is working to develop and deploy four variations of its HDLS system - with build volumes varying from 0.6 metres cubed to 5.2 metres cubed - which will be operated by a suite of virtual reality-trained robotic systems and a built-in adaptive artificial intelligence (AI) cloud-based control system. The finishing touches are currently being put on an initial 30,000-square-foot facility in Austin.
HDLS melts fine metal powder with multiple 1100W laser beams to build up parts, starting first with the raft support structures and then the part itself and other support structures. This standard laser sintering process is to be scaled up and supplemented with automation technology to enable mass production of end-use parts and specialised moulds for injection moulding.
Of the four printers Haute is to deploy as part of its contract manufacturing service, the HDLS 250 Minotaur is equipped with 1-4 lasers and boasts a 600 x 600 x 600 mm build volume; the HDLS 500 Tiamut boasts 4-9 lasers and has a 1250 x 1250 x 1250 mm build volume; the HDLS 1000 Kraken is equipped with 9-25 lasers and a 2500 x 2500 x 2500 mm build volume; and the HDLS 5000 Karathen will house between 64-144 lasers fusing together powder across a 5200 x 5200 x 5200 mm build area. A Stereolithography (SLA) machine with a build capacity of 2500 x 2500 x 2500 mm is also being developed, while Haute will also implement Selective Laser Sintering (SLS) systems too.
This varying build capacity is supported by cloud-based artificial intelligence infrastructure that hosts the printers, robotics and an automation control system. Haute’s automation infrastructure also provides an active feedback loop and learned response from previous builds across the entire fleet of HDLS machines, with a patent-pending system of visual, thermal and ultrasonic imaging analysing the layer by layer fabrication of every part to detect and record defects and anomalies.
The implementation of automation at Haute also includes the use of virtual reality (VR) to train robots and sync production processes for specific client projects. Using VR has allowed Haute to work effectively with its clients, particularly in the current climate where close contact is restricted due to the COVID-19 pandemic, to develop production workflows virtually, offering complete transparency and allowing customers to ask questions and give input. The set-up can be adapted depending on the client's need, with quality assurance processes, for example, being added for an aerospace customer demanding high volumes, and subtracted for the printing of a prototype part. This kind of modification can be done with minimal direct manual labour on the production floor, leading Haute to suggest it has found a way to 'pandemic-proof manufacturing.'
Back inside Haute’s HDLS platforms, meanwhile, are built-in autoclaves which can maintain temperatures up to 2,400F and facilitates the heat treating of components as they are being fabricated, meaning ‘near forged-quality parts’ are produced with just part removal and polishing required post-print. Haute is also confident that, eventually, any metal will be able to be processed on its platforms, though it is starting with more common powders like 316 stainless steel and copper nickel. While the company will also print polymer parts, Haute expects most of its business to be in metals.
Haute Fabrication co-founder & Chief Science Officer talks to TCT
The journey of HDLS technology starts in the 1980s in two states: Iowa, where Chief Science Officer and co-founder Kevin Friesth grew tired of the same old fabrication challenges he was encountering when building rocket engines and eventually got the ball rolling on Haute Fabrication. And Texas, the birthplace of SLS technology and Haute’s soon-to-be home.
It was 2013 when these two arcs met. Friesth, with his visions of a mass manufacturing technology based on 3D printing, met Vikram Devarajan, one of the co-founders of Structured Polymers, and soon found himself being invited to its Austin-based facility to meet with the team, including the late Carl Deckard. The Structured Polymers founders were happy to part with some advice.
“We can help you spend millions of dollars and build a printer just like what’s out there,” Friesth remembers Devarajan and Deckard saying, “but what you need to do is develop the next generation: an automated system. If this is going to be accepted as a mass manufacturing capacity, it has to be automated because one, you don’t want people around powder - [for some companies], 55 to 65 per cent of their cost is labour - and two, the quality of life damages, such as carpal tunnel because of repeated motions. You need to go automated and robotic to eliminate those two and then reduce the cost down to where it’s acceptable.”
Back when Deckard developed and commercialised SLS technology in the 1980s, to automate the technology so that human operators never came into contact with powder wasn’t possible. It was barely possible in 2013. For Friesth, who had this vision of taking the foundations of laser sintering technology and supplementing it with AI-powered control systems and robotic devices, and scaling it up to sizes not seen before, it was more than a little frustrating that this was to be a steady development process, with much time waiting on other pieces of technology to get up to speed.
“Until about four or five years ago, the controller technology hadn’t developed to the point we needed it, you didn’t have a lot of the sensors that we do today. And when you had the sensors, you didn’t have the micro controllers, so it’s been chicken and egg,” Friesth tells TCT. “It’s been frustrating trying to fight through that, but now we’re getting to the point that artificial intelligence, data buses, networking capabilities, the storage tech has finally caught up. It’s taken so many generations – three decades, basically – of allowing this technology to catch up to where we wanted to be.”
With Haute’s 2020 take on laser sintering technology, once the design files are sent to the printer, the robotic system set-up takes over, loading powder into an argon-filled melting system, which ensures no change of oxidation or moisture contamination, and later taking build cakes out of the machines for part removal and polishing. This system will also take care of the unused powder which will be recycled and pushed through the machines again. Job scheduling has also been automated to help prioritise orders, while the AI-driven learned response will help to continuously improve productivity and part quality. This infrastructure is identical for each set of machines, whether it's a fleet of 250 Minotaurs or 5000 Karathens.
It’s a similar story for the hardware itself, with lasers behaving in the same way, intercommunicating with each other and working in designated areas with minimal cross over, just scaled up to manufacture parts at bigger volumes with no compromise on speed. Meanwhile, Haute has integrated Smoothieware’s Smoothieboard fabrication controllers, which have been used on a plethora of CNC equipment, to support many of the automated processes around the machines, and also hired the brains behind it: Arthur Wolf, Chief Innovation Officer, and Mark Cooper, Chief Technology Officer, are responsible for designing much of the hardware and software at Haute.
This infrastructure has been put in place to ensure Haute has a running start once operational later this year. The roadmap has already been laid out; the company knows which industries and applications it will target from the get-go, what comes further down the line in a few years, and how the company grows in response.
“We’re going to go after the markets we know we can get into,” Freisth outlines.
It’s an important distinction. Although Haute has designed its HDLS set-up to deliver the quality and volumes required by the aviation industry, for example, and be robust enough to achieve certification, it doesn’t want to wait around. The company will start by serving lower tech markets, manufacturing components like heat exchangers from ground source pumps and coolers for graphics processing units (GPU).
“It’s a little lower key but, for instance, the water blocks for video cards, if we fill a chamber full, it’s two and a half million dollars of product,” Freisth says. “Whereas we could be like some of the other printer companies, they’re still working on getting certified with [commercial aviation] three and four years later. It’s one of the hardest industries to get into – some of them have spent millions of dollars and never made more than just a few parts. Why would we want to spend 200 million on a product that we might only be able to sell in three to five years?
“We’re talking profitability our first year of operation. And we’re looking at mass manufacturing, not one-offs. Our forte is targeting mass production. If you want 100,000 water coolers, we can get you them; if you need water blocks and you want a million units in a year, we can do that.”
Haute will also get to work right away on rocket engines, aiming to rid people of the anguish Friesth experienced all those years ago - an order to build several rocket engines for Odyssey Aerospace has already been secured. With its Karathen 5000 machine, applications like this will be produced in one primary piece, with just the turbo pump component and piping added after, as will ship propellers and nuclear heat exchangers. Although this machine might not be operational until next year - Haute hopes it will be ready before the end of 2020, but some 'very intensive flooring' needs to be put down to hold the weight of the platform - the first few fabrications are already being lined up, including a cold fire liquid nitrogen pressure test and a live fire rocket engine for testing. Once Haute has established its main business, meeting the volumes demanded for these types of applications and turning the profits Friesth and co are targeting in consecutive years, it will then turn its attention to the aviation industry.
It looks likely that Haute will set up a separate business unit to cater for what it anticipates being large demand in commercial aviation and move through the regulatory processes, while the main Haute business will continue to manufacture low-tech industrial components and rocket engines. On that end of the endeavour, the focus will be on turning old designs into 3D printable ones, consolidating the number of components required and shaving some costs. “If you come to one of our customer facilitation centres, we’ll redesign your part. It might have had 100 parts; it now has 12. If you happened to get your parts from 15 different places, we can get it all at our place.” This way, Friesth says, the company can solidify its base and allow the aviation business ‘the time to grow as it needs it’. In time, the central business will also expand its services to fields like automotive which don’t require so much regulatory work.
Haute Fabrication
Haute Fabrication
Haute's HDLS Kraken 1000 machine.
As the company’s reach grows, it expects to establish a host of additional facilities. The customer facilitation system centres would be multiple, helping companies to redesign components, while a host of manufacturing factories may be set up in regions with industrial clout to cut down on shipping and logistics. The factories will run on-demand, with jobs scheduled across the year to navigate boom and bust cycles, and other locations at the ready in case of overload.
The majority of Haute’s business is expected to be in metals, although the company will also house polymer systems built from much of the same architecture. “There’s not a lot of difference between SLS and DMLS, other than the chamber temperature,” Friesth says, also acknowledging that a lot of Haute’s technology is based off of SLS IP filed back in the 1980s. It has just been a case of adding updated componentry, like a new recoater blade to improve the spread of powder or a new galvo system to enable fine feature applications or thermal management system which is ‘totally unique.’
“What I did was take a system that can go from 100 degrees Fahrenheit, or room temperature, all the way to 2,000 degrees Fahrenheit,” says Friesth. “It’s a very custom design; took me about three years to figure out the thermal management problem.”
Somewhere within those three years, Friesth had gone back to Structured Polymers to show his progress and was turned around with the feedback that the thermal control was still inadequate and the artificial intelligence during build time analysis still needed some work. Advice like this, Friesth and his team have been more than happy to take on board, going to such lengths as moving the company to Texas be closer to these wisdoms.
It was encouragement from the Structured Polymers team that set Friesth on this endeavour, their standards that raised his, and guidance like the titbit from Deckard that follows that went a long way to deciding how to bring HDLS technology to market.
“One of the first things Carl said to me,” Friesth recalls, “he said, ‘you can make a lot of printers, but you won’t make money on them. If you’re going to make money on printers, you’re going to make it on the materials or you’re going to make it on the fabrications. I’ve got 30 years’ experience, I can tell you first-hand, if you try to go into the market selling printers, you’re going to be like all the rest and barely make a dime.’
“[Carl and Structured Polymers] have been very valuable mentors. They’ve ‘been there, done that’ with multiple machines, and we were able to [benefit] from their 100-plus years of knowledge. Unfortunately, we lost Carl early because I really wanted him to see the next generation machine, born from his original technology.”
