Teton Simulation
Teton Simulation's Smart Slice software tool.
A 3D model of a shelf bracket sits on a virtual print bed of an Ultimaker S5 machine, as a user selects the correct print profile, chooses the most suitable material and then begins tinkering with specific parameters to fine-tune the characteristics of the part and meet the end use requirements.
The user defines a fixed surface for the part, then a distributed load of 200 Newtons on its upper surface, then a safety factor of 2.5 times the load to guard against deformation, and then, considering stiffness, a maximum deflection of one millimetre is defined too. Clicking the ‘validate’ button, the software constructs a high-fidelity finite element model based on the STL geometry and data from the selected material. Within 90 seconds, the software has informed the user that the modelled part will meet the stiffness requirement but not strength.
They are now faced with two options: increase characteristics like infill density or wall thickness and rerun the validation or click ‘optimise’ and allow the software’s optimisation algorithm to 'explore the full design space' of the print settings to return 10 different configurations in 30 minutes that satisfy stiffness and strength, listing them in order of print time.
We are inside Teton Simulation’s Smart Slice software, a tool that has been integrated into Ultimaker’s Cura platform since the autumn of 2020 and is being demoed by the company’s Director of Operations and Alliances Rick Dalgarno and VP of Product Doug Kenik, both formerly of Firehold Composites, a CAE engineering firm acquired by Autodesk in 2013. As they play around with Smart Slice’s tools, they offer a glimpse of what the software has to offer, explain the motivation for its launch and outline how they intend to bring the product to market.
Teton was founded nearly four years ago and has been funded by the US Government and the National Science Foundation. It set out looking at how to offer better structural simulation for injection moulded parts but was quickly put on the path towards enhancing additively manufactured parts by its investors, who are said to have a lot of interest in the technology. Once the company’s mindset had been switched towards additive, Kenik, Dalgarno and some of their Teton colleagues harnessed their experiences at Autodesk working with generative design technology to spot a perceived gap in the industry’s software offering.
“Generative design technologies typically don’t take into consideration the actual process itself. When they’re giving you designs, they have a very high level of manufacturability in there as far as support angles, minimum overhang angles, minimum thicknesses and things like that, but there’s very few out in the market that will say if you manufacture it this way, this is how it’s going to behave,” Kenik told TCT. “What we’re seeing is a lot of the times in additive, these parts are completely dependent on the process.”
First targeting polymer extrusion processes, Kenik says the part orientation, infill density, infill pattern, skin thickness and many more design characteristics are ‘highly pronounced’ and therefore can significantly impact the performance of the part once printed. More than that, knowledge on how to achieve the right balance of properties like this is often tribal and therefore many users don’t know how to get the best out of the technology.
“We decided to come out with some technology that can help design engineers and manufacturing engineers design slicer settings for performance-based criteria. This is kind of a paradigm shift away from generative design because that’s on the design side and we’re attacking this problem from the manufacturing side,” Kenik said. “We assume that you have a set geometry and you’re going to bring it into a slicer, and you need help determining what the slicer settings should be so that you get the best part on the other side. That’s what we’ve done. We’ve developed some solvers [on Amazon Web Services] from the ground up and now users can set up a problem inside a slicer of choice.”
In the past, Teton has seen users of extrusion-based 3D printing just produce a solid part ‘because it’s the easiest thing to do’ and if it doesn’t work, it doesn’t work. The company has also seen users produce iteration after iteration after iteration which Kenik describes as ‘just a guessing game.’ Teton wants to provide solutions to both of these scenarios, allowing them to take advantage of more design options and better understand what’s coming out of the other side before it even comes to printing the part.
Within Smart Slice, users are said to be able to ask the software whether their design passes or fails against their strength and stiffness requirements, for example, with a thumbs up or thumbs down response coming back within a ‘matter of seconds.’ Teton calls this ‘as printed simulation’. Beyond that, the company has also embedded custom optimisation algorithms so that if a user doesn’t know which slicer settings they would like to implement, the software can optimise the design to give the user the fastest possible print time, the least amount of material usage, all while meeting the criteria that they have inputted. The gains from this, speaking generally, are said to be a reduction in print time of between 30-50% and a reduction of iterations from between 5-25 to 1-2.
Teton Simulation Smart Slice
In bringing this product to market, Teton decided to quickly align itself with Ultimaker and Cura because of the ‘open attitude towards these kinds of partnerships,’ but also because, “we’re not going to tear people away from their slicers.” Having assessed the industry, the company felt the slicer market was too saturated to build its own, so has instead made the decision to integrate with them and allow users to plug in the Teton capabilities to supplement what the likes of Ultimaker offer with Cura. In doing so, the company has had to also line up with the dozens of material suppliers who have print profiles on the Cura platform and so far 30 have been tested, with ten more materials to be added upon Smart Slice’s next release.
One of those 30-40 materials is the BASF Forward AM Ultrafuse PAHT CF15 which, in a recent case study, was used to print a brake lever with 30% cost savings and a 35% reduction in material usage. BASF also concluded that Smart Slice’s optimisation of material characteristics and the considered load case allows users to minimise print time, increase machine throughout and minimise material usage, suggesting the combination of the software and its materials ‘enables FFF to be profitably integrated into everyday industrial applications.’
During this study, five configurations of the brake lever with varying numbers of walls, top/bottom layers and infill percentage were evaluated, with three of them being optimised by Smart Slice’s algorithms. All parts were printed with a layer thickness of 0.2mm and printing parameters specified for the Ultrafuse PAHT CF15 by BASF. The fourth configuration – and one of the Smart Slice optimisations – was considered to be the best compromise across the five different designs with its 20% infill (compared to 95% in configuration two, one of the manual configurations), 80N/mm linear elastic stiffness (compared to configuration two’s 100N/mm), material usage of 30g (compared with configuration two’s 46g) and 2h 10min print time (compared to configuration two’s 2h 50min). In the study, BASF noted how ‘the infill percentage of the used brake lever can be reduced to 20% while the stiffness of the part is still at 80% of the initial value when printing it dense.’ It also suggested the reduction of weight could lead to new applications areas, like transportation, where lightweight structures are required.
Teton believes the case study – which can be read in its entirety here – proved out Smart Slice’s capacity to deliver ‘as printed performance’ while also ensuring trust.
“That brake lever had a decent amount of load on it when we were designing it,” Kenik said. “We wanted to make sure that was stiff enough because the last thing you want when you’re gripping a brake lever is it be super compliant, or you just mash it down and nothing happens. We were able to do that and also, we were able to optimise the slicer settings to get the performance that we needed out of it while also reducing print time. Another important thing for our customers is to have trust in our software, so a big portion of that case study was comparing the predictions from smart slice to experimental data that BASF collected themselves.”
Teton began working with BASF last year and says it is willing to work with any company that is focused on industrial and structural applications. The company intends to expand the number of FFF suppliers it works with and also broaden out into other 3D printing processes in a bid to put the capabilities of Smart Slice into as many hands as possible. For Teton, the more industrial the focus and the bigger the volumes, the more value its software can offer.
“If somebody spends 100,000 plus Euros on a 3D printer, you want to have the most efficient workflow that you can, you don’t want to print one part and it takes you a month, you want to print parts as quickly and as efficiently as you can,” Dalgardo said. “If you can use technologies like ours or other simulation technologies to reduce the print time per part, that really compounds over the lifecycle. If you’re printing 50 parts or 1,000 parts and you shave off three hours per part, that’s incredible time savings.”
In order to facilitate those kinds of returns, Teton has started to build out its go-to-market strategy, with Dynamism recently announced as a reseller. This strategy is being pursued to fall in line with most hardware suppliers who lean on reseller networks.
“Those resellers are the ones that have all the customer connections,” Kenik said. “When you look at it from an aspect of how we reach the customer base, the reseller network is a no brainer.”
Teton sees that, generally, software lags behind hardware but once it catches up, is able to enhance significantly the efficiency of the machinery it has been developed to supplement. Capable software, the company believes, helps to raise the trust manufacturers will have in 3D printing, in turn driving sales of 3D printers and beyond that, driving applications of the technology. Teton is not blind to the challenges at hand but since its launch in September has demonstrated that steps are being taken to address some of the most prevalent among 3D printing users.
“Design for additive manufacturing (DfAM) is a really nice term, but I don’t think anyone has actually done this,” Kenik assessed. “It’s like, we can produce organic shapes out in the market, and we claim that these are DfAM but really you haven’t designed for the process, you don’t understand the influence of the process on the performance of the part. [And because of that], in the jigs and fixtures market where there is a structural application, the knee jerk reaction is I’m going to mill this thing because I don’t know what’s coming out the other end of the printer. If we can help those users build confidence to say, ‘I’m actually going to print it and I’m doing to do that because I know what’s coming out the other end is going to work the way I intended it to work, I know I can make it right the first time, and then my lead times go way down.’ Where we’re focused is just instilling confidence that, ‘yeah, I know I can do this.’”
Want to discuss? Join the conversation on the Additive Manufacturing Global Community Discord.
Get your FREE print subscription to TCT Magazine.
