The “Art” of Onboarding Materials & Technologies

Carl Douglass (00:00):

Solutionology is about being unyielding with perseverance to get to the solution.

Sean Douglass (00:06):

To not give up and to constantly drive for better. So even when we deliver 100%, I want to deliver 110% next time.

Brian Douglass (00:12):

For me, the constraints of that project are the most important, because that’s what drives us to a solution. It’s all about painting a picture and getting all the details in.

Carl Douglass (00:20):

How do we develop a tool that helps share our journey, educate others, and bring more light to the realities of additive manufacturing? The Solutionology Podcast.

Brian Douglass (00:34):

Welcome to DI Labs podcast number four. We’re going to be talking about materials today, specifically for the Fortus, modified materials, what that means, and some of our recent testing and results from Sean’s work here.

Carl Douglass (00:50):

This is a pretty neat episode, because we’ve got all three Douglass brothers here. This episode is going to be a little different, not only because we have three brothers here, but because of some of the topics we have to discuss.

(01:05):

I’ll start by introducing myself. So I’m Carl Douglass and I am responsible for future planning for the business, strategic relationships, as well as sales and marketing.

Brian Douglass (01:18):

Carl comes up with the … Part of that vision is coming up with the path that we’re going to go down next and aligning ourselves with the new technology or new materials to satisfy our client needs.

(01:31):

Part of that vision is being able to see what others don’t. And that’s been really powerful for us. As we’ve been on our journey of growth, Carl has had insight and vision into areas that were blind to me for sure.

Carl Douglass (01:45):

So Brian is the one that makes those visions a reality. As we’re looking to solve our customers’ problems, bring on new technologies, be on the forefront of the industry, we’re seeking out those visions. And then, Brian is putting the plans in place from an operations standpoint to be able to execute those. Without that, there wouldn’t be a reality.

(02:15):

There’s a lot of other things that Brian is involved in from leading our product development activities, both for our internal projects as well as our customer projects, and making sure that those projects are executed successfully. Either by personally doing the product design, engineering and development, or leading those activities.

Brian Douglass (02:38):

I’m the roadblock clearer.

Carl Douglass (02:40):

The roadblock clearer is right. So those are two parts to the equation, but the third part is just as important, if not more. Once we commit to a new technology or to a new business practice, we put the plan in place to be able to execute that from an operations standpoint.

(03:01):

The heaviest lift of all those activities is actually making it happen. That’s onboarding the technology, onboarding the materials, and being able to do so, so that we have consistent manufacturing, consistent quality, and are ultimately meeting our customers’ needs.

(03:19):

So Sean leads all of our manufacturing engineering activities. Everything from the day the printer shows up to the day that the materials are onboarded to the following days, where we’re optimizing those technologies or machines and making the process run more effectively. Without Sean’s engineering expertise, that stuff just wouldn’t happen.

Sean Douglass (03:47):

And then, I’m able to connect with Brian and go back to the operations side and make sure that we’re connecting not only on what we’re currently doing, but what we’re doing moving forward to make sure that aligns.

(03:58):

Also, going back to Carl and making sure that everything we’re doing goes top to bottom, so that everything’s optimized, but also it meets our ultimate goal of where we’re headed.

Carl Douglass (04:09):

To say that we’re lucky would be an understatement. Being able to work the three of us brothers together and to work together in different segments of the business is really pretty special. We each have an area of our core responsibility or our core capability, but then there’s a little bit of overlap across all three, so we’re able to ultimately drive success working with our team and the technologies that we onboard.

(04:41):

I’ve never experienced anything like what we have working together in our business. I’ve never experienced anything like that anywhere else in my career. I feel pretty lucky to be sitting at the table with you guys. Such great talented individuals and devoted.

Brian Douglass (04:58):

It always takes a team.

Carl Douglass (04:59):

It takes a team. There’s no doubt about that. Back to our Free the Fortus campaign, we’re all here because we have these different responsibilities. Each of these responsibilities have been really critical in order to bring on the Fortus machine and be able to bring on the validated materials in OpenAM.

(05:19):

From an update standpoint, we’re five weeks, maybe six weeks into the beta project. And the project hasn’t gone as we expected it to. Like any other new product launch. We’re on our own track right now. Right?

Brian Douglass (05:40):

Yeah. Since our last update on Free the Fortus podcast, we’ve skipped a couple of weeks because of delays and challenges that we’ve had in going down a path that we thought that we were headed down.

(05:52):

That timeline has changed, and that really brings us to today, where we’ve made some progress with the Fortus and we’ll share that, but then the next steps for us are equally exciting.

Carl Douglass (06:05):

So it’s becoming a journey. I think what I’ve heard from Stratasys, and I think it’s been reinforced for ourselves as well … We keep hearing from them the value that they’re getting from us in fulfilling our responsibilities on the beta and doing the development.

(06:27):

I’m not as close to the details as you guys are, but I feel like we’re offering them tremendous value. Almost to the degree that it’s like we’re part of the team. We’re part of the Stratasys team because of the level of depth that we’ve gone into, that you guys, specifically Sean, you’ve gone into. Pushing the software, pushing the machine, and really trying to bring it online, so that it’s of very high performance and not just checking the box, so to speak.

Sean Douglass (06:57):

Absolutely. We have a lot of time and understanding, not only of the OpenAM, but the 450MC. Making sure that when we’re adjusting a lever or pulling a switch, we know exactly what that means. Whether it’s speed or extrusion variables, we’re trying to fine-tune those in a way that everything that we’re doing is very meaningful. Each step is.

Carl Douglass (07:24):

Sean, we’re five or six weeks in. How many hours do you think that you’ve got invested in the efforts that you’ve done so far?

Sean Douglass (07:36):

Between five materials that I’ve worked on, I probably have 20 to 30 hours a week, each of those weeks. Last week was a little lower than that, just because of the OpenAM, but in that ballpark. Bringing materials to reality.

Carl Douglass (07:54):

Well, it’s a good thing you don’t work a normal 40-hour a week.

Sean Douglass (08:00):

I don’t know what a normal 40-hour week is.

Carl Douglass (08:02):

Well, I’m not going to ask you what your week looks like, because I think it’d be scary for everybody that hears. To say that we live and breathe DI Labs would be an understatement. There’s no doubt. So you brought on, you said five materials?

Sean Douglass (08:19):

Yeah. We have five materials that we’ve gone through and done initial testing and development through. Some of those are anywhere from 75% to 95% complete in terms of development. And then, we’ve got an additional material sitting on the table, which is 9085 carbon fiber, which we’ll be working on hopefully later this week.

Carl Douglass (08:44):

You’ve onboarded those materials. Can you talk a little bit about that process that you go through to bring on a new material? What do the primary steps look like to bring on a new material?

Sean Douglass (08:57):

That’s a really important question. I think Brian mentioned in our previous podcast talking about extrusion parameters. That’s really the first steps, understanding what the extrusion parameters are. Temperature, extrusion speed, and then chamber and bed temperature. Making sure that all those things align, so that when you start your profile that you can work through it.

(09:19):

Without those, it’s really throwing a dart of the dartboard and hoping that you land somewhere inside the range. Until you’ve got that dialed in, it’s really tough. You use the manufacturer’s recommended, and then go from there, but it’s definitely not a one-and-done. There’s a lot of revisions there to get to a good profile.

Brian Douglass (09:42):

As you pick up a new material, it’s got recommended specifications for how to run that material. Why is it that we can’t just plug those values in?

Sean Douglass (09:52):

Thinking through the mechanical extrusion of that, each system operates slightly differently. Even in open source machines, what they say, call it … We’ll say 400 C. It may actually be operating at 385 C, so what temperature it actually is reading.

(10:10):

And then, there’s additional when you’re driving that filament through your thermodynamics through that system. It’s variable as you’re adjusting at different speeds. Identifying those and working through that for each material is very important.

Brian Douglass (10:27):

Those guidelines that we get on a new material, they’re really just guidelines. Fine-tuning it to the machine that we’re running it on due to offsets and temperature variances, that’s where the magic comes in for us to be able to produce a part that’s strong, use those materials, and be successful and go into production.

Sean Douglass (10:48):

Absolutely. That’s really the whole reason why Stratasys is opening up for these materials is because even their base validated materials are really good settings, but there’s a lot of applications where they just aren’t 100%. They’re tuned but not fine-tuned.

(11:05):

Opening it up and allowing the customer to fine-tune it for a specific geometry or their very specific application is very important. We’ve learned that multiple times for our open-source machines. We may have one geometry that is very specific, and then you may have another geometry that operates at either a different temperature or a different extrusion speed because it’s very specific.

Carl Douglass (11:30):

From my standpoint, this is a lot of information that’s tough to unpack. When you’re thinking about buying an off-the-shelf printer, an FDM printer, and then just plugging material in and printing parts and the parts come out okay. It feels like … It doesn’t feel, it sounds like without knowing all the nuances of the materials and the processes, it’s like, “Are we just overworking it?”

(11:57):

And that’s a question that’s not a real question. The answer is, “We’re not.” And the reason that this beta is so important to us is that when you’re printing an ABS or a PLA material, sure, the processing windows matter, but they’re actually pretty wide. You can get away with a lot of process mistakes or lack of fine-tuning and get reasonably sound parts out, especially if you’re just looking at prototypes rather than truly functional parts.

(12:24):

But where I’ve experienced working with you guys, where it’s really critical is when you start working on some of these higher temperature materials, some of these higher-end materials that have modifiers or fillers. Even those without fillers like PEEK, for example, is probably one of the most challenging materials because that process window is so dang tight.

(12:47):

That’s where when we’re talking about dialing-in these materials, it’s generally around really high-end, high temperature materials that don’t work with just general settings. Those are the types of projects that we’re most focused on trying to help solve critical problems for our customers. The PEEK … Is it fair to say that the PEEK has been the most challenging material for us to run in either our open source or the Fortus?

Sean Douglass (13:15):

Absolutely. I think it’s challenging for anyone to run. Even the feedback we’ve gotten from 3DXTECH and from other companies, the reason why PEEK is not in a lot of machines is because it requires a great deal of understanding as well as variance from a manufacturing as well as from an extrusion standpoint.

Carl Douglass (13:35):

Some of those settings for materials like PEEK are geometry-specific, right?

Sean Douglass (13:40):

Absolutely.

Carl Douglass (13:40):

Both from a size and feature standpoint, some of those settings need to be further fine-tuned to account for those specific parts.

Sean Douglass (13:50):

Absolutely.

Carl Douglass (13:51):

So one of the things that I’ve got on the screen here is … It’s just a snippet of some of the data that you’ve compiled, the notes that you’ve taken, Sean, throughout your different studies. You’ve got a lot of different geometries here and you’ve got all these different revisions.

(14:10):

Can you talk through what the process looks like on the screen and how you’re thinking about and executing these setup profiles and the different steps? What’s the method to the magic?

Sean Douglass (14:26):

For each one of these material setups, I start with a basic geometry and I’m trying to dial-in extrusion parameters. So if I were on an open source machine, I would be setting it up, extruding at a certain temperature, understanding how it’s flowing through. And then, is it in crystalline form or not in crystalline form?

(14:45):

And then, from there, I would estimate where I need to be. And then, I would start a test print on the Stratasys. Starting with a range, compiling that information, and then running a part. And then, trying to understand what surface quality looks like, seams look like, and additional information across the range of parts to then fine-tune.

(15:06):

Am I at the right temperature? Am I going too fast? Am I going too slow? And then, where I need to adjust, so that I can meet the surface quality requirements. Particularly, starting with the exterior and then moving into a support and/or infill.

Carl Douglass (15:23):

Just scrolling through this real quick. You call these pages, but I almost feel like you have chapters here. There’s so many revisions that you’re going through. In some cases, it looks like you’re pushing the boundaries on a particular geometry.

(15:37):

Maybe you take it too far and then you back up and then you go a different direction. This looks like it’s just a ton of detail required in order to get this right.

Sean Douglass (15:51):

For the PEEK, it was a lot of trial and error. Trying to push the boundaries, knowing what we knew with the open source machine that Brian helped set up, and using those settings to really say, “Here’s where we think we need to be.” And then, trying to apply that to the 450, and then realizing, well, it operates slightly differently from accelerations and overall setup and trying to manage that in the background.

(16:15):

Then, go through and say, “Okay. Go up five degrees. That was too much. Go down two degrees. Okay. That’s the optimum temperature. Go up 10% on the speed. That was too much. Go down 5%.” Just keep dialing it in. And then, I would go to a new geometry and say, “Well, that didn’t work for this geometry. I’m going to go back and try either a different profile that I’ve already accumulated or make updates.”

(16:38):

And then, I just continued to work through that process until it got to be where almost all the parts were coming out consistently across multiple geometries for different revisions of this code, so that I could say, “For this style of geometry, I would run this revision. For this style geometry, I would run this revision.”

(16:56):

And I just kept working through that. And then, once I got all that done, we rolled into ULTEM 1010 Glass, or ESD, and I immediately took all that learning I did from PEEK and applied that. Within two revisions, we were printing off parts that were pretty satisfactory. So all of that learning goes towards the final solution.

Brian Douglass (17:20):

What’s neat there is the amount of struggle that is faced with a really challenging material like PEEK. Going through learning that process, and then switching to a material that’s a little bit more friendly, which I don’t think anyone would say that the ULTEM 1010 is super friendly, but being able to take a step into that and have quick success by applying those learning lessons, which is really the new product development approach that we take to all things.

Sean Douglass (17:47):

Yeah. I think it’s the fail fast method of go in and try and knock off the easiest things that you can remove from the equation. When we started doing PEEK, we had some issues with extrusion. I switched over to ULTEM 1010, non-ESD, just to understand if it was a high-temp issue.

(18:05):

Validated it wasn’t, moved right back into PEEK, and then fine-tune from there. So all these little areas, if you can understand what’s going on, you can apply your learning lessons to each material. That won’t necessarily apply to ABS, but it will apply to our high performance materials.

Carl Douglass (18:23):

There’s a lot that goes into making this stuff work. When you think about it, you see a lot of the marketing materials and the content that’s out there about additive. Primarily, from the machine manufacturers.

(18:38):

You see a lot of that content and everything just looks beautiful and it looks easy, but as we get into some of these higher-end materials, it’s really anything but. It can be easy to gloss over the details and make parts that look okay or that look good, but don’t function real well.

(18:57):

There’s really a ton that goes into making the additive manufacturing machine work consistently and produce really good parts. We’re talking about a manufacturing process, and I’m not sure that that’s really appreciated by the general public, by consumers of additive, because in most cases you don’t have to battle at that level.

(19:26):

But I appreciate that that’s something that we take seriously at DI Labs. We’re really pushing the hardware, we’re pushing the software to make sure that it’s producing parts that don’t just look good, but function well, and that they’re processed under the right conditions. That to us is really important, and we’ve seen the positive outcomes of that.

(19:53):

When you put a part into a component that’s been printed, you put it into a demanding application and it performs well, that’s worth it. Every one of the materials that we run in all of our technologies, they go through an onboarding process that … Maybe the specifics are different than this, but everything goes through an onboarding process to get validated before we put it into production. Is that right?

Sean Douglass (20:21):

That is correct. Everything on our MJF side, our DLS side, all the FDM has all gone through some level of onboarding to validate geometry, print orientations, strength processing, post-processing, and all the way out to the door to make sure that everything meets our requirements before it even meets our customers’ requirements.

Carl Douglass (20:46):

It’s like a full-time job.

Sean Douglass (20:48):

Some days. Some days.

Carl Douglass (20:50):

Probably, three full-time jobs. No. It’s neat to think about that. Hopefully, sharing some of those details helps shed some light on the realities of using additive manufacturing as a manufacturing service or as a manufacturing tool.

(21:08):

Again, I think a lot of people look at it and think that it’s pretty simple. You push the button and it’s almost like a paper printer and stuff comes out. Maybe at the surface it is that way for simple parts, but it’s anything but when you start getting into truly demanding applications.

(21:24):

We started this podcast with a focus on Free the Fortus. We’ve never done a podcast before. This was a concept for us. It’s like, “How do we develop a tool that helps share our journey, educate others, and bring more light to the realities of additive manufacturing?” So we learned a lot as we’ve gone through it.

(21:48):

We’ve had some great feedback. We’ve had some great questions from customers after the podcast. As we’ve learned this and we also have other projects going on at the DI Labs, we’ve started to think, “Well, maybe this should be more than just Free the Fortus?” So this marks a transition for us to expand the podcast beyond just Free the Fortus.

(22:13):

We’re recharacterizing the podcast as The Solutionology Podcast. Everything that we’ve talked about here, Sean, everything that you’ve talked about is all about problem solving. Sure, it’s for additive manufacturing. It’s for the Fortus. It’s for Stratasys. But everything we do is all about problem solving and it’s taking an engineering approach, a technical approach to systematically solve problems.

(22:41):

Today, the podcast is reformed a little bit. We’re looking at it in that we’re elevating the podcast from Free the Fortus to Solutionology. And of course, Free the Fortus will continue to be a component of the podcast because we’re going to continue on with that effort.

(22:59):

But for me, it’s exciting to think about this in a broader landscape of sharing our journey about how we solve problems, the types of projects that we’re working on, sharing stories with different customers that we’re working with. So I’m really excited to think about this expanding into a broader horizon. There’s tons of applications that we’re working on that this could be pretty powerful for.

Brian Douglass (23:26):

So as we go into the future, we’ll be talking about, of course, modified materials. Mechanically modified materials for FDM. That’s part of our future. Getting into some other new materials that we have, like White Nylon 12, which is a new material that we’re in the process of testing today, as well as how we build and create textures on parts, and how a product might come into our system as a napkin sketch and flow out as a finished part within 14 days or so.

(24:02):

That is really the foundation of what we do every day at DI Labs and zooming out a little bit and getting a bigger picture of these components is part of our journey.