Managing tolerances & quality for mission critical applications

Carl (00:00):

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

Speaker 2 (00:06):

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

Brian (00:12):

And 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 (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.

(00:34):

Welcome to the Solutionology Podcast, #7. And we’re going to talk about enabling additive for mission-critical applications. How’s it going today, Brian?

Brian (00:46):

It’s going well, but not as good as that kickoff.

Carl (00:48):

Yeah. Well, I appreciate that. There’s lots of questions that we get about additive manufacturing and what can be accomplished and the scale of it, but probably one of the most common questions is what are the tolerances and what do you do to control and achieve those tolerances? And there’s some assumptions out there that you can just run an additive manufacturing machine, whether it’s production or prototype, and it just produces the same part every time. And of course there’s assumptions that that doesn’t happen and we get to experience the realities of that.

(01:26):

Can you talk a little bit about from an operation standpoint, what we’re doing to manage manufacturing tolerances and how that’s looked for us? We push the technology pretty hard, but what have we done that’s different from the norm since we’ve been running our machines?

Brian (01:48):

Well, there’s probably two sides of that. We have to manage tolerance expectations for parts that we receive in that we have minimal control over the geometry, and then we have managing tolerance expectations as we are gearing up for production on a product line. And the aspect of gearing up for tolerances on a production line or producing a product at volume and managing those tolerances over thousands, thousands of parts. That’s the key area for us, and that’s what enables us to produce product and mission-critical applications.

(02:24):

So for us, as we’re onboarding a product for production, we’re gathering data, we’re doing analysis on what is the client’s requirements, where do they have tight tolerance, where do we have to mitigate that tight tolerance risk? And then of course, the data that we can get back out of our CMM allows us to work with them to modify the geometry so we can hit their specs and hit the requirements. Or we also use it for ongoing quality checks to make sure that their product is meeting the spec throughout that production cycle.

Carl (02:57):

So basically, we’ve made a change recently, right? You had mentioned CMM and we’ve onboarded a 3D scanning CMM that allows us to basically do deviation analysis and assess the printed part versus the design. That was something that we added on recently. And before that we were using traditional quality inspection tools to be able to inspect parts, and that gave us a solid level of understanding of the dimensions of the parts. But for non-standard geometric features, those traditional tools were difficult and it is not scalable running full-scale production. So we brought on this 3D scanning CMM, and that’s what you’re talking about. So what does that mean and how does it work?

Brian (03:58):

Looking at the past ways of managing quality going through, we’d use go and no-go gauges. We’d use gauge pins, and we use measurement devices to validate the tolerances that we’re getting off the machine, but we’re really limited on if it’s a slot shape hole or if we’re measuring draft on a part, or if we’re measuring a surface profile. We can’t use those tools in most cases to accurately measure what’s the variation surface profile that we have on this naturally non-geometric geometry. So that non-touch-based CMM 3D scan CMM allows us to visually inspect that part and then do a deviation analysis on CAD and even GD&T analysis for datum-required specifications.

Carl (04:55):

So those are tools that we’re using for both quality assurance and quality control, and to a certain degree DFAM, right? So if we’re bringing a new product on board, and it might be that it’s not economical to achieve the tolerances desired with the base design, then we’ll do an analysis and then actually change the design of the part to achieve the desired tolerances. Can you talk a little bit about what that looks like?

Brian (05:24):

The tolerance stack up for additive seems simple on the surface. We get values like this machine can hit plus or minus eight thousandths, but those values don’t take into effect what is the orientation of that part in the build, which is we control that for high tolerance applications. And what’s that tolerance at that orientation for the dimension that the client’s requesting? So it is not a linear plus or minus for us it’s orientation-based, it’s a three-dimensional equation. So what this tool allows us to do is capture that data part after part, understand how that variation is occurring, and then we can either change orientation, change the production process, or we can go back to the design and modify that design so we’re hitting final specs.

Carl (06:11):

And so far as we’ve been using it, it’s given us a level of detail of the part that we simply couldn’t achieve previously. We can only go so far with the traditional tools and go/no-go gauges.

Brian (06:27):

That’s right. There’s assumptions that have to be made when using traditional measurement tools on complex geometries. Simply it’s impossible to measure all the dimensions accurately.

Carl (06:41):

One of the things that we’ve learned as we’ve gone from prototyping to production, or pilot runs then into production is that even with a locked build, there can be variability that exists in the build or locations of that build. So understanding what that variability looks like and then how to design and process for it is really critical to being able to scale with additive and achieve tight tolerances even into long production runs.

Brian (07:14):

And a tool like this allows us to do those long production runs and be able to capture all the data effectively and be able to store that data with that lot controlled product that we’re shipping out to the client. So that everything’s contained, it’s data we can look back upon and we can guarantee that our client is getting product within specification.

Carl (07:37):

So we’ve got traceability, right? We’ve got dimensional traceability through these production orders, and we’re able to go back and trace that using the fair process.

Brian (07:50):

Yeah, so if there’s one valuable piece of that for us is as we’re doing lock controlled orders that are often build based, if there’s been an issue with a machine that hasn’t shown itself from an error code or downtime on that machine, we’ll be able to recognize that defect in the measurement analysis through QC, which will allow us to pull that lot of product out of the mix and QC hold it until we’re ready to either release it or scrap that product so it doesn’t get off to the customer. The CMM allows us the ability to mitigate the risk of our clients getting a part or a product that isn’t meeting their specification, which what sort of applications are we managing that for? Because if we’re making trinkets or pencil holders, that dimensional accuracy isn’t required. So can you speak to the applications?

Carl (08:55):

Yeah. So we’re not using the technology for all the applications and really we invested in the quality technology, the 3D skinning, CMM technology for high value mission-critical applications like medical device production, not just housings, but actual functional assemblies where there’s assemblies that work together to drive to an action. DOD type projects that are in use, tactical components that have a critical level of both traceability and tolerance, and then applications like inflight components where we have to provide and document traceability of the quality of the parts. The dimensions of course, are one aspect of that. So far we found the equipment to be extraordinarily useful in those applications. It just gives us a level of detail and significance of a data set that we haven’t had previously.

Brian (10:03):

Yeah, I think the interesting thing, the tool’s really useful. The tool allows us to collect data, and what we’re able to do with it at DI Labs is we’re able to analyze that data and we work hand-in-hand with our clients to make sure that we’re managing the importance of their project and working with them on the design side to bring it all together. It isn’t just about handing over a dimensional sheet, but it’s helping them as well manage their design so that they can be more effective as their releasing product.

Carl (10:38):

There’s no doubt about that. And the other thing that’s pretty interesting is thinking about how we’ve, our processes and our equipment to be able to produce consistent products time after time. And we’ve pushed that pretty far with traditional measuring tools, collecting the data. And what I see us doing with this 3D scanning CMM is taking that richer data set and taking that even further. We’re able to now fine-tune the equipment and the processes for ongoing production. It should have a cascading impact for all of the production that we do because now we understand the equipment that much better and are able to push it further. And I see real benefit for specific clients and specific projects, but also broadly across the board, we’re going to be able to control the machines that much better.

Brian (11:36):

It’s an important piece of the feedback loop for us to continue to tweak and improve our manufacturing process.

Carl (11:45):

So what is the use of the CMM look like from an onboarding of a production perspective, from a PPAP perspective? What role does the CMM play?

Brian (12:01):

So when we’re doing a production part approval process, we will bring the CMM online when anytime that we have a product that has dimensional requirements with a drawing that has GD&T requirements on that drawing, or the client has expressed concern with fitment or dimensional accuracy. So we’ll bring it in early as we’re doing our first sample runs, and as we’re collecting more documentation from the client on what their expectations are, we’re scanning that product along the way so that we’re preparing ourselves and them for the first article inspection. And that allows us to get to that first stage of handing over the first article inspection and being on the same… We’re at the same place, there’s no surprises. And having the tool and having the process in place allows us to expedite that process and not go back to the beginning and start over because of a missed opportunity.

Carl (13:03):

That has a big, big impact.

Brian (13:05):

It has a huge impact. Huge.

Carl (13:07):

I’m glad I added something of value on that one. It’s game changing.

Brian (13:13):

The topic of quality, it isn’t sexy for the… Most people, it isn’t exciting, but it has to be a matter of fact, and it’s a dry topic. But what happens when you don’t have quality that’s not so dry. There’s a lot of frustration that can occur there. So quality is important even if it is a little dry.

Carl (13:34):

Yeah. So what’s the opposite of sexy then? Ugly. So it gets really ugly when you don’t have quality, and that’s where you see it, and that’s what you’re talking about. It’s been interesting for us as we continue to push the boundaries with production scale on our technologies, primarily powder bed fusion. It continues to be interesting for us to learn the variation that can exist throughout a build process. And now we have to continue to control that under circumstances where for all intents and purposes, it’s locked the design and the builds locked, but we have to continue to control that.

(14:15):

So the takeaway there is that additive manufacturing, even some of the best and most expensive technologies still have to be controlled on an ongoing basis. And it’s not pushing a button once you have the design locked and then forgetting about it, it still has to be managed. And that’s been something that from my standpoint, we continue to learn. We continue to push the boundaries as we experience new learnings at different levels of scale, and it continues to evolve as the scale increases.

Brian (14:48):

A lot of our clients have found us because they haven’t been happy with the quality of product that they are receiving today, and they come to us looking for a solution to that quality question. It’s been our focus from the very beginning to deliver high quality product and to make sure that the parts that they’re requesting, we’re meeting their requirements.

Carl (15:11):

And when you think about it in that way, this podcast is called the Solutionology Podcast. And so our investment in the CMM technology is really to solve those problems. And in some cases, we’ve pushed the equipment as far as we can with traditional quality inspection processes. And this tool allows us to push it even further. And why a customer cares or why a client should care about this is that we’re essentially working hand in hand with you to push the technology as far as it can possibly be pushed, understand what levers we have to be able to achieve the desired tolerances, especially for high tolerance, high value applications. And I don’t see us at the end of the road now we’ve got the technology, we’re going to continue to push it, and I see us continuing to push the boundaries on additive manufacturing tolerance. And I still think there’s a long way we can go. We’ve come a long way.

Brian (16:16):

Yeah, it’s really another tool in our pocket as our focus has been designed for additive. And maximizing the capability and the efficiency of that product, whatever it is. And this additional tool allows us to fine tune that maybe another 10%, which is big when we’re trying to get everything we can out of the production. Another 10% is high value for us.

Carl (16:41):

We’ve been able to achieve tight tolerances with additive manufacturing. And you might ask, how have we been able to do that? And from my standpoint, there’s several aspects of achieving and maintaining tight tolerances. First, you have to want to. And then second, you have to know what your tolerances are. And then third, you have to be able to design the manufacturing process to achieve them. And in an area where I think we’re a little bit different than some others out there, is that we’re not running our machines to push as much throughput through them as possible. We’re running our machines to manage consistency and quality, and those are two different sides of the coin.

(17:29):

And we found that when we start running those machines really, really hard, increasing pack densities, increasing throughput, unpacking them fast, there’s a great deal more variability in the outcoming parts than there is if we’re running them for quality. So I think that’s the first part of starting to achieve higher tolerances. And now that we’ve got this other technology, we’re going to be able to understand what other levers we have to pull in order to achieve even greater consistency.

Brian (18:02):

And our culture is really founded in quality is a piece of our culture. As we’re bringing on new work cells, as we’re educating our team members, as we’re going through that process, we’re automating the mundane tasks. And that gives all of our team members a focus on the quality output. We’re focused on what are the parts, are we hitting the requirements of the customer? We’re not worried about the buttons and knobs in most cases. We’re focused on the parts.

Carl (18:31):

We had been looking at the three-D scanning CMM for some time. It’s been something that’s been on our radar for over a year, and it’s an expensive piece of equipment. It doesn’t generate any revenue. It’s hard to justify. And we finally made the decision to pull the trigger. And there was a couple of reasons for that. One was because we had a project that required a significant amount of data collection, and doing that by hand was tedious and not the best use of time for our team. So they didn’t really love it. And I don’t blame them because quality’s not sexy. But at the same time, we started to uncover some opportunities to achieve better performance. So we made the investment to buy the technology. Can you talk a little bit about that project and how we’ve been leveraging the CMM on that project and how it’s changed the process for us and also the deliverables for our customer.

Brian (19:40):

So being able to amass that data collection, all the values, the critical values of the product for our client, we’ve been able to relay that information to them and we’ve been able to be open about dimensions that we just wouldn’t have been. You can’t measure with traditional measurement tools. And that was eye-opening for both sides because their expectation was that those parts were in spec. Those parts were manufactured by another before we got on this product.

Carl (19:40):

Another manufacturer?

Brian (20:15):

Another manufacturer before the project came to us. And those dimensions were assumed to be accurate, but having the 3D, CMM allows us to guarantee that those measurements are in a value. What we found is they weren’t. So we were able to work with our customer to either adjust the model or adjust the requirements so that they knew what they were getting on the delivered goods, and we knew that the delivered goods met the requirements.

Carl (20:50):

Yeah, so basically, if I understand, the project came to us, it had been previously produced, it was essentially locked with the assumption that the parts were meeting the tolerance requirements. And then we did a study ourselves using the same parameters that were previously used and found that they weren’t actually meeting the requirements. So we made adjustments that otherwise we wouldn’t have known about had we not leveraged that technology. The CMM.

Brian (21:21):

Yeah. So we’ve reduced the liability for the full program because the dimensional values are transparent. We know what’s going on, and so if there’s any questions about where we are, it’s all documented. So we’ve eliminated that risk or that liability for both us and the client.

Carl (21:40):

Obviously there’s a great deal more confidence for all of us as we’re working together to bring the product to market.

Brian (21:47):

Yeah, we’re aware of what’s going on and there’s nothing hidden for us to navigate.

Carl (21:53):

Do you see us using the scanning CMM? Do you see us using that for all applications?

Brian (21:58):

Not all applications. It’s really about assembly, tight tolerance fitment. If we have multiple parts mating together, and especially if there’s moving parts within that assembly, there’s a tremendous advantage for us to be both on the product development side, helping design that product with the deviations that we know that we have to live with, and making sure that we’re working with the design team so that they’re incorporating those values, they know what they are, and they can make adjustments from today and in the future that are functional.

Carl (22:36):

Well, it’s a great resource and I’m sure that we’ll be leveraging it for more and more high tolerance projects, and we’ll continue to push the boundaries on the additive manufacturing technology.

Brian (22:51):

So if you find yourself in a place where you have product today and the quality isn’t meeting your expectation, we’d love to give us a call. We’d love to work with you on your project and find a way to use our tools and our process to meet your project requirements.

Carl (23:08):

Right on. Thanks. See you soon.

Brian (23:11):

See you soon.