Q&A: Pipeline Design & Engineering

What is the story of starting Pipeline?

Pipeline Design & Engineering started about 13 years ago. It was 2008-2009, kind of the recession period, and I’d just got laid off from a previous job at a different engineering services company. My employer definitely made the right call, though — I had grown disengaged from the work. I wasn't super excited about it, and I didn't know why.

I thought maybe I'd fallen out of love with engineering. But after some searching and personal reflection, I realized that it wasn't necessarily engineering, I had fallen out of love with just the way in which I was doing engineering. So I decided to give it a shot on my own and started Pipeline.

What was it like building your business and getting Pipeline off the ground?

It was just me in the beginning. I took on some really small jobs, got paid peanuts, while working, you know, 70 hours or more each week. But I absolutely loved it. I wondered why I was so much happier despite working way more and not making any more money, and realized I loved having ownership over the entire process.

A few years later we brought on our first employee, then another one, and another one. Now at this point, there are 20 of us plus a few contractors.

How did Pipeline evolve since its beginnings?

In the early days our focus was general product design. Like consumer products and medical devices, also some solar power equipment: it was pretty broad in general.

But as the years have gone on we've refined our focus. At this point, the vast majority of our business is developing custom equipment, test fixtures, and automation. We also do some software and programming in-house. Most of our business is in the machine design space, which is why Markforged is so perfect for us. It’s an incredible replacement for parts that are historically machined out of aluminum or stainless steel, or even Delrin.

How were you introduced to 3D printing?

I was first exposed to 3D printing at an engineering company that I worked for right out of college. It was an engineering services company for product development, and they actually had an entire division dedicated to just rapid prototyping through 3D printing and urethane casting. A lot of our work was in product development designing medical devices.

This was back before 3D printing was mainstream, but access to printers became normal for me since we used them all the time.

What was your experience working with 3D printing back in the old era of rapid prototyping printers?

We didn't have the materials then that we have now, so there was limited use for 3D printing back then. We used a lot of SLA, some SLS, but not a whole lot of FDM back then though. The prototypes we got were very dimensionally accurate, they were great to see, touch, and feel. But they weren't so great for actual use in functional prototyping, in testing and validation: the materials were generally fairly brittle and tended to break easily.

So they were great for making sure things fit together, but not so great for evaluation from a functional standpoint.

When did you first hear about Markforged?

I had been tangentially aware of Markforged from seeing booths at trade shows and what I’d read online on different engineering sites. It was our local sales rep who really clued me in to the value Markforged machines could bring to our business, though.

He reached out and said Markforged printers would be a perfect fit for Pipeline, with all the fixtures and custom equipment we use. I’d been curious about Markforged printers anyways after seeing the machines at trade shows here and there, so figured I’d take a deeper look.

Sure enough, we found out Markforged printers are perfect for what we do at Pipeline. It’s a perfect replacement for machined parts.

Did Pipeline use 3D printing before purchasing your first Markforged printer?

We had dabbled in 3D printing early on at Pipeline. Before Markforged, we had a couple SLA machines. They were okay, but weren't super reliable. And again, the material choices were kind of limited, so there was only so much we could do with those.

One of the things that I really disliked about our SLA machines was the amount of maintenance they required. Prints would fail, and then you’d have to mess with all this goopy resin to clean it all out and get it set up again. I hated doing that, it really did not feel like a good use of my time.

What did you think once you brought in your Markforged printer?

I was amazed with how reliable it was — there were essentially zero failures. We probably printed over 100 parts before we had our first print issue. Very low maintenance, a massive shift from the other 3D printers I was used to dealing with.

Making any significant purchase as such a small company is a little scary. We're not a huge company now, but back then we were an even smaller company. Around $20k for a Mark Two was a big expense for us. I did worry and wonder to myself if it would pay off.

After the first print it became pretty clear that yeah, this is going to pay off. I don't remember what the exact payback period was to break even on the capital equipment expense but I think it was around six months.

How immediate was the impact?

From the first project, which was a tray table for an automotive application. There were around a dozen parts that we needed to prototype and test out, many of which were structural parts. In the past, we would have had to machine a lot of these parts — costing many thousands of dollars and taking several weeks for them to arrive.

We were able to print all the parts in one build. In around a day and a half we had all the parts, including the structural ones, and we were able to put them together and test.

That was kind of a lightbulb moment. Like wow, this is really cool. A day and a half and a few hundred dollars instead of three weeks and many thousands of dollars.

What would the project be without a Markforged printer?

Without our Mark Two most of the project’s parts would have been either aluminum or stainless steel, maybe a couple of Delrin parts in there. Together, the dozen parts would have come to maybe $3,000-4,000, somewhere in that neighborhood. It would also have been three weeks or so at the earliest to receive the parts.

It was under $500 in material cost with our Mark Two, and with huge time savings. It was only about a day and a half to print all those parts.

Has use of your Markforged printers expanded beyond what you initially brought it in for?

Yeah, it's been a paradigm shift for us being able to make so many of our own parts quickly and cheaply.

It opened up a new revenue stream for the company: selling our own 3D printed parts. We still order some parts through machine shops, but most of the time we just print them ourselves and save so much time and money doing that.

There's another project that we did that is a really good illustration of this. It was a quarter million dollar project to develop a custom piece of inspection equipment for a medical device company. We ended up printing most of the custom parts for that inspection station, and there were a lot. I mean, there were probably 50-60 custom parts that we 3D printed.

What were the benefits of 3D printing that many parts for a project?

Printing those 50-60 custom parts easily saved us over $30k, plus months of time. To be able to accomplish a pretty ambitious project like that using so many printed parts doesn’t just save a lot of money, though. It also allows you to deliver the final product to the customer months earlier. That's huge, not just for us, but also for our customers.

What can you tell me about the production parts on that inspection device? Any functional requirements the assembly was held to?

One of the high-precision parts on the inspection station utilized a laser vision camera. It needed to perform a pressure decay test and measure RPM of a shaft on this medical device needed to measure input torque. It needed to accommodate around five different devices within this family of medical devices.

When we delivered this station, our customer saw how well all the parts worked and held up to daily use. At that point, floodgates opened up for them. They said print as many of the parts as you can moving forward: we see the benefit, the cost savings, the time savings.

Has getting Markforged printers changed the way your company works or operates?

Well, we print everything we can now. Our customers always need to reduce their costs. So that carries over to us, and we get pushed to find ways to bring our costs down.

Markforged has been huge for doing just that. Not only can we reduce costs, but we can reduce lead time now as well. It’s been huge in prototyping as well. Our engineers just print everything and test right away. You come up with an idea, you print it, you test it, you find some aspect that doesn't work quite the way you expected. Then update the design, print it overnight, and the next day you have the next iteration.

In the past we've used machine shops that have a pretty quick turnaround time, right, maybe three to five days you get your parts. But even then you're still waiting three to five days to get parts. And it's certainly more expensive than printing something in house.

One example is a project that we did for a different large medical device company. There was a two week period during which we were printing a new variation of a part and testing it nearly every single day. At the end of those two weeks we had the perfect geometries for a critical part with a few different components: a disc component that needed to swivel around a ball joint, with air flowing through as well, kind of like an upside-down air hockey table.

It was such a huge improvement over what they've been using in the past that they ended up replacing their entire inventory of these tools — about 15 of them made from machined aluminum. And they're not cheap tools. So they replaced 15 of these because the 3D printed version we delivered worked that much better.

This was due to a combination of factors: both the quality of the printer and materials as well as our ability to rapidly prototype effectively.

That’s an interesting example. How much of that project’s success would you attribute to the quality of the printers and materials, versus the ability to just run a lot of iterations efficiently?

Oh, I’d say 80% the quality of the print. 20% the ability to run many iterations.

Have you had success using Continuous Fiber Reinforcement (CFR) in your parts?

Using CFR helped us deliver on time when we didn’t have time to send a part to a machine shop. Going back to that inspection station I mentioned earlier: I remember this one keenly, it was a Thursday. We were supposed to deliver the whole thing the following Tuesday, and I realized that we weren't going to make it at that pace.

The inspection station had this structural bracket that needed to be really stiff, and we needed to make a design change to it. We didn't have time to get it machined out of aluminum again so we just printed this part, we stuffed it with reinforcing carbon fiber and printed 100% infill. It was awesome, it did exactly what we needed it to do and worked functionally. CFR was definitely one of the big reasons we were able to deliver on time.

What is your experience using our Eiger slicing software?

We love the software from a user perspective. The cloud based aspect of it is very useful, especially with remote work these days. Our engineers who are managing 3D prints love being able to start something remotely — which could be the difference between having a part that's ready to test the next day or not.

Distributed manufacturing also lets us successfully accommodate last-minute modifications for customers without any shipping delays. For example, we had delivered an electronics test fixture to one customer before they realized they needed to use a different mount camera than what we originally designed for. They needed a quick change before an upcoming customer meeting where they planned to showcase this fixture.

We made the design change on our end in a day or so, but we didn’t even have time to print it and ship it to them because they were out of state. So we just exported the updated part as an STL and sent it to them. The customer had a Markforged printer of their own, so they loaded the new STL into Eiger and printed it. The fit and everything worked on the first attempt — they were able to have it ready for their customer in time.

I’m sure customers still have a lot of apprehension about using 3D printed parts. Do you have any compelling example of a lightbulb moment when a customer finally bought into today's 3D printing?

Yeah, I think it's still a little bit of a challenge to convince people that these parts really are strong enough. We did a lifecycle test fixture for a weapon, basically a machine gun. This thing needed to automate the firing and reloading of a machine gun.

But due to the aggressive mechanical nature of this piece of equipment, our customers said they don’t want any 3D printed parts, as there was a lot of abrasive motion going on. Which makes sense — think about shooting a machine gun.

Well, when push came to shove, they needed to accelerate their timeline. We said the only way we’d be able to do that is by 3D printing certain parts. Initially they said no, no, we definitely can't do that — but otherwise there was no other way for us to meet their new deadline.

They finally relented, so we ended up getting some 3D printed parts in there. And you know what? They never came back and asked for replacements for those parts. Last I heard they were about half a million cycles into that lifecycle test.