How do you research effectively when beginning to prototype a solution? How do you know you’ve done enough research to troubleshoot a problem?

In this episode, we’ll chat with Erin, one of our R&D engineers, about her process of finding the best solution through rapid, iterative prototyping. 

 

Julia

Erin, thanks for being with us today. Tell me what you do here.

Erin

I’m a research and development engineer. I’ve been here a little over four years.

Julia

And you started as an intern, right?

Erin

Yeah, not counting the summer I started as an intern, I guess that would be four and a quarter years then.

Julia

Talk about that research part of research and development and prototyping, because I know that’s a lot of what you do here.

Erin

Yeah, definitely! As an R&D engineer a big part of my job is doing research, especially at the start of a project. We’ll come in, the client will have an idea for a concept but there’s a lot of research that has to be done to figure out what components you need, how to source those components, sourcing materials, and finding people that can produce the parts we need. I do a lot of research in my day to day job.

Britt

You’re about to do some research, where do you go first? Google?

Erin

Yeah, I mean honestly, most of the time Google. I try to start out with a broad search and cast a wide net. I’ll see what key words I see popping up a lot and use that to modify my search, and then I seek out experts, whether that’s someone at Blur or someone out in the field for their help. I also keep a log of every person I’ve contacted, what suggestions or notes come out of that conversation. That way I have a log of everything I’ve done and I can go back to that research, or if someone else at Blur is researching a similar part or concept I can go point them to that. 

Britt

It’s great that everyone seems to be creating these libraries that you can reference instead of recreating the wheel every time.

Julia

I think it shows just how collaborative we are too. The resources that we find or the processes we use, it’s not just about us and getting our work done, it’s about saying, “I did this and found it useful, let’s keep it so that someone else can use it because I know someone else is going to run into this problem at some point.”

Erin

Right. Someone else is going to need a camera, so let’s compile a list of all the cameras we’ve used so we don’t have to repeat that work on the next project. 

Julia

Yeah, definitely.

Britt

Do you ever have times where you’re having trouble finding data? Just altogether you can’t find any information, and how do you handle that?

Erin

Yeah, we definitely run into that more than we would like. I think that’s sort of where the testing and prototyping comes in. For instance, if we’re looking for a specific material property that we can’t find, can we get the material in and prototype something ourselves? Is there a simulation we can run with CAD software, find a similar material, and see if that gets us close enough? If we can have a simulation and a test that point towards the same answer, maybe that takes the place of whatever online answer we’re looking for. 

Julia

How often would you say we run simulations here to help in that research process?

Erin

I say we do it on almost every project, whether it’s a finite element analysis (FEA) or a flow simulation. I feel like I’ve done that with a lot of the projects I’ve worked on. We’re always 3D printing components of a device and testing those before we’re going to full scale production. 

Julia

What would you say in your research process is the most challenging part and what do you find the most rewarding?

Erin

I’d say the most challenging part of the research process is if it’s a really specialized component and there’s not a lot of people who make it or know a lot about it, you’re really relying on people to respond to your cold calls and emails. That can be challenging sometimes. I’d say the other big challenge is when a part has a lot of requirements, so it needs to be low cost, high lead time, really durable, juggling all of those different requirements can be a challenge. Or, if you’ve thought you found a new part that would work and a new requirement gets added, you have to go back to your spreadsheet that you’re keeping of all your options and reevaluate that. 

Britt

Yeah, I was going to ask how do you manage that with a client? They’re adding new requirements, how are you able to communicate with them that this new requirement means extra time or extra cost?

Erin

I really think you just have to be transparent with them and say, “We can switch to this part, that’s not a problem, but that may mean the lead time changes, the overall cost of the device changes.” If you’re switching from a smaller part to a bigger part, that housing that it lives in will probably have to change and that part now is going to get more expensive. Understanding that changing that one part can have a ripple effect is important to make sure the client understands as well.

Julia

I think that’s a super important part of presenting anything like that to a client. Having all of those details worked out so you don’t just say, “We can’t do that.” Instead say, “If we do that, here’s what’s going to happen” and then letting them make the decision. At the end of the day it’s their decision to make, and maybe they don’t mind that it’s going to cost more and they really want that specific part or to use that specific process. Definitely make sure you’re getting them all of the information they need to make that decision. 

Erin

Yeah, I agree.

Britt

When you’re first starting to prototype with a new client, what are you go-to initial questions to get the ball rolling?

Erin

When we kick off a design, the thing that I care about most is what function it needs to serve. I’m not as concerned with the overall size it needs to fit within, or if it has to meet all of these certifications. I’m most concerned with function at first. After we have an initial design, the next things I’ll look at are the material it should be made out of and the quantity we’re looking at. If we only need three for some demos we’ll probably 3D print it, if we need one thousand we’ll be injection molding it. And then past that, it’s good to look at usability. Does it need to have a handle, does it now have to fit inside this other part, what constraints does that mean? Also, being aware of color, does it need to be transparent, does it need to be waterproof, biocompatible, food safe, all of that. Those are all important to know, but they’re not the first things I worry about when starting a design.

Julia

How do you know or make a decision that , “Okay, I have done enough research to confidently go with the solution to the problem I’m experiencing in this design.”

Erin

It depends on the test. A test is targeting a specific function, whether it’s usability or waterproofing. Typically there’s a sequence of tests. I’ll start small, like if it’s how two parts interface I’ll just prototype and 3D print those two parts cropped to where that interface is. Once I have that working I’ll print the full-size parts, make sure that works, share it with the client, and have the users handle it and make sure they don’t have any complaints. Past that, what are the differences between the part you’ve prototyped and the production part? If one’s 3D printed and one’s injection molded, is there going to be any difference between them that would cause it not to seal properly, or be waterproof, and making sure you’re confident that despite those changes the design you have will still work. I think that’s when you can check that box that testing’s passed.

Julia

Getting prototypes in the hands of users is so important because, as we have experienced many times, they will give you feedback you haven’t even thought of. 

Erin

For sure, yeah. The earlier the better.

Britt

So you’ve been at Blur for four and a quarter years, what’s something you wish you would have known your first year that you could advise others?

Erin

I’d say just not being afraid to ask people questions. Everyone at Blur is so smart and so capable, so if you get stuck and you can’t solve a problem, don’t spend time just noodling away on it yourself. Go take a walk around the office and bounce some ideas around with people because someone is going to have an answer and it’s going to be a lot quicker than if you were just to sit there alone in your corner of the office.

Julia

Yeah, muddling through it on your own. 

Erin

Right.

Julia

What was the most intimidating part of our machine shop when you first started?

Erin

It’s not really part of the machine shop, but the casting-molding station was really intimidating. At first, just because there were so many components and steps to the process. I was so worried about messing things up. You have to measure stuff initially, make sure you have the A and B components at their correct ratios, degas it and pour it into the mold. Make sure there aren’t any bubbles and put it in this pressure pot; make sure that’s all sealed correctly, which always freaked me out. Then you just have to wait a couple hours until it finishes molding. You don’t know if you’ve failed or not until you take it out.

Julia

I’m sure there were times when you took it out and you were like, “Aww, dang.”

Erin

Yeah, like last month. I had a part that I thought didn’t need mold release, which helps it come out of the mold easily. I went to take it apart and it didn’t move. I was trying to use all these tools to pry it open and it was just chipping away at the tools. I had to just throw it away and start over, so now I use mold release

Julia

As you’re prototyping and testing, what parts of our labs do you use most frequently?

Erin

I use the 3D printer probably the most; right now it’s every single day and I’m kind of hogging it. The laser cutter I use fairly often, that’s really handy when you have larger assemblies that you have to build really fast. I work a lot with the UV cure adhesives, so I am using our UV light pens pretty often. Those allow you to bond things together really quickly, and we have a whole bunch of different UV adhesives for different materials. 

I don’t venture into the machine shop with all of the blades very often. When I first started I was using a circular saw and I didn’t realize there was a safety so I revved it; Jeff was standing next to me and I think I really freaked him out. He was like, “Yeah, I’ll do that for you” and I said, “I think that’s probably a good idea.”

One part that took a lot of testing and troubleshooting and iterating on was we had these two components that had to interface and seal in two different chambers. We wanted that to be done in a single connection so that the user just had one piece they had to connect. We looked at a couple different ways of doing that, either a press-fit, a snap-fit, a press-fit that’s just plastic on plastic or silicone on plastic. 

So, I did a lot of prototyping back in the lab. I would print those two parts on our 3D printers, and really just segments of the parts to speed up the prints and save us some money as well. I’d just print a bunch of different interfaces and see how well they sealed. 

The silicone on plastic interfaces I would overmold the plastic with one of our silicones we have back in the lab and test those. We narrowed it down to three different options, and we’re working on getting each of those full-size parts printed to send to the client. They’ll evaluate each of the three designs and see if there’s one they really like, or if they like all three we’ll bring them to the users and see if there’s a favorite among the users. If there’s not a favorite among the users, we’ll just pick the one that we like the best and will be the easiest to manufacture. That’s been a lot of iterating on and time with the 3D printers and molding station, which has been a lot of fun.

Julia

The benefit to having this in-house prototyping capability is we can come up with a lot of ideas for how to do things. You were just saying, you were just saying you made so many… how many prototypes do you think you made?

Erin

There’s like a hundred different pieces on my desk right now. 

Julia

Exactly, we have all of these different ideas for how these two parts can fit together, but by being able to do this in-house and fairly quickly we can throw out the ones that don’t work.

Erin

And each part is really only about an inch big since I’m just cropping it to that interface, so it’s really not wasting too much material or time to get those printed.

Julia

How long would you say it took you from the time you started 3D printing these up until you had the three core ideas?

Erin

When I am working on it, I’ll print a couple designs one afternoon, test them the next morning, figure out what I don’t like about each design, make those changes, and get them back on the printer that afternoon. So, the iteration cycle is pretty quick. 

Julia

Thanks for coming on Erin and chatting with us.

Erin

Yeah, thanks guys, thanks for having me.