Product Design

  • 06 Jul
    Process Drives Product Design Creativity, Part 3

    Process Drives Product Design Creativity, Part 3

    The ideation phase of product design is too often glorified. You’ve seen it — the beautiful sketches, the compelling imagery — the fake stuff. Don’t get me wrong, this is a fun and important stage, but a pretty sketch is meaningless if not supported by logical, critical, and thorough analysis. That kind of analysis — that’s what I do! For more on the work leading up to this point, please review part 1 and part 2 of this series. We ended part 2 with a definition of our product. Now we embark on a three-pronged process of ideation, design, and prototyping.

    We will design and prototype a planter consisting of a hard vessel using a removable filter (in place of rocks) and an integrated drain. The planter will use changes in mass to signal “time to water” with an indicator.


    Now we design a first draft of our product. This isn’t going to be the be all end all, instead, this is the model that will test the basic principles of our concept.

    First, determine the necessary components of the design and write it all down. When I get an idea down on paper, I almost always come away with something better than what was in my head. Here are some snippets from my notebook.

    Wait, this is basically a finished design, where did it come from? When did I decide to use a compression spring?

    Cutting Corners

    That’s right, I’m cutting some corners. I could excuse it by saying that this is just a prototype, blah, blah, but the truth is that I have a mechanical design more or less in my head that I want to implement. Once we selected our concepts from the morphological chart in part two, I knew what I was going to make. Sometimes this happens. It can be a problem.

    It is likely that there is a better solution than this one. I won’t uncover it because I’m skipping opportunities for design and analysis. On the flip side: we need to get this thing done! I want to test the basic principles and then use the thing and see if I like it. If I like it, then I can go back and explore the missed opportunities. If I don’t like it, or if I’m the only one, then this is good enough!

    What is important, though, is that we are intentional in our corner cutting. I want to be able to look back and quickly see those corners. For me that happens two ways: one, physically marking a relevant page and making a note in my notebook; and two, maintaining a list of assumptions and corners cut. OK, now that I have appropriately excused any shortcuts you may notice…let’s get back to it!

    The System

    While I’m making the sketches shown above I’m concurrently identifying, naming, and defining the parts of the system. Once I have a good grasp on what parts are involved I make a list of them and their necessary functions. Let’s do it here.


    • Contains filter
    • Allows plant to be easily removed
    • Fits tight against pot
    • Does not move in operation
    • Could be rigid or flexible


    • Fits tight against basket
    • Supports basket
    • Allows water to flow out through bottom
    • Engages with spring
    • Reveals indicator
    • Moves up and down with changes in mass
    • Can be removed entirely, but only intentionally


    • Retains spring
    • Acts as bearing surface for pot
    • Contains drained water
    • Has water feature

    At this point in the design questions come up. We know that the plant sits in the Basket/Pot assembly and the whole things moves up and down using the force of a spring. How much does it need to move? What exactly is the signal? How much will the mass change from water loss?

    The first question to answer is the latter: how much will the mass of the pot assembly change with moisture loss? To answer this, I use the internet. A quick search for something like “weight of wet potting soil” will quickly get you the information you need. I ran the numbers for 4″ and 6″ pots and found that for a 4″ pot I should plan to detect a change of somewhere between .3 and 1.5 pounds. That’s a big range, so I did a little testing.

    At the time I had four starts (basil, mint, lavender, and tomato) that had been sitting out for a couple of days. They were dry and needed water! I took their mass before and after watering, and I found that for a 2″ pot the change in mass between dry and fully saturated was between .15 and .41 pounds with an average of .3lb. Thus, for a 4″ pot (4X the volume) I should expect a change of about 1-2lb, average of about 1.2lb.

    Now that we know our change in mass: how precise do we need to be? How much travel do we need for our specified weight range? Can we find a spring that can do it?

    I started with a reasonable estimate of the amount of travel needed: somewhere between 1″ and 2.5″ of spring travel in the trigger range. That leads me to a spring constant of around 1 in/lb. Well, turns out a spring with that much travel and that light a spring constant is hard to find! McMaster had nothing, but I found a likely candidate on Century: spring 12420.

    Great, next: what is the signal? And what does this thing look like?

    Let’s start with the signal.

    I started this sketch thinking that I could have a signal band on the pot that moved up and down. The position of the band relative to the window on the base would be your indicator. Then I thought, what if the window were adjustable instead? Well, to do that, all you need to do is put a band on the outside and make the signal feature on the pot fixed. Good enough!

    Concept Sketches

    It’s very common to make an underlay as a sketch aid. Underlays provide a framework for your sketch, letting even poor sketchers like myself communicate a concept. To start, I modeled the system components roughly (as cylinders). Here’s a screenshot from Fusion 360 of the mock-up parts plus a couple 3D printed vases (one is self watering) that I imported for reference.

    I took screenshots of line-representations and threw them into a word document to make underlays. They look like this:

    The left image is the trigger point and the right is fully saturated. Now I have something that I can use to guide my concept sketches! Before we move ahead…be warned: sketching concepts is not one of my strengths. This is when I most often partner with an industrial designer or two. Here are some highlights from my sketches…

    This one, above — just no.

    Names are important: let’s call this one Hot Rod. Hot Rod really appeals to me on a mechanical design level. There’s the opportunity here to design a perfectly constrained mechanism! I modeled it quickly in CAD, though, and found it visually awkward with no obvious fix. Check out the models, below. I actually got the design all the way to “ready to print” but just couldn’t tolerate how it looked. Here’s a screenshot from my working environment (how I see it while I’m modeling) plus a quick render.

    Ugly, right? Let me know if you have any ideas about how to make this approach more attractive. I know there is a good solution! Make note: this would be the fastest way to test the concept. Rubber band, spring, two bic pens, and two printed parts.

    Next up, our winner!

    Let’s call this one the Maesprung Planter (Mason + Sprung). The Maesprung uses the same basic principles as Hot Rod, but does it in a way that is visually simplified. It’s also a little less mechanically compelling, but that’s OK. I quickly mocked this up in CAD to help visualize the concept.

    CAD Design

    We already have a solid start on the CAD design. I often work in CAD while I’m deciding between concepts. With the work we’ve done, finishing the 3D model is a fairly quick job. After about 4 hours of work I had parts ready to print. Let’s take a look at some screenshots.



    As I designed the model aspects and features of the product came and went. For example, I designed a coaster and a cap to be printed in the same elastomer as the band, but the cap was too much, and the coaster seemed awkward and was certainly not necessary to test the product. I’ve also abandoned the root basket part, choosing instead to pot the plant simply in the pot. We can set aside the root basket for a product enhancement down the road.

    Print, Prototype, and Plant!

    Here’s our planter! Next: print the three parts, add plants and dirt, water, and see what happens. I chose a much neglected spider plant to be our test subject. It was bone dry!

    I measured the weight of the plant/pot once potted (but dry) and again after watering. Dry: 1.45 pounds. Saturated: 1.96 pounds. So almost exactly a half pound difference between bone dry and saturated. It seems like we might want to water again at about 0.1-0.2lb heavier than dry. Here’s about how high that is:

    So we can see that we’ll want to have the band slammed all the way down on the base if we want to use the bottom of the pot as the trigger level. That’s lower than I was expecting. In the next version I may have a color change on the pot to serve as an indicator. Then I can have the band sit higher on the base.

    OK, we got this thing printed, planted and ready to go. But we forgot to ask an important question:

    What do we need to learn from our prototype?

    Here’s what I did with the prototype.


    By David Perry 3D Prints Product Design
  • 06 Jun
    Process Drives Product Design Creativity, Part 2

    Process Drives Product Design Creativity, Part 2

    In product design, brainstorming is too often a process that lacks all structure, and thus consistently generates mediocre solutions to a poorly defined problem. Brainstorming is useless without structure! Using the methods I’ll describe here, you can generate creative solutions for any product design challenge. Let’s continue with the design of a better tabletop planter. In Part 1 we defined the problem. If you haven’t read Part 1, go do it now. Our challenge is to:

    Design an indoor planter that assists the user in maintaining optimal moisture content.

    Then we spent time identifying and developing problem constraints, requirements, and objectives. We also took a close look at all of the variables and components of a planter system. Which variables have the most effect on soil moisture? Now we know, and we know exactly what we want our design to do. These are our design objectives.

    Now that we understand the problem it’s time to generate creative solutions. I’ll show you the process I use to “brainstorm” awesome things: the morphological method.

    This is about following a process in order to generate innovative solutions. Every corner cut along the way is a missed opportunity to generate a unique solution.

    We know what we want the design to do, but we don’t know how to do it. That’s actually exactly where we want to be! What will determine whether or not we meet the design objectives? How can we ensure that the user is happy with the product? The answer is simple: the function of the product. But what is the primary, overarching function of this product?

    Primary Function

    What is the function of the product we are designing? Don’t try to jump ahead to the answer here, even though it may appear straightforward. This is about following a process in order to generate an innovative solution. Every corner cut along the way is a missed opportunity to generate a unique solution. Back to it: we need to name the function; the primary, overarching function; the single function that captures what this thing does. Follow these steps:

    1. Identify the system and its inputs, outputs, and interfaces. Include all exchanges of energy, matter, force, and information.
    2. How will the customer know if the product works?
    3. Now identify the overall primary function.


    By “identify the system” I mean to map out and understand thoroughly the system in which our product operates and any systems within our product. Make a sketch with all of the interacting elements; all of the exchanges of matter, energy, force, or information; and you’ve got it. Here’s mine.

    From this diagram, we can identify our inputs, outputs, and interfaces.

    Inputs include: water, CO2, light, and the supporting force from the table

    Outputs include: water (liquid and vapor), O2, and information regarding dryness

    Interfaces are: between table and planter, between planter and soil, between soil and air, between plant and air (leaves), between plant and soil

    With the system identified, we now ask ourselves: how will our customer know that the product works? Well…it will work if the plant doesn’t die. The user will know if the design works by whether or not the plant thrives.

    <<Begin primary function generation>>

    We must design the system shown above such that the plant thrives.

    The system’s function is to support plant growth.

    Contain soil and root zone.

    Physically support the plant and allow exchange of water.

    Physically support root system and maintain acceptable conditions for growth.

    Physically support root system and exchange liquid water at optimal rates for plant health.

    I think I got it…

    The primary function of this planter is to physically support the plant’s root system and exchange liquid water at an optimal rate.

    We can break this into two primary functions (it’s OK to have two).

    • Physically support plant root system
    • Exchange water at optimal rate

    That’s it, we have two primary functions! Now we need concepts that perform these functions. To do that, first, we will break the functions into the simplest sub-functions possible. This process is called functional decomposition. From our sub-functions we will select a few critical sub-functions and use them to drive concept generation.

    Functional Decomposition

    Functional decomposition is an iterative process, so don’t worry about getting it just right the first time. If you find that you have other functions you want your product to serve, but they don’t fall under the primary function, list them separately as secondary functions. Here’s my first functional decomposition. After this I went digital.

    And after five iterations, my final.

    Now, select the most important and most basic sub-functions. Typically these will be the functions at the bottom of the tree. We will use these to generate creative mechanical design solutions.

    Morphological Chart

    Functional decomposition is the bones that support this product design process, but the morphological chart is the heart and soul. In the morphological chart you’ll generate concepts for each sub-function. You will generate concepts that you were not expecting. The trick is to generate concepts independently for each of your selected sub-functions. Each concept is written and roughly sketched in the morphological chart. You can spend 30 minutes generating concepts, or you can get a team together, use various techniques for concept generation, and spend a month. Either way, the process is the same, and it’s simple: make a chart with your functions at the top and ways to accomplish those functions below. Here’s a template as a pdf: Morphological Chart Template, and here’s my morphological chart:

    For each sub-function I generated a handful of concepts. This is the third chart I made, and I spent about an hour generating concepts.

    Now that we have all these concepts, it’s time to combine them to form several cohesive product design concepts. I usually aim to make three to five concepts out of my morphological chart. Here’s my chart of concepts:

    So we have these great concepts. All have pros and cons, and I have a sense of what I like, but I’d like to take my opinions out of the process as much as possible. To do this I use a points system to rank the concepts as they relate to the performance of the product.

    First generate a list of questions that assess all aspects of the product’s performance. Included in this list should be feasibility. Feasibility can be yes/no, or it can be assessed in degrees from 1-5 (not feasible to certainly feasible). My questions are:

    1. Is the design feasible to prototype in 48 hours using the available equipment? yes/no/maybe
    2. How reliably does the concept signal appropriately? 0 – 5 (completely unreliable to 100% reliability)
    3. How easy is it for the user to not notice or ignore the signal? 0 – 5 (very easy to very hard)
    4. How easy is it to adjust the dryness trigger? 0 – 5 (difficult to easy)
    5. How easy is setup? 0 – 5 (difficult to easy)
    6. How well does the concept encourage proper watering technique? 0 – 5 (not at all to very well)
    7. How likely is it that the concept will break or need repair? 0 – 5 (sure thing to certainly never)

    Now, let’s make just one more chart… Rank each concept A through E using the questions, above.

    In this example, I calculated a straight sum of the values given for each concept. For some product design challenges I calculate a weighted sum — applying greater importance to certain attributes or risks.

    We now have a straightforward ranking of our product concepts! If we cross-reference with feasibility, then we have our concept: Concept B.

    We will design and prototype a planter consisting of a hard vessel using a removable filter (in place of rocks) and an integrated drain. The planter will use changes in mass to signal “time to water” with an indicator.

    So…what’s that look like? In part 3 we will do a bit of ideation around what the planter could look like and how it could be put together. Then, and a little concurrently, we’ll head into CAD to design the planter, prototype it, and test. Go to Part 3.

    By David Perry Product Design
  • 16 May
    Process Drives Product Design Creativity, Part 1

    Process Drives Product Design Creativity, Part 1

    This August I’ll lead a group of high school teachers in Clackamas County through a 3-day seminar to help them teach students to design and be Makers. What’s the best way to do that? Teach the teachers to make stuff! First lesson: you can’t just say “Here’s a problem, go make a solution.” Instead, you teach a consistent process that can be applied to almost any challenge. I know that I will always find creative and effective solutions to a wide range of problems. Why? Because my design process ensures it. The process gives me something to lean on when I get a difficult project, and it will give students (and teachers) the courage to go be Makers.

    I know that I will always find creative and effective solutions to a wide range of problems. Why? Because my design process ensures it.

    I’m working through a design of my own in preparation for the seminar, and I’ll be documenting the process here on the blog. I’m breaking this accelerated design into three chunks:

    1. Problem definition: requirements, constraints, and design objectives
    2. Concept generation and selection: functional decomposition and morphological charts
    3. Design and test: paper to computer to prototype

    I’ll tackle each chunk in a separate post, and they’ll be released weekly.

    Part 1: Problem Definition

    Each challenge is presented as a new client showing a “design firm” a potential project. With only this written statement from the client the teacher teams must develop design objectives, requirements, and constraints. Remember: the best solution may not be what the client has in mind! Here’s the challenge I pose myself, spoken from my client’s perspective.

    “Here’s the problem, and lots of people have it — I have a hard time keeping my indoor plants alive, I just can’t remember when I watered them. Did I do it recently? Did I just casually dump the rest of my water glass in, or did I give it a good soak? If I water them now will they be too wet? I can’t tell by looking at them and I don’t want to get all dirty sticking my finger in and also ruining my arrangement with a bunch of holes! So instead I just water whenever and hope I get lucky…wait, is that mold?”

    “I keep these plants on my desk. I tried a small self-watering planter. Great idea, really, and they can work great, but I bumped it the other day and sloshed dirty water all over my sketchpad! Besides, I can’t use those for my succulents, here, because they need to dry out almost completely. So there’s gotta be a way to do this better. Here’s what I’m thinking: an internet enabled pot that tracks moisture content!”

    “Look here, I’ve put together a little something. The pot for the plant has this sensor built in, right here, where it detects the soil moisture. Then, that runs into this little computer taped over here and the data is sent through this ethernet cable and into the cloud. Then you log in on your phone and voila! You could have an app that says, ‘hey, water your Jade tree.’ Great, right?”

    “It’s working great but I can’t seem to get the sensor to read very reliably, oh and I fried this board the last time I watered. So we need a water-tight and nice pretty way to hide all these cables and electronics. I need your help to make this work.”

     This is a problem I have all the time, so I’m thrilled to get to solve it! Now, I’m not sure that the electrified-sensor-internet approach is the right one. I’d prefer something simpler, something that’s purely visible and tactile. I think that’s what most people would want, so I’m going to back up, frame the problem, and come back to my client with three compelling solutions.

    Regularly and intentionally reviewing and revising your constraints, requirements, and objectives is an important part of the design process.

    Defining the problem means developing three things: constraints, requirements, and design objectives.

    • Constraints are somewhat firm limits imposed by physics, manufacturing methods, available resources, or design decisions you already made.
    • Requirements are features or functions that the product must have or perform to be acceptable.
    • Design objectives (the most fun) are the goals of your design, which you will aim to accomplish as best you can given the constraints and requirements.

    These three crucial components of any product design will change and evolve as the design develops. Regularly and intentionally reviewing and revising your constraints, requirements, and objectives is an important part of the design process.

    So how do you start? For a major project with sufficient budget I employ a technique known as QFD, or Quality Function Deployment. For this, we’re just going to read over the problem statement, use our brains, and write stuff down.

    First, write down the problem as you understand it. Write it again, and try to break it into more measurable language.

    Help user keep plant alive.

    Remind user to water plant.

    Design an indoor planter that assists the user in maintaining optimal moisture content.

    Great! That’s the goal. What is required of the finished product? These are our requirements.


    • Product shall not leak if bumped during regular use.
    • Product shall allow re-planting.
    • Product shall be waterproof, but not submersion proof.
    • Product shall be manufacturable at a reasonable cost.
    • Product shall accommodate a variety of plants: from garden herbs to succulents.


    • Product must not be larger than 150mm in any direction. (about the size of a tissue box).
    • Product must be prototype-able using the tools at our disposal.
    • Product must not require any special tools or consumables.

    Design Objectives

    In order to decide on design objectives, we need to better understand and define the system. In particular, what are all the things that effect the soil moisture content of an indoor plant? Write down all the variables you can think of. Ask other people, search the internet, and write it all down. I like to make a table of variables. Then I rank each variable for various categories. For this design, each variable gets three ratings from 1-5. Those ratings correspond to these questions.

    A. How much influence does this variable have on moisture content? 1 (little) to 5 (lots of influence)

    B. How much difficulty does this variable cause for the user? 1 (user doesn’t think about it) to 5 (constantly nagging)

    C. How easy is it to influence this variable with the container (our design)? 1 (practically impossible) to 5 (very easy)

    Next, combine those rankings to get a single value. I have chosen to weight the final value as: A*(B+C). This gives more weight to the degree of influence of any single variable. Now — none of this is set in stone! Use your common sense. Are you going to design a container that controls plant respiration rates? No, because you can’t.

    Here’s my variable selection matrix. Pardon the informality (and slanted lines), you’re seeing the inner workings here!

    Using this matrix and my sense, I rank the top five variables and eliminate others. Now I know what I should focus on with my design!

    Now I carefully analyze the selected variables to understand how they work and what they do. It’s a thorough but informal analysis; typically a combination of rough sketches and text that more resembles conversation than engineering analysis. Here are some sketches from this process:


    After careful thought, here are my design objectives:

    Product shall signal or prompt user at a certain soil moisture level.
    Product shall encourage and reinforce proper watering technique.
    Product shall maximize the ratio of volume to surface area, within reason.
    Product shall include features such that the user need only add plant and soil.

    OK, let’s take a break!  We know what our design must do, we know what constraints we need to work with, and we have a list of objectives that we’re designing for. Next, we’ll dive deep into the product’s function with a process called functional decomposition. Then, using the functions produced by decomposition, we’ll start generating concepts with the use of a morphological chart. Then we’ll combine concepts to make more concepts and analyze them all with another selection matrix! You don’t want to miss it. Go to Part 2.

    By David Perry Product Design
  • 22 Oct
    The F-F-Fiddle Means You Can Make Anything; Here’s How

    The F-F-Fiddle Means You Can Make Anything; Here’s How

    You don’t need to be a master craftsman or an industry expert to make something new and exciting. If you are willing to try and fail (repeatedly): you can make anything! Before I made the F-F-Fiddle, I’d never made a musical instrument. I started playing the violin when I was in grade school, and I’ve always wanted to make a violin, but I thought I’d wait until retirement — the barriers to entry for that kind of craftsmanship are so high. Then, in early 2013, I bought a 3D printer. Suddenly I had this robot that could make complex, accurate parts that I modeled on the computer. It could make anything! All I needed to do, then, to make a violin, was to design and model it on the computer and print it out. Shoot, I can do that all day! Let’s walk through the process I used to make the F-F-Fiddle: Research, Ideation, Design, Prototyping, Iteration. This is the same process that you can use to make, well — anything!

    1. Research

    All of my projects start with some research. For the F-F-Fiddle, I had to learn about electric violins. What do electric violins look like? What are the important components? I pinned violins that I liked to a Pinterest board to collect inspiration.


    I also took apart an old (junk) electric violin that I got for $20 from a friend. Now, I love destroying things, but taking things apart is seriously important — it’s the best way to learn how your parts need to go together. I re-used the components from the junk violin for my first F-F-fiddle.

    You gotta break it to make it.

    You gotta break it to make it.


    2. Ideation

    Armed with enough info about electric violins to be dangerous, I began the ideation phase. I modeled the necessary components (that I took out of the junk violin) in CAD and printed underlays — or outlines of the required parts that I could then sketch on. My sketching skills are poor, but regardless of skill level, the process of drawing is vitally important! I think it wakes up parts of your brain that (for me) typically lay dormant.

    I brought in my friend and excellent industrial designer, Dan Nicholson, during this ideation phase. Dan helped me refine my design intent and was able to beautifully communicate it with pen and paper. By working side-by-side, Dan and I developed the same vision for the F-F-Fiddle. Even though his sketches were relatively rough, we both knew exactly what we wanted to see in the instrument.


    3. Design

    With Dan’s sketches in hand, and a clear vision for the F-F-Fiddle, it was time to make the thing real. This is where the design goes digital. During the computer design phase, I have to make decisions to balance my design intent (my design vision) with the realities of violin ergonomics and my manufacturing method — 3D printing. For an design engineer like myself, these hard decisions mark the beginning of the real challenge (and fun). This process took a couple of weekends, about 30-40 design hours.


    4. Prototyping

    I printed out the first F-F-Fiddle as quickly as possible — there is no better way to learn than by making! Sure enough, it didn’t work — the strings were way too far from the fingerboard, the bridge too flat, and the electronics were terrible. What this first one did do was validate the concept — it help the string tension, and it made noise. I knew at this point that it was possible to 3D print a violin!


    5. Iteration

    The rest of the process is iteration: perform a redesign based on feedback from the prototype, then make another prototype — repeat. My first order of business was to make a playable instrument, then to make it more durable, and then to continue with improvements for playability and ease of assembly. After another 30-40 hours of work I got to the point where the parts printed reliably and the violin felt more-or-less like a real violin — it was ready to be released to the wild!


    A lot of these steps may be familiar to you if you’ve made stuff before. The process used to make things is at the same time both universal and unchanging while it’s also being transformed by advances in digital design tools and 3D printing. No matter what you’re making or how, what’s important is that you work the steps — Research, Ideation, Design, Prototyping, Iteration. Each step is crucial to a successful project, but the real magic happens when they’re combined.


    By David Perry 3D Prints Product Design
  • 02 Apr
    3D Printing Biomimicry Leads to Righteous Ripping

    3D Printing Biomimicry Leads to Righteous Ripping

    Where the soft calls of ocean mammals meet carefully extruded polycarbonate plastic,  high-performing and beautiful surfboard fins are born. Boardshaper Roy Stuart has combined two things that will save the world: biomimicry and 3D printing.

    [I wrote this blog for product design blog, Solidsmack. It’s re-posted here with permission.]

    Most surfers these days ride with three fin, or ‘thruster,’ setups. Thruster fins offer a good balance of performance and stability. Some use two fins, which is less stable and more maneuverable and typically used only in small water. Beginners, long boarders, and folks that like retro equipment use single fin setups. Offering much greater stability, single fins also limit performance significantly. To get a sense of the difference between thruster and single-fin performance, imagine a ‘70s surf video of riders smoothly flowing along a wave, and then contrast that with modern surfers shredding the heck out of it. That difference is due, in part, to the better performance of thruster fins.

    Roy Stuart’s Duke board is a beautiful $369,000 take on classic single-fin style boards.

    From what I read on the internet, Roy Stuart’s focus is on the basic essence of surfing — catching waves. He likes to do it with large wooden boards that are elegantly crafted over long periods of time (years). To improve the performance of his single-finned masterpieces (as well as others’), Roy optimized the flow across the fin by adding a series of bumps, known as tubercles, to the leading edge of the fin. This forces water into faster currents which in turn increases performance. The use of leading edge tubercles (bumps) for improved performance is not new, Humpbacked whales have used them for millennia!

    A humpback fin, photo taken by MrMoorey on Flickr.

    Humans have long been inspired by nature, but in this day of pocket screens and wearable computers it’s easy to overlook the natural solutions that surround us every day. Despite this, there is a rapidly growing trend towards biomimicry — the use of the natural world as inspiration for solutions to human problems. Bio-inspired solutions often result in astounding efficiency gains, and fluid flow is a common application. If you’re interested in learning more about biomimicry, check out Ask Nature as well as Jay Harman’s recent book, The Shark’s Paintbrush.

    Biologically inspired designs often require complex geometry that can be difficult to make with modern production methods. 3D printing offers a manufacturing technique that is similar to many of those found in nature — like a sea creature building a shell — and presents a new set of design constraints that encourages organic shapes and experimentation. For his fin, Roy considered a variety of manufacturing methods, but found them all either too difficult (machining a tall skinny thing causes deflection) or too expensive (tooling costs). He ended up partnering with a local design and 3D printing firm — Palmer Design and Manufacturing, for the development of the fin.

    I contacted Palmer Design to get some information on the project, and was very pleased when Andrew Palmer responded with insight into their design process.

    Yes, it is this easy. From Palmer Design and Manufacturing.

    Palmer designed the fin in Solidworks, as you can see from the process shot, above. While Andrew declined to comment on their specific slicing software and printer setup, he did imply that they use a secondary piece of software that allows them to specifically control internal part geometry. By controlling infill geometry and using polycarbonate — a very strong printing material — the folks at Palmer were able to overcome their biggest challenge: fin strength.

    The majority of the fin looks like it’s printed with rectilinear infill — often considered the strongest choice. 

    The fin is printed vertically with an FDM process. This build orientation means that as the fin flexes against it’s broad faces it is pulling the deposited layers apart. The bonding strength between layers is often the weakest link in a FDM part, but for part resolution and printing ease, it doesn’t make sense to print the fin in any other orientation. The strength Palmer has achieved here is impressive, and you can watch them break a fin.

    Roy’s original fin, right, blows my mind with beauty. Picture from Roy’s website.

    There are presently 17 versions of the fin. Picture from Roy’s website.

    Each fin takes 2-6 hours to print. Using 3D printing, there’s no penalty to making a greater number of unique parts, and at present they have about 17 different versions. (3D printing will save the world.)

    By combining his knowledge of surfing and board construction with biologically inspired aerodynamics and additive manufacturing techniques, Roy Stuart is now able to sell performance enhancing and affordable surfboard fins. For those of us that work in product design and manufacturing, biomimicry and 3D printing will continue to produce better products, more efficient processes, and game changing (anti-gravity) manufacturing methods. Considering the global problems we face, it’s important that we understand, embrace, and contribute to these changes.

    Title photo of Roy Stuart by Alan Gibson from the New Zealand Herald, copied from Matrix Surfboards.

    By David Perry 3D Prints Product Design
  • 05 Sep
    How to Design for 3D Printing

    How to Design for 3D Printing

    3D printing is the fastest way to go from CAD model to physical part. 3D printing also allows the designer to break many of the rules that apply to subtractive manufacturing techniques like CNC machining. With these powers combined…3D printing is awesome! But it’s not a free ride, to make great products you have to learn how to design for 3D printing.

    3D printers, like any manufacturing method, have limitations. Inexpensive machines (<$3000) are generally all Fused Filament Fabrication (FFF) types, so that’s what we’ll focus on for now. These machines construct 3-dimensional parts by building them layer-by-layer with extruded plastic. Generally these printers print one layer at a time on a fixed X-Y plane. After one layer is complete, the build platform moves down by one layer thickness (or the print head moves up), and another layer is deposited.

    So, how do you design for 3D printing? Easy — learn these ‘rules’ that I use when I design 3D printed parts.

    Keep in mind, rules are only guidelines; please stretch and break these as much as you can to learn about your machine and its capabilities.

        1. Design your part to have a flat surface. While it’s not critical, it’s very helpful to have a nice flat surface to select as your first layer. You can have your slicer fake it with support material, but support material is a pain to clean off and can mess up your beautiful layered surface finish.
        2. Keep under-hanging angles near or below 45°. If you keep your underhangs at or below 45°, you won’t need support material! They’ll all come out looking very nice and you’re more likely to have successful prints on the first try.
        3. Keep bridges to about 30mm or less. Bridging–the act of trailing filament between two supports (forming a bridge)–is fun. As long as you keep them short, it’s easy, too. If you’re planning a bridge on a large part, try to test it out first to make sure your part won’t fail. Nothing’s worse than having a print fail towards the end of a large part!  EDIT: I just bridged 80mm. It took a couple layers to firm up but I was blown away that it worked as well as it did. It just goes to show…BEND THE RULES!
        4. Design your own support material. You can minimize the supports needed and the effort required to remove them by designing them yourself. Obey rules 2 and 3 and keep the supports small enough to easily trim with a pair of snips.
        5. Use the X-Y gantry for complex geometry. The X-Y gantry moves with very good resolution. The Z: not so much. I think the topographic look formed by the layers of material is really cool, but if you have complex geometry that you want to come out just right, try to use the X-Y gantry as much as possible. Common examples include holes, text, and logos.
        6. Load along the layers. If you are designing a part that sees some load, design it in such a way that the loading does not try to pull apart the layers. Any applied forces and bending should be along the layers. Think about it like this: if you only printed half of your part, do you still have the necessary geometry to support your loading scheme?

    Now you know it — how to design for 3D printing. Go make stuff! Keep these rules in mind, but be sure to stretch them.

    By David Perry 3D Prints Product Design