There are many techniques to prototype the products you want to create in the future. I have had the chance to work with some of the most talented people in the San Francisco Bay Area. Through their inspiration, I wanted to share some of the tricks of the trade making something from nothing. Here are their and my favorite ways to prototype:
Note: this post is lengthy. If you are particularly interested in a certain technique see the list of links below:
- Thinking outside the Box
- Industrial design with Renshape Foam
- Subtractive machining
- 3D Printing
- Laser cutting
- Silicone Molds
- Moldable Plastics
Let’s get physical. Not the way your thinking though.. 😜 Prototyping tangible things is at the core of any good consumer (or otherwise) product. If the physical portion of a product does not work then it completely kills the user experience causing your product to go into the trash bin. So, before that ever happens you should test your product in as many ways and contexts as possible. One of the main ways to do so is prototype it to ensure the end user uses it as you intended. There are many ways you can do this so I will go into detail about some of the top ones below.
Thinking outside the box
One of the folks I had the great opportunity to work with at Sifteo was Dave. He was one of the founders and the main reasons why I was there in the first place! He had some interesting points about the methods on prototyping electronics before getting too crazy with coding, circuit board board building and 3D printing. Enter Dave:
- Repurposing something that is close to what I am thinking about, rather than starting from scratch.
- First figuring out a way to try out an idea without writing any code (e.g. acting it out with props and another person, putting a cell phone inside a cardboard box to make it look like a new device, etc)
- Time-boxing the activity: What can I put together in 15 minutes?
Especially as an engineer, these few principles will save time and frustration down the road and prevent you from burning out of your initial idea inertia. It’s too easy to go down a rabbit hole of an problem only to realize that it’s not what you need. The worst part is that you spent a whole month putting it together! Sunk costs are brutal and it’s a topic for another blog post all together. I digress, avoid this at all costs and take Dave’s advice.
Industrial Design with RenShape Foam
One of my all time favorite people is my buddy Greg. He was our industrial designer at the last startup I worked at. He has many tricks up his sleeve but one of them included the use of RenShape Foam. I remember him toiling away in the garage with little foam pieces scattered about. It didn’t seem fun to me but then again i’m not an industrial designer 😎. Here’s how he says it’s best used:
When 3D printing is unaffordable, and for early in the process form and scale studies, RenShape foam is irreplaceable. It can be easily cut with a saw, shaped with multiple sizes of files and smoothed out with sand paper. It can also be hot glued with other pieces of RenShape and then re sanded. Finally, it can be smoothed with automotive bondo and then painted for a high res, although delicate, beauty model.
As you can see it’s versatile and cheap. I’m a fan of Automotive Bondo in limited cases but it requires an advanced ventilator and it’s very messy to work with. If you ever get it on your clothes (which you will) it won’t come out, ever. Here’s a picture of the ventilator I used when working on my project way back:
You can find something like this at Lowes, Home Depot, or Amazon. Always use protection working with less than human friendly materials. Your braincells will thank you later.
As an aside, back in my young days, I got so excited that I ended up making a giant soap costume out of Bondo and cardboard. It’s crazy what the things you can make with the limited resources you have
Additionally to Renshape, I always thought clay was a viable option. I have used it sparingly but more like a moldable plastic more than anything (see a few sections down) Greg also goes into the details of why it’s better than using clay.
I do prefer renshape to clay cause it can shaped very quickly and holds its form. So you can do 5-10 ergonomic studies of say a screw driver grip handle in 1 hour, take those studies to review with the team and then iterate on 2-3 shapes again with small files at a higher fidelity (include knurled thumb pads or tpe over mold using bondo). Plus they won’t slump overtime or become more brittle like Sculpey will - as long as the RenShape is kept out of UV light which will discolor it and turn the exterior into dust as it breaks down.
Looking for more? Here is one particularly good article I found on Hackaday about Renshape.
Subtractive machining involves purchasing a block of raw material that is then milled down to fit the specifications of the part you submitted to the workshop. This technique has been around the block once or twice before 3D printing really took hold.
Here is an example of the process:
You can take a block of ABS that you purchased from McMaster Carr, bring it to a machine shop and have them cut it apart using the step files that you provided from your CAD work. In some cases, because the material is very uniform, it’s more ideal for testing reliability and function. Also, when there are more delicate features that protrude, a solid block of plastic will always be stronger than a 3D printed version. Additionally, 3D prints are generally hollow with a lattice structure on the inside to give the part some strength versus, as mentioned before, a solid brick of plastic.
I definitely recommend going into the 3D printing realm before ever getting something cut by a machine shop. I will go into more detail about 3D printing in the next section.
Here are some players in this space. You can check out some their design tips and suggestions to get your prototype in your hands that much faster.
- Plethora for making 2.5D metal things quickly (Thanks Dave)
- Cycle Start (Local Bay Area)
I remember when the rep-rap came out and it blew everyone’s mind. Now, 3D printing is just as ubiquitous and using a sharpie marker. Machines are everywhere and it’s relatively cheap to make the things you’re dreaming about a reality.
During my time as a product engineer, I have also had the chance work with another talented industrial designer out of Canada. Brett Molnar has excellent design sense and he had a bunch to say about some of his favorite prototyping techniques. Specifically, he is a major fan of using the SLA type 3D printing techniques and he specifically recommends the Formlabs Form 2. An aside: You can check out Brett’s website for more information about what he’s been working on these days.
In regards to prototyping my go to tool as an industrial designer is definitely a 3D printer, more specifically the Formlabs Form 2 as it offers such a range of materials and very high quality surface finish. As designers and makers we are very fortunate to have such amazing technology available to us as a tool for testing, prototyping and even creating realistic painted and finished models that can be used to sell new ideas to customers, investors and to validate new designs. It cuts the lead time of projects down by being able to test my ideas in house without having to source them or wait for a manufacturer to prototype it. It has always been a part of my design philosophy that anything can work on paper but it is not until you physically begin to build the idea in true 3D form that it comes to life. I still rely heavily on making quick mockups of initial ideas out of foam or other readily available materials and then move into 3D CAD to eventually 3D print a model.»
As you can tell, not only 3D printing can be a useful tool but it can directly correlate with getting someone to sign on with your project whether it be a customer or investor. I’ve personally used 3D printing for all types of work including printing out full models of a charging adapter for Pryme Vessyl and placing the electronics inside so we could charge our test units. If we didn’t have a 3D printer we would have been at the whim of our manufacturer who didn’t deliver the first-shot prototypes until about 3 weeks later.
3D printing has also evolved a bunch over the past 10 years. You now have the ability to print many other materials other than plastic. One service based in the Bay Area called Fathom can print a variety of materials on their machines. These machines work a little different than the SLA or additive types (like a MakerBot) but in a good way. They’re able to make extremely fine detailed models that other processes just can’t touch.
Here are some of my favorite resources for 3D printing:
One of my all time favorite mechanical prototyping tools is laser cutting. I was first introduced to Ponoko back around 2011 and realized it was a fantastic resource for creating custom wood and plastic cutouts that could be used for products or even gifts for my family.
I particularly find it useful to use a program like Sketchup to refine what the outline of a particular piece may look like and then export it to a vector image which is then used by Ponoko to make the necessary cuts. I recently made a full prototype of an electronic chessboard using laser cut parts. Here is a snapshot of what it looked like:
The coolest thing is that I was able to make it almost to scale with the features that my industrial designer and I had dreamed up. It was a perfect proof of concept if there was ever one.
The process for creating a design using laser cut parts is somewhat straight forward. It requires working in some type of 3D design tool. I personally have used Sketchup with some success but it’s limited and requires a few steps to generate the correct file formats to be imported into a vector graphics program like InkScape.
I first learned about Ponoko through the maker community and I haven’t looked back since. Here are some of the most useful links that I have found that teaches about how to make your designs and get them cut by the friendly folks at Ponoko:
There are two ways to think about silicone. The first way is that you use silicone as a mold to cast hard plastics. The other way to think about it is to use hard plastics to cast silicone! I’ll dive into casting silicone in hard plastics first.
During the prototyping phase of yet another defunct project of mine, we decided to 3D print out a 3 piece mold that would be used to cast an egg shaped object with a hollow middle. We printed everything using the services at Fictiv. I’m actually very happy with the results as you can see below:
To modify the overall thickness of the silicone we ended up using wax that we could dip the center piece over and over until the right thickness was met. In the end we had about 4-5 samples of a moderately translucent silicone model!
Alternatively, there are ways to create silicone molds which then can be used to cast solid materials. This way you can create extremely detailed prototypes which are also quite robust and can stand the test of time. These prototypes are the ideal prototypes, besides using ones that are injected molded, which can be used for reliability testing and user testing.
There are a few processes of creating a mold. One way you can actually 3D print the negative of a silicone mold and then cast silicone around it. Once cured you can remove it
Casting silicone can be messy but can produce some excellent results. It involves submersing a model (3d printed or otherwise) into RTV silicone. That mold then can be cut in half and reused to cast new parts in the place of that original prototype. This is a very basic way of creating batch prototypes for a product where you can meld the electronics with the mechanical design.
Sometimes you need more moldable material that conforms to a surface or provides protection in a certain area. I’m not sure if there is an official term about this type of moldable plastic but i’m going to call it just that for the purposes of this article.
The nice thing about this plastic is that it’s multipurpose. It can be used to fix your furniture just as much as it is useful for prototyping a feature on your product. With that being said, that’s all it’s really good for. You won’t be able to cast a full full product in this stuff but it’s useful for small bits of detail.
Here are few types that I can think of that are quite popular these days:
When prototyping a previous product, I had been introduced to a special kind of plastic called Instamorph that, when warmed in a boiling water, can be shaped or formed. Then, when cooled maintains that shape as if it was hard plastic. The downside of this stuff is that it’s hard to work with and mold into a particular shape that you want.
There are many ways to use it but as I mentioned before, expect the final product to be crude and not polished.
Another option is Sugru which is also another great little tool to create rubber like finishes. The difference here is that it is a chemical cure plastic vs being mailable when warmed. So, with that in mind you only have a limited amount of time before it becomes hard and you’re stuck with a chunky blob of Sugru. I have personally used this stuff to fix things more than prototype. For instance, I have used it to fix my motorcycle turn signals countless times as they are the first thing to go when I go toppling over on the motorcycle. Sugru comes in many colors and, if not anything else, is a great tool to keep in the toolbox.
Websites and Apps
There are many ways that I’ve seen apps prototyped. When working at my last startup, our design team used used sticky notes, to define the flow and human interaction within our application. (Talk about low tech!) I have also seen digital mock ups of the same thing. They contain only the basic amount of information needed to convey the state of the application and what’s possibly going to happen next. This process is called wire-framing and its a summation of every single screen or interaction that you may see during the use of an application. Having easy apps to use has been a high priority for many businesses and having a User Experience/User Interface designer is critical to meeting that goal.
One of the first people I ever met who did interface design was my friend Emily. She was the UI/UX designer and had the responsibility of determining the direction how things looked and worked for the end customer. She is extremely versed in her field and has even taught some General Assembly courses on the matter. For more information about her, I highly recommend you check her website here. She weighs in on one of her favorite tools InVision. Enter Emily:
For apps or websites, I recommend using InVision. It allows you to make an interactive prototype at whichever stage of fidelity your design is, whether it’s rough wireframes or pixel-perfect mockups. The ability to have an interactive prototype is key for user testing and understanding how to iterate and improve on a design. InVision also allows other people to leave comments on screens, which facilitates and consolidates feedback from the team.
Emily concentrates her efforts on application usability and informs her design decisions with tons of user testing. One of her main concerns is to ensure an interface works and works the first time around. She also helped design a flow for an app that I was working on for a bit. Here’s an example screenshot below:
Electronics are their own beast and require a whole different set of knowledge compared to simple physical things. The best way to play around in this area is learn a little about electrical engineering and start making boards yourself. I have no formal training in PCB design. Let me repeat that, I have no formal training in PCB design. All my skills have been learned on the job working at companies like Apple or through tutorials on the internet. A great site to learn more is Instructables they have almost everything you can think of.
I made this discovery on Texas Instruments’s website a while back. It’s a development kit for different size IC’s. It provides 10+ tiny printed circuit boards with 0.1» pitch headers for each pin which then can be placed on to a breadboard or plug right into a devboard for testing. The only catch is that the integrated circuit has to be soldered to that board. This is by far my favorite way to deal with small IC’s these days and I highly recommend if you need to prototype small analog circuits go this way before creating a custom board.
You would be surprised how much you can get done with a development board while developing your product. Since the advent of Bluetooth Low Energy, development boards have become the example everyone copies from for lots of the IoT products.
At Sifteo, we actually used a development board coupled with a daughter card that I had designed and hand assembled by hand to get our first prototype working. (The boards originally had no solder mask or silkscreen, quite a pain to assemble.. lesson learned, never again) My amazingly talented co-worker at the time documented the work that went into that prototype at the time. The article can be found at Adafruit’s website.
Additionally, you can combine developoment boards and bread boards to prototype analog circuits. By using the resources I mentioned above, you can not only write some basic firmware but you can test the functionality of your hardware with your firmware before spinning a board. I highly recommend this approach as it’s a great opportunity to test the most risky portion of a system before moving to a full scale prototype.
There are lots of options out there to choose from. One of my good friends and collegues from Sifteo still stay in touch from time to time. Cedric thrives in the make things cheap and fast world. He comes from a scrapy hacker background which is sometimes useful especially when you’re on a budget. Enter Cedric:
Single board computers such as the Raspberry Pi have become insanely cheap these years. (Bunnie has an interesting theory about Moore’s law end. (See his article on Wired)) Apparently, we won’t see much more powerful computers, but damn cheap CPUs, yes!
Just recently the folks who came out with the Raspberry Pi released a $10 version that has wifi and a smaller footprint. When considering a platform for a production product, I highly recommend use something that is production ready. Modules are almost always a bad deal unless your Apple. The only situation I can think of, when it comes to modules, is that when you’re short on time they’re a good choice to get a product out the door and subsequently cost down to recoup some of the losses by using a module. Remember, a module can be anything: bluetooth, wifi, zigbee, etc.
Custom Printed Circuit Boards
If you want to bring your project to production, it will likely require a custom circuit board which fits within the product that you’re creating. In all of my experience, most companies will have their engineers farm out the layout work to someone else. Whether that to be a team member at the same company, an expert at another company or a freelancer. I have had some great luck finding someone on [Upwork]() to do some of the work for me.
But first, you need to create the schematic that will be used to create the circuit board layout file. Bear in mind that you can’t have one without the other (in most cases)!
I have tried all the most major CAD packages out there and they all have their plusses and minuses. More recently, i’ve been sticking to OrCAD as it’s still the industry leading product compared to the others. There is also EagleCAD which is a favorite among the hacker community. Additionally there is also Altium which is also a favorite among professionals at small to mid-sized companies.
There is way too much information to go into on this post but here are some good getting started guides with EagleCAD which, by far, has the lowest bar of entry.
Sparkfun was the place where I spent lot’s of time learning my skills back in the day. They are still a great place to learn and plus they have a completely free parts library which saves time and headaches!
Going to fabrication
Once you have a design in place it’s time to build it! For blank printed circuit boards nothing beats OSH Park right now. Unless, you have a giant board like I had spun around this time last year. It’s too cost prohibitive with OSH Park. I ended up going with 4PCB of which i’ve been a long time customer.
If you wnat to assemble yourself and need a stencil, I highly recommend OSH Stencils. But, if you’re boards have lots of parts, I highly recommend you get it done by a machine. You’re eyes and patience will thank you. Speaking of, if you’re using Eagle to design your projects Circuit Hub is the place to go. They will fab and assemble your circuit boards for reasonable prices.
This is just the tip of the iceberg when it comes to prototyping products. There are many ways to get the same task done but it may take more time or may be way more messy. (Avoid the bondo!)
I hope that you’ve found this post valuable. If you have questions about a particular technique or want to learn more I want to hear from you because a) it’s awesome to hear from my readers and b) it will help tailor my content for you in the future! Leave a comment here or fill out my survey and I’ll be sure to give you a shout when it comes time to publish my next article.