While reading Arduino and DIY hardware forums, I noticed that quite a few people have questions about moving from an Arduino prototype to a (selling) product. Since I have done this a few times and have learned a lot during the years, I decided to share what I think is the best (quickest, cheapest and lowest risk) method that I have used. In addition, I will try to address some of the choices that come with taking a product into (pilot) production.
- Built prototype using Arduino and breadboard [few days]
- Draw schematic, design printed circuit board [PCB] with CAD program [e.g. Eagle] [1 day]
- Order protoboards [wait 2-4 weeks]
- Solder protoboard, test and adapt if needed [1hr to several days]
- Get quotes for manufacturing based on Bill-of-Materials and Gerber design files [1 week]
- Select manufacturer and order (pilot) series [few weeks]
Things to consider:
If you are going to make a (small) series of products, you want to make sure that you select the right components. For me this means checking at the large component suppliers (Digikey, Mouser, Future electronics, etc.) that they have the components I use in stock (at least several hundred), have a good price (2$ component vs. a 5$ one) and that there is decent Arduino support. Although your producer will likely use another source for components, you at least have an idea of pricing and a backup supply channel.
SMD VS. THOUGH-HOLE COMPONENTS
Whether you will want to design with though hole (TH) components (= with wires) or SMD (surface mounted device), depends on how many products you expect you will create. If you are going to make less than 50, it is probably cheaper and easier to use though hole (which is done by hand at these volumes). Above this number, it makes more sense to use SMD, since these components that are handled by fully automated robot pick-and-place machines. However, there are a few things to consider when SMD is used: Typically larger start-up cost because of machine programming and paste stencil (200-1000$ is not untypical), less flexible because of this and harder to build prototypes by hand (certainly possible, take a look at the Sparkfun or Adafruit website). But advantages are significant lower cost at volume and sammer designs.
FINDING A PRODUCTION PARTNER
Finding a production partner for smaller volumes (<1000 products), new company or individual is often harder than it sounds. Reason for this is that every customer and design takes time for the producer, but profit and revenue are low and uncertain for a new customer. For this reason, I advise that (with Bill-of-Materials (= BOM), Gerber design files and volume estimate), you request quotes at several production companies, also known as EMS (Electronic Manufacturing Service).
FAR-EAST OR LOCAL
It depends on your experience, plans and sense of adventure. But since you are reading this, I assume that you have some kind of Arduino prototype and are thinking about giving or selling a few (up to a hundred?) online or to friends. In that case I would advise to start local, since it makes communication so much easier, you can visit or call easily, when something is not the way you expect you can go back easily and work things out and (maybe most importantly) you have a smaller chance of being delayed. I have heard and experienced firs hand several cases, where production of a widget was delayed for months due to unclear reasons (of course there are exceptions, but these are just myself and my friends experiences). I have never have had this with local producers. But if you are going to produce 10.000-millions of widgets, then probably you should look into going to the Far East and get some experience there.
The first steps from Arduino to soldering prototype and quotes are relatively cheap:
- Proto PCB(s): 20-100$
- Components for DIY soldering: 20-100$ depending on # and type of components
- 200-1000$: Setup and stencil (only for SMD) cost
- 5-20ct per component placement + component cost
- Other costs: Testing (often optional), packaging and shipment
If you have built or are working on a neat design and have some time on your hands: Why not get started on the process and start building your own gadget. In the beginning ith might all seem very complicated, but once you start going, you will learn and things will make more sense. And if it doesn’t work out the first time, you know a lot more when you want to do it again. Good luck!