20 years design & manufacturing electronics in China & Taiwan
CEO of www.titoma.com
MD of WeLL Design Taiwan
I have developed & manufactured electronics in Asia for the last 20 years.
Most people go through different versions of prototypes, putting it in the hands of target users is essential, keeping your idea secret is a big mistake.You will get a lot of feedback which more often than not lead to drastic rethinks of the whole device. So I strongly suggest to go to China only after you are VERY sure what you want. At home you can iterate a lot faster. Best would be for you to start playing around yourself with Arduino, which is a very newbe friendly way to make and program hardware. If that's really not for you then find some enthusiast at a local makerspace to do it, they're all over now http://spaces.makerspace.com/makerspace-directory.
You can also quickly iterate on the housing design using free 3D CAD software and 3D printers.
Do realize that Arduino is a great way to validate your specification, but not a cost effective architecture for a mass manufactured product, so a lot of the work will need to be done all over again, see https://www.linkedin.com/pulse/kickstarter-prototype-production-100k-enough-keesjan-engelen
So don't spend months and months to get your firmware just right, to get your cost to an acceptable level you likely will have to switch to another processor which needs to be programmed in a different language.
Once you're happy with the way the prototype works I would look at manufacturing as follows:
1. China is the best place to manufacture electronic products. If only because they have the most complete supply chain in the world. https://techpinions.com/why-cant-the-us-build-consumer-electronic-products/41950 China has many thousands of factories making any kind of part imaginable. They are especially unmatched in any part which needs to be custom made: housings, batteries, PCB’s, motors, LCD’s, cables….
2. One of the keys to reducing cost, time and risk in product development is to use existing parts as much as possible. Quite a few firms for some reason still regularly have LCD’s custom made. Why pay $250,000 in set-up, and months in mold production and sampling, if you can design around an LCD which is already being manufactured in huge quantities?
3. 60% of the unit cost of a product are decided in the architecture stage, and 20% during the design. As they say at Toyota: ‘‘Skillful improvements at the planning and design stage are ten times more effective that at the manufacturing stage.’’ http://www.design4manufacturability.com/DFM_article.htm
4. Doing your design without close interaction with the China vendors means you’re missing a lot of information, which leads to sub-optimal design decisions. The further you advance in the design process, the more expensive it gets to change anything. So if 98% of a product’s design is already done and validated in testing, you can hire a really smart China manufacturing consultant, but there is only so much they can do because most aspects are frozen already.
5. Another important tenet of Design For Manufacturing (DFM) is early supplier involvement. If you need custom parts better make sure that the factory who will make them actually feels comfortable manufacturing them. This requires a lot of communication in the design stage.
6. If you’re in the US the 13 hours’ time difference and language barriers make a smooth & fast collaboration with Chinese engineers very difficult. Plus if you have never worked with the factory before they’re not going to put much time in optimizing a design they may never get to make. Only for established clients does a factory see a very real chance that they will be responsible to make the part efficiently and reliably, and so they will be very keen to help optimize your design for their processes and capabilities. The flip side is that you are optimizing your design for this particular factory, so you’d better be sure that they can indeed deliver.
7. Picking a factory is not easy, the proof is always in the pudding: how responsive are they, how do they react when there is a problem. Are they going to say: “Oooh you do not want the batteries to explode, you never specified that, well that actually costs a lot more”. You have to kiss a lot of frogs to find your princess. Except that in manufacturing kissing is not good enough, you actually have to get into bed with them, only after you have wired your deposit and have been working with them for 6 months do you really know your bedfellow. This goes not just for the factory doing the final assembly, but for every factory making custom components for you. Building a reliable roster of suppliers takes many years, and a lot of “tuition”. In China going for the lowest bidder in the end invariably ends up being the most expensive option.
8. The more complex your product, the more inter-dependencies between parts, the bigger the mess trying to get a product manufactured in China when it has been designed in a vacuum somewhere else (and it’s not just Kickstarters who run into this problem).
9. The first samples for most any custom made part in most cases need some improvements. Some companies work will send their “China manufacturing guy” but it takes a lot for 1 person to master all the intricate considerations which went into the design of the cosmetic appearance, the construction of the injection mold, the layout of the circuit allowing it to pass FCC certifications, the adjustments needed in the calibration software… So at IBM for example knowledge transfer in manufacturing is considered a dirty word, the original designer of an aspect of the design has to see his design through all the way into manufacturing. Flying each of the 12 different functional experts over to China quickly becomes expensive, and not every engineer wants to camp out in a China factory dorm for months.
10. So doing electronic product design as close as possible to the factories making the parts and doing the final assembly will significantly lower your development costs, your unit costs, your time to market and overall risk.
11. The knee-jerk reaction to get this done is to asking a China ODM factory to adapt one of their products to your wishes. This can work well if your changes are minor. But if you want them to do more there are some serious risk with this approach: you don’t have any ownership of the design IP, and it more likely than not it will take ages. https://www.linkedin.com/pulse/china-factory-designs-7-pitfalls-avoid-keesjan-engelen
12. If you're ready to camp out in China for a good 6 months then https://hax.co/ would be a great place to be mentored. https://www.hwtrek.com/ provides a lot of China & Taiwan resources for HW start-ups. You could also give the job to a consultancy such as www.detekt.com which specializes in helping start-ups.
13. Setting up your own design team in Asia can be tricky, the success hinges heavily on picking the right CTO to manage and build out the team. And in China this kind of talent tends to switch jobs very often, especially after they have obtained a foreign “stamp of approval” on their resume.
14. It is true that hardware can get copied very quickly, but copycats are unlikely to invest their efforts in an unproven product. So this risk comes up only after your product starts being successful in the market. By that time everybody can copy it, no matter whether you manufacture in the US or in China. In fact products manufactured in the US likely have to sell at a higher price, making them a more attractive target for copycats.
15. Furthermore, the painful truth is that excellence in hardware no longer gives your company a lasting competitive advantage, to build barriers to entry hardware companies these days need to work not just on brand and distribution, but also on Community, DataBase and Software. https://www.linkedin.com/pulse/toasters-roller-coasters-how-hardware-startups-can-john-melas-kyriazi Doing this well is no mean feat, so I would focus most of your effort there, most of my clients are companies which realize that hardware no longer is their core activity.
My firm www.titoma.com works with well established companies, the reason is that 7 or 8 vendor partners put a lot of effort in collaborating with us from the very beginning to optimize every aspect of the design for manufacturing. They invest this time because from experience they know that the products we design will be manufactured with them parts for years. Unfortunately not every start-up becomes a runaway success, such is life, but I cannot afford to burn bridges with my partners, sorry!
When looking for good partners in China you will need to spend (and keep spending) a lot of effort on selling the factories on the fantastic prospects for your product, otherwise your 1K pieces project will quickly become their No. Last priority. So being there very often is essential.