[TriEmbed] Advice Needed

[John Vaughters]

Charles,

It's great to see people trying to turn ideas into actual products. for a significant portion of my career I was involved in design, product development and pricing in the late 90's, with significant experience in board assemblies. I know that technologies have improved and costs have come down, but the development and sourcing has not really changed that much. Even during the rest of most of my career, I have been involved indirectly with the development of products.

My advice will be based on lack of clarity of your true intentions. Are you selling because you just want to learn and maybe pay for your hobby?, or are you serious about getting a product out there? You don't have to answer that question to me, but you need to know for yourself. The advice below is geared towards getting a working product out with minimal cost and reasonable speed. 

1. As already stated, reduce your BOM to the bare minimum, I like the ATiny idea.
2. Forget about CAD for now, make prototypes with your reduced BOM and prove your product works
3. Use perfboard or bulk FR4, superglue and wires to keep prototype costs down
4. Buy manageable components that you can solder yourself for the prototype
5. Find a very basic cheap project box to house your prototypes
6. Demo the product and test the heck out of it

Now you have an idea what your product will look like so you will want to begin the final designs which will take several iterations in the design process. This is where you have decide what direction to take. Costs go up rapidly from here. There is no low cost options once you start building boards and design cycles will kill your cost and your time. You may want to sacrifice cost by looking into perf-boards that are a little more expensive, but will allow you to create an actual product that you can sell without custom board build. I say this, because it is probably a very good idea to create an early production product that you can build completely on your own to test the market. Even if you loose a little money on your test sales, it may be much cheaper than going down the road of custom board builds.

If you screw up custom board builds, you are left with a pile of scrap and a good chunk of change gone. For this reason, I go back to the question of your intentions. Are you bent on learning board building? or you are looking to sell a product? This is very important. Board building is a craft, a real skill. Bad designs cost real money. I have often destroyed things in the sake of learning and consider it more efficient than paying for college classes, but make no mistake it is a learning cost.

If you get to a point where you think yo have a real product, consider talking to small build board manufacturers. They will save you alot of money in scrap by allowing them to design the board for you, or look for people that contract those services. Don't get me wrong, building your own boards is achievable and not all boards are difficult, but trust these words from significant experience, weigh your choices carefully. Even with an experienced designer there will be design iterations. 

A little advice on cost of your project. Take the cost of the typical Newark or Digikey 1000pc price and cut it in half. This will be approximately what a board builder will pay as true cost. Also, do not trust the cost that a board builder shows as component cost, they love to squeeze 20-30% markup into the components, but keep in mind that you will not get that cost reduction until you ramp up the volume. They will tell you your volume is not high enough, but it doesn't matter, they get volume pricing better than they will show you in most cases. Again, you have no recourse for negotiating until you get your volume up. If you find yourself at this point, I cannot stress enough how important it is to have your board design solid. You DO NOT want to have a box full of assembled product that does not work.

One last piece of consideration. The lowest cost board will be with the smallest components and smallest board size, but you may want to consider a design that is repairable. There is a cost cut off in product volume where boards become throw away, but you will probably find that volume rather high. Investigate several designs that may allow you to repair vs. throw away.

I wish you the best of luck and please keep us informed of your progress.

Thanks,

John Vaughters
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.triembed.org/pipermail/triembed_triembed.org/attachments/20131004/01876e66/attachment.htm>