Creating a finished product

So I have a hobby of indoor gardening :wink: I've spent a lot of money on these microcontroller systems that help automate a few devices. I think I can improve/add-on to the existing concept with a couple of ideas of my own. After doing some preliminary research, it seems all the devices that I want to be controlled have an arduino compatible counterpart. Is it possible to build a device and have it look professional enough to wholesale/retail in the open market? What I'm concerned about is the aesthetics of the the interface. The devices on the market now have a plastic case with analog buttons and a small LCD screen that produces readouts of the measurements taken by the sensors. They are produced in China.

So what do you guys think? doable? I have no engineering background or computer science. I am willing to learn and do have a science background. Also, I'll probably outsource a lot of the work. Has anyone ever sold a home made microcontroller and made it look like it rolled out of the factory and not your garage?

Any input would greatly be appreciated. If I said something totally inept, please excuse me. This type of thing is not my specialty, but I do see a potential in the market.

Thanks,
Calvin

Where there is a will, there is a way. Your thoughts are doable. First, you would need to build a functioning prototype. This will give the components necessary to build into your final product. Then, I would design and fab a custom PCB. Eagle is the common design platform that most hobbyist use. Then you would send the design files to a fab house that can produce your board. Then it is up to you to design and fabricate the final enclosure. There are literally dozens of sites that can help you through the process.

http://www.cadsoftusa.com/

I seem to have misplaced my bookmarks for enclosures.. I know someone here will post it.

I can tell you that you will need patience. If you are just starting out, then learn while your building the project. I got into Arduino and electronics about 6 months ago and I am only a quarter of the way finished with my project of an interactive humanoid robot. But your project seems doable if you have the will to learn. Any questions you can't find, ask here. That is what this forum is all about. Look around, read some posts.

http://www.okwenclosures.com/ - Enclosures
Enclosure Hub | EnclosureHub -- Enclosures
http://www.budind.com/ -- Enclosures

Has anyone ever sold a home made microcontroller and made it look like it rolled out of the factory and not your garage?

My product isn't microprocessor based, but yeah I've been producing goods in-house and they look just as good (and even better) than my competitors.

Let me tell you, the design, electronics and PCB is the EASY part... enclosures and packaging will be the hardest and expensive part! -- but this is what your customers will see - the packaging and overall look.

You can take several routes here.

  1. Buy a pre-fab enclosure, and CNC mill it yourself (assumption is you have a CNC machine). The CNC machines makes all your work look consistent, exact.-- without manual labor or manual mistakes. Cons: Expensive, have to buy the CNC machine, you'd have to learn CNC stuff, and it's another "to-do" item on your busy day. Pros: Other than the initial CNC machine investment ($1K+), you can produce enclosures "on-demand" and cost per unit is reasonable.

  2. Buy a pre-fab enclosure, and have someone else manufacture your front and rear panels, or send them the enclosures you bought and let them CNC it. Cons: Expensive, Expensive and Expensive! Pros: Outsourcing, no worries, frees up your time. But your unit costs will be very high. Can you price your product to recoup this?

  3. Have a metal shop design and make you a custom case. This is probably going to be cheapest per unit piece, but will require the most capital outlay. You'd have to order several hundreds to bump the price down. Pros: You get exactly what you want, cheapest price per unit Cons: Big money invested... and if this is for a new product, you don't know yet whether your product will sell or not. So could be risky.

If I were you, I'd go #2... then down the road do #1 .... then #3.... (assuming your business is still alive).

So you start out with the most expensive option #2, and as your product sells and you got money rolling in, you'll have the money to do #1...
then as you get more and more orders, you'll have plenty of money but not enough time.... so you go to #3.

Makes sense?

Wow, great information guys. I am grateful you all took the time to share that. This is going to help me GREATLY!!! I'll keep ya posted :slight_smile:

I've made lots of boards - you can see many of them here.
www.crossroadsfencing.com/BobuinoRev17
and I have more in the works.
Can customize an arduino for your needs as well.

CrossRoads can be a huge help :wink:

I agree, the electronics and firmware is the easy part. The enclosure is the hard part. Best to use an off the shelf enclosure if you can. If not, have a look at shapeways.com.

The hardest, and possibly the most expensive part, is certification. Europe and Japan need "CE Certification." Depending upon what you're building, you may be able to "self certify" but you may have to pay a "noted agency" to certify it for you.

In the US, if your clock speed exceeds some amount (I think 7 MHz) you'll need FCC certification. Similar deal.

I'm also contemplating using a 3D printer to actually build my enclosures and a few other parts for my product, but still worrying about building materials they use in the affordable printers. Does anybody know if something as tough as MDF can be used in a 3D printer? The shapeways.com samples all look somewhat fragile.. :~

Use the MDF in what manner? That's like really fine sawdust & glue particle board isn't it? Used for kithchen cabinents & speaker enclosures?
You would mill that out to make an enclosure?
I generally see 3D printing as melted plastic being built up in layers to make a part.

evkar:
I'm also contemplating using a 3D printer to actually build my enclosures and a few other parts for my product, but still worrying about building materials they use in the affordable printers. Does anybody know if something as tough as MDF can be used in a 3D printer? The shapeways.com samples all look somewhat fragile.. :~

Are you contemplating using MDF as the chassis of a 3D printer? Our hackerspace has 3D printers made from plexiglass and from plywood as both are easy to laser cut and are dimensionally stable enough. MDF is very dense and has good dimensional stability and routes good but I haven't seen it used for a printer chassis yet. I guess you could try it.

3D printing with plastic still looks a little wavy on the surfaces to me unless you use a really high grade printer.

Back to MDF, it's easy to machine and can now be powder coated which looks really good. It's pretty dense therefore pretty heavy so that could be a plus or a minus for an enclosure. Worth considering for an enclosure for an electronics project.

Doable, totally.

I just suggest you do as your future competitors do: outsource production to China.

Let me share my experience with your, since I am undergoing the exact same situation as you.

I have a working prototype of my product, which I drive around with all day. It has a home-made, project-specific Arduino board, with all through-hole components. I also used MDF wood as a "chassi" to the assembly. the product also consists of 10 identical daugherboards, that are connected to the Arduino. Since these are double-sided PCBs, and 10 units, I had them manufactured by iTeadStudio in Hong Kong (total for the 10pcs of double-side 2"x2" boards: $17 shipped to Brazil).

Prototype 2 will have a lot of features over prototype 1, including better enclosude, which will be made in acrylic. It isn't cheap to have acrylic laser cut just for me, but for 1 unit it is great.

If I don't need a 3rd prototype, I'll then just find a manufacturer in China to whom I can send my project files. It is a lot cheaper than having it manufactured in Brazil.

I already filed for a patent for this product, based on Prototype 1, so it is currently "Patent Pending" (or Required). It may take up to 3 years until my patent is granted.

So, my suggestion is: before working on an enclusure, work on prototypes, just so you can test your ideas and concepts, change them, create more, and over the weeks you are using your own product you'll come up with thousands of ideas on how to improve it, and most of them will make the final product.

If you can create more than 1 prototype, have a very trustworthy friend/relative use it as well and give you feedback.