Manufacturing a product
darren
If I've had an idea for a product, how can I go about mass-producing
it in order to sell?
For instance, a small gps transceiver in a simple plastic or rubber
casing.
- can I find out how much this would cost to produce? (and can i find
out for free?)
thanks for any help!
darren
Smári McCarthy
Hash: SHA1
Hi,
I must say, this question is alarmingly on-topic; grounded in a
sensibility that is not often found here. But Darren, beware, this list
waxes philosophical about the tiniest of questions, so the answers you
get are probably going to be misleading at best.
That said, the general path is:
1. Develop the product concept, plan for target groups, etc.
2. Build a prototype.
2a. Plastics could be done by manually milling the
casing or creating a cast mold from urethane or wax.
2b. Electronics can be designed in a program like Eagle
and either made by hand (for small quantities) or
purchased from a third party; various companies do
cheap circuit board fabbing.
2b1. If your design includes new microchip
designs, you're going to have a harder time
prototyping, but I'd suggest writing the
VLSI and then working with a FPGA for the
first round. Remember that microchips are
products in their own right!
2c. Metal parts required can be machined easily enough,
just avoid awkward sizes and shapes. If you don't
know milling/lathing yourself, buy the work from a
third party. Companies with CNC mills and lathes
will often output from good CAD files for very
little money if the design seems interesting to
them. (These companies are run by enthusiasts who
suffer horrifying tedium; surprise them)
2d. Getting the initial prototype 3D printed to some
extent is not a bad move but beware that it can be
costly if the object is big. Z-Corp 3D printers are
everywhere nowadays and give a nice finish, but also keep
your eye open for SLS's, SLA's, FDM's and other
acronymic digital fabrication technologies.
2e. If the assembly of the prototype from the parts
gathered doesn't work, it's because you didn't do
the CAD work and the homework well enough.
3. Figure out who's going to bankroll the mass-production.
3a. This is where the market analysis stuff comes in
handy.
3b. If it's going to be you, try and make sure you
will be able handle the product flopping.
4. Find a fab that's capable of mass producing to your demands.
4a. Many smaller ventures hand-craft each item, but this
raises price and slices profits.
4b. China is popular.
4c. So is Taiwan.
4d. Bear in mind that unit cost tends to drop as unit
number goes up, in part due to externalization of
the true costs of production (read: larger companies
use slave labor more liberally)
5. Profit.
The open path, being advocated here, is much the same, except:
Exchange step 1 for building a website.
Exchange step 3 for learning how to do things yourself.
Exchange step 4 for uploading the design to the website for free
download and stick an ad there offering the object
pre-fabricated with a personalized design.
Exchange step 5 with the knowledge that you have helped humanity
by not being a gluttonous bastard.
Welcome to the list, good luck on step 2 (it's the only one that
matters) and viva la revolution, or something.
- Smári
Smári McCarthy
sm...@yaxic.org http://smari.yaxic.org
(+354) 662 2701 - "Technology is about people"
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.6 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org
iD8DBQFI9wSi9cJSn8kDvvERAlwTAKCFVarqOckbf2N1JJbZ+0s88Q1BjACgtXJt
EkysNQp0ctx2Bp8c4/bFTdU=
=9K+z
-----END PGP SIGNATURE-----
Mel Chua
Smári offers some great tips here especially from the "what techniques
can I use to get a first functional prototype?" perspective. I'd like to
address your questions on DFM (design for manufacturing) and production
from the perspective of a working engineer.
>> If I've had an idea for a product, how can I go about mass-producing
>> it in order to sell?
Smári more or less covered this - I will point out that making a
functional prototype can be very different from making a mass-producible
design (example: I can make a motor controller in <1hr from "I have an
idea!" to "I've breadboarded a circuit and have a motor spinning on my
table," but laying out the PCB, making sure the case design I have in
mind is injection moldable, etc. would take days or weeks more to verify
(ship it out to manufacturers, argue with them about tolerances, pick
materials, get it back, complain about the solder masking, repeat...)
>> For instance, a small gps transceiver in a simple plastic or rubber
>> casing.
That's still a pretty generic statement. What does the GPS device do?
What chip will it use? What requirements (does it have to be durable?
small? light? waterproof?) does it have? (All these will dramatically
affect cost; I'm talking a 2x-10x or more difference and that's a rough
guess, even.) Plastic or rubber - what type? The exact material will
determine what fab methods you can mass-produce with, which will
drastically affect the cost. Also, what quantities? As Smári pointed
out, setup cost is usually the bulk of what it takes to produce a first
round of mass-manufactured prototypes.
>> - can I find out how much this would cost to produce? (and can i find
>> out for free?)
If you have a detailed mass-production design, yes. You can send that to
the companies that would mass-produce it for you and get a quote. If you
have a working prototype, you can usually work with an engineering firm
or an engineer to give you a rough quote on how much it would cost (1)
them to redesign it for mass-production, and (2) mass-produce it. (The
first is often dependent on the second; it will take me more time to
reengineer something to make 10,000 $10 prototypes than it will take me
to figure out how to make 100 $1000 prototypes, and there are boundaries
to how cheaply I can make something at all - for instance, a laptop for
$5 isn't going to happen soon.)
Given the description above ("small gps tranceiver in a simple plastic
or rubber casing") you're going to get one of 2 answers:
(1) Exactly $X. (Run away. Run away VERY fast. They can't possibly know
this. It's like saying "How much will my birthday party cost?" and
someone saying "$125" without asking you first whether you'll be sharing
a single bottle of cheap wine with 3 friends, or flying in performing
elephants for a 2,000 person gala.)
(2) It Depends. (or "It depends, maybe between $X and $10X or so, what
are you trying to do?")
Good engineering firms won't work with you unless you have a solid plan
and resources to back it up - a working prototype and the knowledge of
how you're going to finance it, take it to market, etc. Their work will
also almost certainly not be free (example: as a lowly intern at the
product design consultancy I worked for, customers were billed $75/hr
for my work, and this was when I was 19 years old, a student, etc - so
think about how much an experienced engineer costs. Think order of
magnitude.)
A good firm will, however, talk with you (if they think you are serious,
not just a person with a cool idea who has no idea what they are getting
into and might flake out when they realize how difficult it is - because
it can be *really* hard to get something to market) to explore the
possibility of a relationship, and you'd work out these kinds of details
during those conversations before you talk about how much you'd pay the
firm. Some firms - not all, but some - will do pro bono / reduced rate
work for things that they *firmly* believe in, if they see somebody
standing there who can and will take it "all the way" except for this
One Part that they don't have the resources for ("if only I could
redesign my mold to use this eco-friendly plastic.")
This is probably more than you wanted to know - and not to scare you
away from doing things, it *can* be easy - you can make something, fab
it at home or in a friendly local shop, maybe it's a design that people
can do themselves ("hey, get a milk jug and some scissors, cut like
so...") but it can also be a difficult, long, and complicated process
and some people don't expect that and that hangs them up.
My philosophy is that if you acknowledge that the worst-case big
expensive blown-out thing *could* happen and learn about that, and
*then* go "Ok - how do we get this done without expending all those
resources and time?" you end up with fantastic results with clever
shortcuts that don't degrade your process or your product, and you're
not continuously having heart attacks because you just learned how many
thousands of dollars it would cost to make a mold.
I am a hacker. I make things in my living room, in friends' warehouses,
garages, everywhere; I love the cheap, last-minute, quick-and-dirty late
night hack. I'm also an engineer, and I know how far hacks tend to go
(and not to go), and if you want to make a big impact and change the
system later, the easiest and fastest way to do it, imo, is by going
through the current systems, learning about them, fixing them as you go.
What I described above is the current system. We learn from it, we make
our own, we change it from the inside.
-Mel
darren
I wonder how small a gps+cellphone combo could be made, thinking about
theft prevention for bikes, snowboards and so on - and also wearable
units, eg disguised/embedded into a necklace/flipflops for children.
(assuming basically no physical controls - just the simplest/smallest
unit possible).
ie your bike gets stolen, your device every 15 minutes makes a quick
phone-home to report its location. You log on to the website and track
your missing bike/child/loved-one/loved-one's-lover, whatever.
darren
Smári Páll McCarthy
hang around sheep's necks; I haven't seen any smaller but there's nothing
beyond antenna size that prevents it from being reducible further.
Antennas are always a bit of a bitch.
However, anybody who wishes to hide tracking/surveillance equipment in
their children's clothes deserves to be shot.
- Smári
Smári P. McCarthy
sp...@hi.is - 662-2701
darren
smartphone(+battery), not the most simple combo possible that I want
to investigate! :)
Preventing child abduction in many parts of the world is still a huge
problem to be tackled. I think there is potential to market such a
product to rich¶noid Westerners wanting to safeguard their child,
eg when on holiday. But also bikes, snowboards, cars, handbags,
murses - you name it!
-darren