[stewi]

Assuming I've a new concept of a high speed surface mount assembly machine and I'm well capable of building a prototype. Where else could I sell it, if not to China?
Further assuming electronic assembly is done to 60% in China. Where are the other markets? Can anyone point me to a more accurate market study than my own guess?

[stewi]

No numbers here either Global Surface Mount Technology (SMT) Equipment Market to Reach US$11.2 Billion by 2017
unless I'm willing to pay $4,800 for the full study.
However, some comforting words about market recovery.

[RomanLini]

Well China and India are just going to make their own machines and not buy western technology (unless they buy ONE, you know what I mean).

And most of the west will buy new machines if they have big budgets and used/surplus machines if they have small budgets. Or (possibly more likely) western companies will just out-source their SMD and other work to the cheap shops in China and India since those countries are getting so much better now at English language communications and fast postage, and their incredibly low labour costs makes is more and more silly trying to manufacture in the west.

Most of the PCB makers in Australia have closed down now as electronics manufacturers here just outsource work direct to China now. The Australian SMD and soldering production will be next to go.

[stewi]

Looking at this thread I'm not afraid of Chinese copies
http://www.cnczone.com/forums/vertic...ml#post1007663
It takes many years experience to make a copy work well.
I wouldn't even mind selling to the Chinese market, but for an entrepreneur like me, this market is inaccessible, unless I would approach larger US or German corporation, who are already established in China.
New projects will likely require massive funding to make an impact and be recognized as a potential competitor.
Nowadays even angel investors in California sitting tied on their money and I don't have any money to spend either.
However, I'll keep on posting open source ideas. Maybe one day I find an investor for ideas I have not revealed yet.

[keebler303]

Some American companies have brought their assembly in house to reduce lead times, reworks, etc. One that comes to mind is Gecko Drive. Mariss has discussed his operation in several threads here on the zone.

Matt

[RomanLini]

Looking at this thread I'm not afraid of Chinese copies
...
It takes many years experience to make a copy work well.
...

Unfortunately not so much time is needed these days. Basically they just need to buy a mechanical product, pull it apart and spend a couple of weeks measuring all the parts and whipping up some CAD models. Then they just reproduce it using their CNC machines and masses of very cheap human labour.

Keebler303- Yes some smaller manufacturers are still doing in-house. These are some of the people buying 2nd hand PNP or new low-end PNP (Madell etc) for $10k to $20k price.

But over the last years I've watched a few trends happening. Where people used to make their own PCBs for hobby and small manufacture now they email a file and in a few days receive back perfect PCBs. Almost nobody makes PCBs at home any more.

The next step is soldering and SMD, people are sending it out more and more to production houses, and it's getting common too for the houses to stock the parts at least all the common stuff like resistors and caps and common semis.

So int he future I think it will basically go like this;
1. You send a file, they make your PCB,
2. They say "Do you want fries (SMD) with that?"

People will say "Yes" and get back a finished PCB with all parts soldered, at a competitive price.

That kind of thing is also getting common with production houses to do contract CNC/lasercutting/printing etc.

...
However, I'll keep on posting open source ideas. ...

It's a bit of a catch-22 of course, if you post as open source you can not sell the idea or patent it, and someone might even start in competiton making stuff based on your ideas before you can even start production of your own ideas.

[stewi]

Thanks for your insight feedback. I have some good friends small shop owners, who are trying to get into the online ordering market as well. All I heard it is tough and already very competitive.
However, although I like the challenge of the high speed machines, this market will likely always go to cheap labor countries. That said, there may be a market for low cost and bench top machines. Low end usually also means no good, the Madell machine, from what I see in videos, may be here an exception.
Assuming I could re-focus on a small machine, likely single head, capable of 0201 to BGA, vision system, some feeders for under $10k, I could likely make these until copy clones from China.
No matter from which angle I look at surface mount assembly machines, I would loose.
Rather than loosing some $100k investment in a new start up business, I don't loose anything in posting some of my ideas in open source feeder and rotary head.
Sure, this forum has also Chinese attention and they read this stuff as well. Yet, in open source everybody has a chance to make a better product out of it, but now they can do it before this product comes out of China.

[Nails]

>>Low end usually also means no good, the Madell machine, from what I see in videos, may be here an exception.

View topic - Madell SX1020 Pic & Place machine • adafruit industries • Customer Support Forums, DIY Electronics, Open Source Hardware, Arduino

Not a lot of good things to say about the Madell machine.

[daedalus]

just to put another thought out there, but look at companies like makerbot, they produce a product that is inferior in every way to the commercial systems, but with open source and a good price point they have created a massive amount of goodwill and publicity.

Their last round of investment brought in $10 mil.

[stewi]

Open source for a company (Makerbot) in California is different from my situation living in Georgia.
I used to hear people talking about new ideas in California all the time, at breakfast in Los Gatos or at the San Jose airport. They have to be open to investors or potential partners and at the end of the day they have likely revealed 3 to 4 times their ideas of the next I-pod.
Georgia has one of the highest unemployment rates of the nation.
Should I ask my neighbor sitting in the Marta on the way to the airport, if he wants to invest in my ideas? According to the music, he is listening, he hates me, or actually he hates everybody and everything.
At age of 6+, bi-lingual, bi-cultural I know quite a bit about engineering. I don’t have money to invest and hardly find anybody here to invest in my ideas.
I do admit, that I use this forum advertising my concepts, hoping somebody in California will find me. Since I’m also contributing to this site, the operators of this site shall forgive me.

[daedalus]

stewi: they got their investment due to having a strong brand, market awareness and a popular product. I highly doubt Foundry Group are interested in the technology per se. In a sense their web address had more to do with their investment opportunities then their physical one. That said im sure California has some advantages in terms of networking.

The benefits of the OS approach is you can get your ideas out there, and recruit people to help with the design / testing effort, in the hopes of getting enough critical mass to attract investors. The downside is you end up loosing some control of the idea, and you loose the commercial routes (patent, licence) but from your previous posts I got the impression you weren't too keen on them from past experience.

There are also the likes of kickstarter and other social investment schemes which might be useful for early-days funding.

for a good example of what distributed teams can achieve, check these guys out: http://www.wikispeed.com/

[stewi]

Can I make a quick poll, who would be interested in a Wikipick&place site?

[daedalus]

stewi: it might be worth doing a new thread, that way you can use the poll feature on the forum, and more people will find it. I would be interested in such a site, as a user and potentially a contributor.

[stewi]

Sure, I want you on board.

http://www.cnczone.com/forums/rc_rob...ick_place.html

[scsi]

Let me join the discussing if you guys don't mind. I seriously believe that there is a fairly large international market for a decent desktop PPM under $15K, which is curretly occupied by madell etc. They have been trying to make "an electric bicycle" for years and it's still falling apart in customer's hands. It is certainly possible to design and build a sub-$10K machine that will greatly outperform what is already out there and people will put it on their desks and use for rapid prototyping and small batch production.

Rapid prototyping is one area where outsourcing to an assmbly house becomes cost prohibitive. In case if you need to assemble only one (or ten) boards, you'd rather do it on your desk. And then re-do it if necessary.

Same goes for stencil printing. I'd rather have my desktop machine do solder dispensing rather then spending hundreds of dollars on a proper stainless stencil just to find out that you need to change your pcb layout slightly a day later.

In rapid circuit prototyping, there is only one stage that can be worth outsourcing and that is PCB manufacturing. The costs are low and turnaround time is great. Milling pcb's on your desk is a messy business and doesn't allow you to use more than 2 layers. Lack of solder mask is another reason not to mill your own pcb's. The rest can be done and should be done on a desktop PPM.

[RomanLini]

I wish you guys luck, but I still get that feeling that by the time you have all worked hard together for months and got a good opensource design close to done, someone with money and cheap manufacturing facilities (Eastern Europe? India? China? A hungry manufacturer in the US?) will just pick all the good bits out of it and start making machines, and get all the customers (and all the profit or course).

[stewi]

This is a good and valid point RomanLini and I'm scared about this as well.
There are a few things I would propose for this new design to make copy clones more difficult.
1. We could develop our own custom motors (linear motors) and drives.
2. Avoid using off the shelve parts or aluminum erector sets
3. Use custom tooling and/or fiberglass, carbon fiber parts
4. All upgrade parts, multiple nozzle heads, vision system will only fit on our base machine
5. Since everybody in the team has already developed a machine or parts of the machine and volunteers his knowledge, we have zero initial costs.
6. We will likely develop a new machine much faster than any other company.
This will be hard to compete from larger corporation and old style management, even with lower cost of copying.
7. We have a better moral, not being profit driven. Eventually, customers will reward organizations with ethics and engineers standing 100% behind the product instead of products created with child labor and no support at all.
8. We have likely better chances to get funded than a hungry US manufacturer.
9. We'll likely launch in multiple countries. Since I'm bi-lingual, I'll spread the word in Germany as well.

Go Wikiplacer!

[scsi]

For that same reason I would highly recommend keeping software in closed source domain.

On the other hand, developing own linear motors and drives would not stop competition from copying them. There is nothing particularly difficult in doing that and pretty much any off-the-shelf servo amplifier would work.

[RomanLini]

I can tell you guys are enthusiastic, so please let me apologise for posting a negative slant on it.

I'm excited to see what you come up with, especially linear motors which I think are some of the coolest things ever.

[stewi]

Here is another option of linear motion, but less expensive than linear motors:

[ame=http://www.youtube.com/watch?v=a4c7AwHFkT8]Control of Inverted Pendulum with Servo Pneumatics - YouTube[/ame]