Do you think there is a possibility that electronic-device-making will also one day slowly evolve into a Home/Individual/Self-manufacturing process, a la 3d printing?
3D printing seems to be one of the big things forecast for 2013-14, and is predicted to slowly but eventually enable small-scale material/physical manufacturing at millions of homes.
So I was discussing with a friend whether the same for electronic devices might ever happen. After all, as is obvious from the diversity of projects described in the Project Guidance thread of the forum, it would appear that there are tens of thousands of diverse custom applications for different individuals -- whereas, there are only a few hundred categories of consumer electronic products that are mass-manufactured out there.
If this kind of customized home self-manufacturing of electronic devices were to materialize, it probably wouldn't happen in the exact same way as 3D printing; at least that's my opinion -- it's hard to imagine everyone having their own PCB-making and pick-and-place machine, not to mention an infinite supply of parts.
However it could work modularly perhaps, e.g., you might order modules and put them together. (I think the Arduino has been a great first step in this direction.) But this type of modularity will have to evolve into a greater number and variety of common, connectable electronic legos, or perhaps there are some clever solutions that can simplify it? Will this ever happen in the next decade or so, or are there inescapable hurdles to this kind of thing that I'm not seeing?
Silicon fab will never be a home industry. Too much needed to make transistors.
Assembly of modules, that's a possibility. I am busy designing cards for folks to support that. Pick N Place would be nice, hand placement & reflow oven is certainly managable, and when gets high enough, can be farmed out to a small business like cbas-usa.com that can do it for you with stencil, pick N place.
If get big enough, move on from drop-ship all material to a real turn-key operation.
Assembly of modules is indeed probably the only way.
While the already-existent consumer electronics manufacturers may try to fight this trend, this will still be welcomed by the electronics-parts industry in general because it wouldn't cut down on their sales, it would just distribute or displace part usage into another style of production -- if anything, an even greater number of parts would be purchased.
It would be pretty neat if society headed that way -- Imagine if anyone could create their own custom device for whatever they wish -- for keeping track of temperature in a certain part of their home, or a custom-alarm device, or a device for tracking their infant when it goes out of a certain range, or a fully personalized home-automation system, or an LED-based decoration that's controlled via smartphone, or devices for single-experiment-scientific-research, or devices that teachers can use to demonstrate projectile concepts during a physics class about gravity, etc., etc., etc..
its sort of there now if you have money to toss around
a desktop 3 axis cnc for ~1000 bucks
vinyl cutter for stencils for ~ 200 bucks
reflow ovens can be half assed with a toaster oven
all thats missing is a cheap PnP machine, other than you ... I just placed 600 components by hand for a panel of pcb's at work ... took less than a day (it would have taken just as long to program the juki's, and only for one panel total, course once the line is fired up and running it can populate boards in seconds, but for one panel its the same time)
Osgeld:
all thats missing is a cheap PnP machine, other than you ...
If you already have a 3-axis CNC machine - you are about halfway there: Simply attach an SMT vacuum picker (cheap on ebay) to the spindle, and set up some kind of self-feeding rack for parts - and hopefully have everything aligned...
Ok - so I missed a few steps; it wouldn't be easy, but a 3-axis CNC machine -could- be re-purposed for such a use...
@Osgeld, Do you guys (i.e., at work) presently use the mentioned $1000-desktop CNC for the PCBs? I'd appreciate a reference to the product if you have it please.
Apart from the tape dispensing, which could perhaps be solved, I like @cr0sh's idea of attaching a standard vacuum picker/pen to the CNC spindle. I just don't know if a $1000-desktop CNC would have accuracy of 02.-0.3 mm that is necessary for placing parts.
there seems to be a substantial gap between the $1000 amateur CNC machines that people are using occasionally for PCB making, and the "pro" models from LPKF, T-Tech, and similar.
There's another VERY substantial gap between the Pro PCB routers and a real PCB house. Pin pitch, plated holes, soldermask...
You can buy a lot of high-quality PCBs from Seeed, ITead, or OSHPark, for the price of even a pretty bad home fabrication scheme. I bought an LPKF some years ago (pre Seeed, etc), and I don't think I'd do it again!
Osgeld:
I think the tape dispenser is much harder than the actual pnp action, its the biggest fidget on our machines at work
As someone who's tried to design several pick and place machines and built one, I can emphatically back this up. The "pick" and "place" pieces of a pick and place machine are relatively easy problems to solve, and/or have already been solved by other people so you can build on their work.
Reliable, automated, tape advance continues to be my #1 area of pain; particularly with parts smaller than 1206.
giantsfan3: @Osgeld, Do you guys (i.e., at work) presently use the mentioned $1000-desktop CNC for the PCBs? I'd appreciate a reference to the product if you have it please.
no we order most everything, and in a pinch I take some blank pcb, hit it with a coat of spray paint and send it though the laser engraver
I have a CNC machine.
I use it for making cases, front panel, back panel.
3D printing seems to be one of the big things forecast for 2013-14,
3D printing is great for making a one-off prototype, or a one-off custom job.
It's very useful in visualizing something and holding the part in your hand. Then you can test it's actual fit in your project in all 3 dimensions (X,Y,Z).
3D printing is not the answer to everything.
It's slow, it's expensive, and the part it produces is weak (because it's only plastic, and not metal).
Once the 3D part is verified to fit/work, then traditional methods are used to mass produce it (stamping, pressing, injection molding, CNC, etc).
There's 3D printers that fuse sintered metal to make strong parts. There's one that prints complete working bicycles with bearings in place. I've asked when they will start using ABS plastic with glass or carbon fiber core because wow the strength you could build in!
If they can handle fiber core plastic then maybe they could lay wire?