Whens RepRap likely to take off?

Hello, it's been a while, sorry i haven't replied to my own thread until now, just got that spam in a private message heh , so now is as good a time as any to say hi and thanks for all the replies, interesting stuff, its a pity im not quite ready for building one at the moment but looking forward to seeing what happens in time, ill make sure to have a look now and then.

Just taking any surface mount component and intermix it in my own layout with as many layers as I want at my home for a small cost would be a dream for any Electronics Enginering guy. Adding the capability of removing parts and material and rebuild when you've screwed up the design would also be useful!!

Suppose you could just place a BGA chip on its back and then start to coat its "legs" up layer by layer and add insulation inbetween, then fiddling in some 0403-size coupling capacitors somewhere in the "pcb".

Especially BGA's are a pain to solder since they have the pins not only at the edges but all over the bottom, (making the possible number of connections N^2 instead of 4N).

Perhaps printing of whole components would make most sense when it comes to power components such as driver transistors, current limiting resistors, rectifiers etc.

Print a "tray" that will hold all your components, upside down as you said, and that can be reliably fixed relative to two edges of the rep-rap tray. Place your parts, replace the tray in the rep-rap, (possibly the first time it's been put in place correctly,) and inform the machine you're ready for stage two.

For this reason, it might be good if there are two slots, to engage such a tray, on perpendicular edges of the z-axis table.

wonder what sort of precision and resolution you need to do that, legs have spacings of 0.5 mm today, maybe you can help the machine a little with a large magnifier and a keypad for small corrections?

... the fab@home-guys made funny electronics with embedding conductive trays in elastic polymer, cover the tray-canales with polymer too, so that's a closed sandwich-structure and then insert the electronic components through the cover ...

So you have an elastic sheet, which you can bend and roll on a curved surface and you can replace blowed components, if needed.

Play this a bit further, with inserting SMD-parts in thr trays, so you can build your multi-layer-electronics with complete housing too ...

Viktor

Edit:
sorry, the direct link didn't work because of ':' in the namespace, so look at the fab@home-Homepage -> 'fabbing things' -> 'Gallery of ideas'

[www.fabathome.org]

Edited 1 time(s). Last edit at 11/16/2007 07:04AM by Viktor Dirks.

This thread could be profitably calved off in at least 3 directions at this point so to focus on the "Can/should we print circuitry?" leg:

Woods metal definitely has an allure resulting from ease of use (low melting point) but the same trait is also a drawback (circuits melt when left in car). I would be inclined to look towards relatively high temp (epoxy) substrates teamed with "easy" (moderate temp) silver solders in the near term. Silver solders have good conductivity and are relatively robust (flexible, tensilely(sp?) strong, heat resistant). Eventually we might look at electrochemical deposition of relatively pure metals, but that raises problems related to specialized feed stocks and waste products (imagine a machine which could print IC chips to order but required plutonium as an ingredient and produced lots of waste nitroglycerin). I agree with the idea of evolving the ability to print electronic components one at a time. I think that resistors would be dead simple (carbon plus a binder connecting 2 wires, length = ohms), capacitors seem a little tougher but doable with in-reach techniques. Fab@Home has already done batteries and switches. Diodes and transistors I don't know enough about, but my gut says that cost effective transistors, much less IC's are at least a paradigm away. Same goes for motors.

Thinking about resistors...
If you can mill, and you have pick-n-place, you might add a probe.
Make the component a little oversized, test, then mill down to the proper result. Make your resistor or capacitor overwide, test, then grind down until the resistance is high enough, or the capacitor capacitance low enough.

Definitely blue sky. though. Maybe recursion three, or four.

why would you make resistors and capacitors when highly accurate ones cost fractions of a penny and come on reels that a machine can eat?

---

[www.hydraraptor.blogspot.com]

Fewer specialty resources to hang off the machine.

I've thought a little differently about the resistors... develop a way to vary the resistance in the circuit board pathways themselves and they become the resistors... The width and length of the pathways would play a big part in it... varying the conductive qualities of what you're extruding would also help...

You've also got to look at the evolution of it... yea you can easily drop in mass produced electronic components... but then you're still reliant on that sort of manufacturing to give you exactly what you want...

I'd eventually like to see the prototypers capable of building the PIC's and everything else... You're not truly self replicating if you're buying your electronics from a factory...

Another Very Interesting Topic that has been spawned by this thread is on the question of "Who would want a RepRap and why and at what price?" and the related (and so far mostly unspoken) question of "What manufacturing capabilities should RepRap have?

Hi Brian,

... a fair conclusion and summary

My path ist something between - i from time to time used 3D-mills in job and at home, actually i'm on activating an old CNC-system as 3D-mill again, so i can make milling at home too and tinker with my children in ideas, designing and fabbing freely ...

And i'll tune/update the system with a paste- and FDM-extruder, so i can play with RepStraping too ...

As with my tripod-concept i'll try to design 3D-systems, which are more simpler and easier to assemble, then Darwin or a typical cartesian robot, so it's a bit of a sideway, but very with interesting consequences for 'anarchic fabbers'

My far-goal is a kind of evolution in the RepRap-idea combined with the concept of 'fogletts' or 'utility fog' ( [en.wikipedia.org] ) - a very high count of micro-robots, capable of interconnecting to a kind of crystallisized bulk, capable of morph the inner structure and apply force and stiffness to the surrounding ...

A similar concept for a macroscopic 2D-structure you can find in 'claytronics' ( [www.cmu.edu] )

Viktor

An errant observation...

Between Mouser, perfboard, soldering iron, and some bus wire, knocking together a circuit by hand is cheap and easy. Add something like BatchPCB and making a "production quality" unit isn't much harder. All the tools fit in a shoebox or two.

OTOH, try making a case that looks anything like an iPod. I got into DIY CNC machining because I wanted to make nice-looking cases and mechanics so that the electronics could do more than beep and flash a bare LED at me. So far this has gotten me a 7x10 lathe, a mini-milling machine, a 12x18 CNC router, and next up a bandsaw. All to make a case and some knobs!

With regard to making cases for electronics that look as good as iPods; have you considered polyester resin embedments?

Considering the reprapable electronics discussion, there is an idea that has recently haunted my mind: If you had a wire deposition toolhead and an extrudable ferrite material, you could reprap core memory and transfluxor logics (that is logic circuitry that works without transistors). The product would almost certainly be more expensive than commercially available ICs, but custom-designed, easy-to-reprap (with only wire-, ferrite- and base material deposition heads and no pick and place mechanism) microchips would be nice, though.

Extrudable ferrite would also take us a step closer to a reprapped stepper motor, if possible at all.