Homebrew PCBs

Richard,

Based on what you've written here, I'd suggest avidly pursuing the 
printing of flux. That approach seems the most likely to yield decent 
results. Solder dust printed to a board without flux is gonna just 
run all over the place, IMO.

Oooorrrr...

What about using a laserjet and having the toner--which we already 
know is thermoplastic-- become the binder for the solder/copper 
powder??

Seems to me that if we printed trace outlines directly to the BARE 
FRP board--or even ironed it on! after printing to paper-- and then 
put a layer of solder/copper dust on top... That if that went into 
the oven(set for toner melt temp), the toner would bind the 
solder/copper to the board. 

Since the toner melts at a lower temperature than the solder copper...

The NEXT step is to brush away the solder copper dust-- which is not 
sticking anywhere EXCEPT where the toner was-- and then put the board 
into a hotter oven for reflow OF the solder/copper!

Just might work.

Ballendo

P.S. I have several samples from the Zcorp RP machine. Got 'em at 
last year's Westec show. One is a ball bearing, workable, all printed 
at one time. Pretty cool, but not nearly as durable as the other RP 
processes. In times of high humidity the parts tend to crumble.   


--- In Homebrew_PCBs@yahoogroups.com, Richard Mustakos 
<rmustakos@a...> wrote:
> Hey,  I was just surfing, and ended up at a site for a 3D prototype 
company.
> <http://www.zcorp.com> (interesting in it's own right)
> They use an inkjet printer to print a liquid that binds plaster 
dust 
> together to
> make their models.  Sounds almost like regular inkjet except they 
add 
> another
> set of nozzles that prints clear, and always output the same amount 
of 
> liquid.
> They print with a full color inkjet head and give full color 3d 
parts.
> But I digress.
> Direct printing of a pcb, version 0.01
>   Make the inkjet head transform in the Y, as well as X axis.
>   Put a piece of fiber-glass on a flat bed.  
>   Spread powdered solder on the board to an even level.  
>   Print on it, wetting the powder where you want a trace.
>   Dump off the dry powder.  
>   Put it in a toaster and reflow the stuff the ink kept in place.
>   If you have surface mount components, put them in place prior to 
the 
> reflow.
> 
> I think the resolution would be controlled by the fineness of the 
grind 
> on the solder
> more than the resolution of the printer.
> 
> I suspect there are issues with:
>   grinding the solder fine enough
>     (not an issue - it's manufactured to the tiny as s--t level)
>   wetting the powder enough with a regular inkjet
>     (how thick does powder layer need to be?)
>   keeping the wet powder right where you want it
>     (need something that wets the board as well as the solder)
>   getting the reflowed solder to stay where you want it
>     (would some kind of flux work as a wetting agent and keep the
>       reflowed solder from running or balling? means printing flux)
> 
> The Dow Corning PI-1000 copper based conductive ink is made with 
copper
> powder and solder powder.  The curing process has two purposes:
>   1) it actually cures the binder
>   2) it melts the solder powder so that it produces the copper to 
copper 
> connection
> 
>   Using copper keeps the resistance down, and is exactly what we 
would want.
> Using the solder binds everything, just like we would want.  I 
don't 
> think the fiber
> glass would take the temperatures required to 'reflow' the copper 
powder 
> by itself,
> (not to mention my toaster won't, either) so the solder powder is 
required.
>   A quick look showed a number of copper powder manufacturers, and 
a solder
> paste manufacturing uses solder powders.  I'll try and get more 
> information on the
> materials next week.  it's tough to do on the weekends.
>  Richard