Homebrew PCBs

.....also , If you plan to use your software "auto-routing" you must inform the software(simply by creating a new "micro-hybrid" component for your component library. This way the auto-router knows NOT to treat the SMDs as discretes , but as a hybrid lumped module.



--- In Homebrew_PCBs@yahoogroups.com, "Richard" <richard.liberatoscioli@...> wrote:
>
> Hi Andrew!
> 
> I'm not sure exactly what your saying.
> 
> First, did this make sense to you?:
> http://tech.groups.yahoo.com/group/Homebrew_PCBs/message/25979
> 
> Basically, what I'm doing is making micro "hybrids" out of the top LCR circuitry before I commit them to bottom layer routing.
> 
> IE: a top layer 16 point(connection)circuit then becomes only 4 points of bottom layer(connections) Your PCB placement and routing software now sees the this "hybrid lump" as only 4 points of connections.
> 
> In other words, this method is not just one routing algorithm, but divided into two, a separate "top layer optimization" and a separate "bottom layer optimization". 
> 
> Also, Inkjetting the "muti-layer bottom layers"(conductive and insulating) affords a unique optimization in routing in that you can now "spot mask" areas without masking the entire layer.
> 
> If you dont understand what I'm saying here, I can give you a link to my webmeeting for a audio/visual explanation of this.
> 
> 
> 
> 
> --- In Homebrew_PCBs@yahoogroups.com, Andrew Villeneuve <andrewmv@> wrote:
> >
> > How often can this be practical? It seems to me that the vast majority of
> > signals would not happen to align with the physical pin you needed in order
> > to attach components side by side. You'd need to space them out and put
> > small traces.
> > 
> > I have seen some clever IC configurations along these lines, though - I've
> > once seen on a production board, SMT memory ICs literally stacked atop one
> > another, all the top pins soldered directly to the bottom pins, with just
> > the chip-enable and write-enable signals broken out independently.
> > 
> > -Andrew
> > 
> > On Tue, May 25, 2010 at 10:28 AM, Richard
> > <richard.liberatoscioli@>wrote:
> > 
> > >
> > >
> > > .........Yes!
> > >
> > > Any RCL electronically grouped is "physically grouped" into a very tight
> > > formation during the "pick and place"
> > >
> > 
> > 
> > [Non-text portions of this message have been removed]
> >
>