[Coco] Multipak redesign/replacement

[Gene Heskett]

On Monday, February 23, 2015 10:16:31 AM Ronivon Costa wrote:
> And the outsourcing pcb print companies are happy to receive KiCAD
> files? (osh-park...)
> 
> On 23 February 2015 at 15:05, Chris Osborn <fozztexx at fozztexx.com> 
wrote:
> > On Feb 22, 2015, at 10:55 PM, Gene Heskett <gheskett at wdtv.com> wrote:
> > > I have no clue if pcb-gcode can deal with its data format to make
> > > gcode out of its finished board file either.
> > 
> > KiCAD is most definitely for making schematics and PCBs, and is not
> > an architectural CAD program. I've used it many times to make PCBs
> > on my CNC. I definitely recommend switch to KiCAD and abandoning
> > Eagle. KiCAD is free, you can make a board any size you want. It's
> > also open source so the community that is growing around it can't be
> > shut out by a proprietary company.
> > 
> > I've got several article on my blog about using KiCAD to mill PCBs:
> > 
> > http://www.insentricity.com/search.cl?q=kicad
> > 
> > --
> > Follow me on twitter: @FozzTexx
> > Check out my blog: http://insentricity.com

Neat link, thanks. Your depth problems begin with how the dremel (a very 
poor choice) is being mounted around its body.  But even mounted by its 
nose nut, it will grow with the heat, either telegraphing out of the 
bearing, or down the shaft from the motors armature.  For various reasons 
including much enhanced xy accuracy from its solidly mounted to the 
armature chuck, the Proxon tool is much the better choice.

But mount it by its nose.  Flattening the bed is a large problem, one that 
would have me replacing it with a 1/2" thick alu sheet at some point, so 
that once it was flattened, it would not need to be done again for many 
moons.

An air bearing and air driven dental tool that spins 100k revs would be 
even better for this as the air in the bearings also serves as an 
excellent collant. Or even some refrigeration effect from the adiabetic 
expansion in the turbine that drives it.

My somewhat larger xy table & post mill also suffers, but I've found that 
a fan blowing on the post area, started about an hour ahead of the job 
start, serves to equalize the temps and keep it from growing near as much. 
As my Z drive is anchored about 2 feet above the table, there is a lot of 
growth in the drive screw as it passes by and absorbs heat from the 
spindle motor, and the spindle itself may even grow with the heat, so 
enough air moving to keep it all near the same temperature helps a lot. So 
does ball screws, but I have not yet converted my z drive, its a 1/2" nook 
acme thread, double nutted to keep backlash in the 2 thou max range.  I 
also have an autozero function that can be edited into the gcode file 
which will repeat that compensation during the actual milling operation, 
but so far have only plugged it in after a tool change because my collets, 
M2 taper, have no zeroing stop when changing tools. I use tools with the 
stop collar, but thats nowhere near close enough in the real world.

These g-code snippets work well with linuxcnc, allowing me to drill a hole 
halfway thr the board, turn the board over and do the other side, and meet 
in the middle with an unmeasurably small error, looks like one hole all 
the way thru.

But they may need things that mach may not have.  You don't say that I 
caught, what machine driver software you are using. I would also point out 
the pcb2cgode does no longer work with eagle, the much better pcb-gcode 
does. But I've no clue if it works with kicad either, its a bridge my 
horses haven't had to cross yet. ;)

Cheers, Gene Heskett
-- 
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
Genes Web page <http://geneslinuxbox.net:6309/gene>