Re: PCB milling - getting started
|
Glad to hear things worked out. Care to share some pictures?. I'm planing on spending some money on a CNC. Not sure if I'll be doing milling, as it seems to deliver mixed results. I plan to use it for drilling and maybe a laser to expose the traces. Bruno Basto Software Engineer - Brazil Liferay, Inc. Enterprise. Open Source. For Life.
On Sun, May 12, 2019 at 5:44 PM Mark Pilant < mark@...> wrote: After finally getting to spend some more time, I have a good PCB :-)
I did encounter one minor anomaly when "drilling" but it was easy to
work around.? (The anomaly was the drill file generated from the Kicad
drill file had a different reference origin than the PCB.? I'm not sure
why, more investigation needed.)? I also didn't actually drill, but
rather just used the V bit to "center punch" where the drilling was
actually going to be done.
Here are some of the milling information I finally used:
? ?Bit: 0.1mm 30 degree "V" bit
? ?Spindle speed: about 8000 RPM
? ?Depth of cut: .4mm (0.0078")
? ?Feed: 50mm / min
After all the milling / drilling I took down all the sharp edges left
on the traces with some very fine (400 grit or so) sand paper.? After
that I used a small brass bristle brush to clean out anything left in
the milled groves.
Time for drilling and soldering :-)
Here are some of the things I learned.
Making single sided PCBs, with the traces on the "back", is not all
together straight forward.? Although it is not all that difficult to do.
Larger pieces are more prone to flexing than smaller pieces.? From the
same stock, some 30mm x 60mm milled without problems, but a larger board
(65mm x 130mm) had sufficient flexing to have problems with auto-leveling.
The PCB stock *really* needs to be securely fastened to the table.
Not only to eliminate X & Y axis shifting, but to eliminate / minimize
the PCB stock flexing for Z axis cutting.
Auto-leveling is really important.? Fractions of an inch/mm matter.
Not only might the table not be exactly parallel to the Z guide rods
and drive screw, but it may be possible to have PCB stock of varying
(but small) thickness.
Auto-leveling may not be able to address flexing.? As the probe is used
for auto-leveling, not enough force is applied to cause the stock to flex.
(As would be expected.)? However, milling may exert enough force to cause
the stock to flex and the bit to ride up / skate on the copper.? Using an
end mill seems to be more likely than a V milling bit to ride up / skate
on the copper.
I hope this helps others in the future.
- Mark
|
