[Noisebridge-discuss] CNC Shop Class

[Jonathan Foote]

Progress!

Last night we spent some time experimenting with acrylic and the
smallest end tool (0.031 inches) to do some engraving and mock-PCB
routing. Things came out pretty well for a first try! Examples are by
the mill in the space. I got toolchains working from EagleCAD and
Inkscape, the latter with some glitches. Still need to hack the gcode
a little, details on the wiki. Went as far as I could with the PCB
routing -- basically to the limits of our existing tools, so we need
to get some smaller/better end mills and some small drills for PCB
work.

Really feeling like we're close to being able to do useful things
(circuit boards, sign etching) at this point!

Next steps:

* Still need to get the limit switches working. Means digging a little
deeper into HAL documentation.

* Better method of homing Z axis. I think there's a special
height-measuring tool for that?

* Mike noticed some play in the X axis bearing. I think that's an easy
adjustment but needs small hex wrenches, will try to remember to bring
mine in. Inch not metric I think.

* It would be great if we could run the EMC2/AXIS code in simulation
mode, so you wouldn't need a VM to run it. Right now it checks for the
rtai kernel and craps out if it can't find it. Any gurus want to
tackle that? It may mean recompiling, but hopefully it's just a config
option someplace I couldn't find. This will let anyone tune up their
gcode  and dry run sessions without tying up the machine.

* more GCode toolchains. QCAD(Solidworks?)->DXF->GC?
Inkscape->SVG->GNUCAM->GC? Blender->?->GC. If we had simulation mode,
anyone could do this on their own machines.

* EagleCAD header options (it needs to set spindle speed) and possibly
design rules for DRC check (depends on tools).

* Figure out tool width compensation stuff in tool.tbl. Then we can
try cutting some things so we can get a start on good feed rates and
spindle speeds. Can we increase traverse velocity?

* Get a set of better tools and clamps.

* Dust maintenance. Shower curtain over table to keep it out of PC
underneath? Dedicated dustbuster/small vac? Build an enclosure?


Anything else?




On Fri, Oct 9, 2009 at 3:34 PM, Michael Wright <mike at smallip.com> wrote:
> The Mill is operational in a safety 5th sort of way.  This means it will in
> fact cut patterns from g-code, at the moment it will only do that with the
> spindle running at full speed any time it's plugged in, though I'm planning
> to fix that first.
> I consider it safety 5th because it doesn't have travel limits like the
> manufacturer intended.  The manufacturer relied on software to keep the
> travel in range, we're not using their software and haven't sorted out how
> to make EMC2 do that (or a hardware solution).  In short we need to be
> watching it all the time, and only run things near the middle of it's ranges
> (so we have time to turn it off).
> If it appears to be getting out of hand, please is turn it's power switch
> off (the one on the side of the control box).  Note however that after doing
> this you'll probably need to restart the software because it will be out of
> sync with the controller (it gets out of sync).
> If it only looks like it might be getting out of hand, the F1 key should
> toggle Emergency Stop in the software (which at present won't stop the
> spindle).
> In terms of learning CNC, EMC2 has a really really nice simulator (well,
> actually it's a 3D representations of what it thinks the machine it's
> controlling is doing).  It runs reasonably well on VM-Ware (even on my
> little MSI Netbook).  If anyone wants a copy of the VM I use it's about 6.4
> GB (alternatively you can just download the .ISO and make your own.  It
> seems to run much better installed on the VM hard disk with 2 virtual cores
> than it did running diskless (from the .ISO file).  Currently it has 700 MB
> available and is using 360.
> Effectively this means anyone can sort out their G-Code file before ever
> touching the mill (or while someone else is running a different job on the
> mill).
> I'm defining "getting that g****d mill to work" as getting it to where EMC2
> accepts G-Code.  I think Jon has done some work on setting up a tool-chain
> and put links on the wiki.  It would be wonderful if someone sorted out a
> tool-chain and possibly set it up to run on something other than the Mill
> controller at NB.
>
> mike
> On Oct 8, 2009, at 6:43 PM, Shannon Lee wrote:
>
> I believe the shop class tomorrow will address this particular area of
> idiocy (I am hoping so).  Whether it will increase or decrease the general
> level of CNC Idiocy remains to be seen.
>
> On Thu, Oct 8, 2009 at 6:35 PM, Al Billings <albill at openbuddha.com> wrote:
>>
>> So I know nothing about CNC stuff but have seen all the traffic about
>> the mill. Will the shop class tomorrow night rectify my idiocy (at
>> least in this limited area)?
>>
>> Al
>>
>> _______________________________________________
>> Noisebridge-discuss mailing list
>> Noisebridge-discuss at lists.noisebridge.net
>> https://www.noisebridge.net/mailman/listinfo/noisebridge-discuss
>
>
>
> --
> Shannon Lee
> (503) 539-3700
>
> "Any sufficiently analyzed magic is indistinguishable from science."
> _______________________________________________
> Noisebridge-discuss mailing list
> Noisebridge-discuss at lists.noisebridge.net
> https://www.noisebridge.net/mailman/listinfo/noisebridge-discuss
>
>
> _______________________________________________
> Noisebridge-discuss mailing list
> Noisebridge-discuss at lists.noisebridge.net
> https://www.noisebridge.net/mailman/listinfo/noisebridge-discuss
>
>