On Sat, 17 Dec 2011 19:52:23 +0100
Michael Haberler mai...@mah.priv.at wrote:
just a vague idea - it would occur to me that transforming the probe
result into a parametric model for the Z surface, feed that to the
trajectory planner and make the TP 'do the right thing' as a move is
executed
actually it probably can go into a kinematics module, like below
I think it could work out as follows:
- take grid of Z probes
- compute surface parameters and pass to kinematics module via HAL pins
- mill unmodified G-code
The advantage would be - beyond 'no gcode translation needed' that it
2011/12/17 Michael Haberler mai...@mah.priv.at:
actually it probably can go into a kinematics module, like below
I think it could work out as follows:
- take grid of Z probes
- compute surface parameters and pass to kinematics module via HAL pins
- mill unmodified G-code
The advantage
On 17 December 2011 20:16, Michael Haberler mai...@mah.priv.at wrote:
actually it probably can go into a kinematics module, like below
Ideally you would pick up 3 fiducial marks and do a full 3D tilt and
rotation compensation.
(There already is a kins which corrects for somebodies twisted
2011/12/17 andy pugh bodge...@gmail.com:
On 17 December 2011 20:16, Michael Haberler mai...@mah.priv.at wrote:
actually it probably can go into a kinematics module, like below
Ideally you would pick up 3 fiducial marks and do a full 3D tilt and
rotation compensation.
(There already is a kins
Am 17.12.2011 um 21:45 schrieb Viesturs Lācis:
Michael,
I like proposals fo solving different tasks in kinematics module :))
I understood everything except that I have no idea, how do You intend
store the probed surface and then feed into kinematics module through
HAL pin.
It seems to me
On 17 December 2011 20:45, Viesturs Lācis viesturs.la...@gmail.com wrote:
It seems to me that creating a mesh from probed points will require to
extend EMC to be able to do so and then there needs to be some tool
that reads that mesh and tells, what is Z in a given X,Y point.
This is
On 17 December 2011 21:01, Michael Haberler mai...@mah.priv.at wrote:
and, of course, come up with a good interpolation algorithm - but there's a
large body of work there; for instance that is exactly the stuff the GIS
folks use for map coloring from discrete digital height models
Linear
2011/12/17 Michael Haberler mai...@mah.priv.at:
remember, its almost 2012, and the interpreter sports full extensibility by
Python methods these days so you're not limited to the RS274NGC pits any more
for extending emc - this is not limited to 'scripted toolchange'
if you get really
i go to the terminal to load rtapi as docs say cd emc2 halrun, then it gives
error consistig of rtapi kernel not loaded. i would include the file but ive
not yet figured out how to save it to send it to the paste bin this install
disk was burned dec 16 thanx jeremy
I am not good with programming, but I work with surfaces in mining
applications all the time.
Why not create an XY grid of the z surface and then load that into EMC much
like one loads in a map for backlash. The grid can be created with as few
as three points or as many as the operator desires.
Thanks for the link. Much appreciated.
I have a machine that was formally a CNC router. The gantry ways are pretty
parallel with the table surface to begin with. The real problem comes in
the pcb mouning to the table. You need to make sure that the PCB does not
bow upwards. I try to pre-flex
Have you seen this? Z-adjustments for non planar surfaces.
http://phk.freebsd.dk/CncPcb/
http://ing.dk/uploads/society/content/201.png
There's a longer thread
herehttp://www.cnczone.com/forums/pcb_milling/82628-cheap_simple_height-probing.html
.
I haven't seen this one before, even though the
2011/12/17 Sven Wesley svenne.d...@gmail.com:
Have you seen this? Z-adjustments for non planar surfaces.
http://phk.freebsd.dk/CncPcb/
Thank You very much!
The solution to isolate PCB and use existing tool in router for
probing is genious!
Since I might need to build a small machine for PCB
2011/12/17 Viesturs Lācis viesturs.la...@gmail.com
Thank You very much!
The solution to isolate PCB and use existing tool in router for
probing is genious!
Since I might need to build a small machine for PCB routing, I have a
question to those that do work on PCBs and would like to test
Here's another interesting little development:
http://www.gabotronics.com/development-boards/xmega-xprotolab.htm
http://www.gabotronics.com/development-boards/xmega-xminilab.htm
Saw these on another mailing list this morning.
Mark
On 17 December 2011 10:43, Viesturs Lācis viesturs.la...@gmail.com wrote:
The extension of the test g-code file is .ngc:
http://phk.freebsd.dk/CncPcb/testpcb.ngc
Are there any other CNC controlling applications that use this file
extension or does it mean that this solution has actually been
well it turns out if you do it exactly as the install exactly as the docs say
it will work. just fyi anyone downloading emc to usb should get the infra
recorder and use it to burn the image as windows defaualt burner will not work
nor will nero9 as it will not burn an image from usb, well i
At the link it says it's emc2 code
On Sat, Dec 17, 2011 at 7:04 AM, andy pugh bodge...@gmail.com wrote:
On 17 December 2011 10:43, Viesturs Lācis viesturs.la...@gmail.com
wrote:
The extension of the test g-code file is .ngc:
http://phk.freebsd.dk/CncPcb/testpcb.ngc
Are there any other
2011/12/17 andy pugh bodge...@gmail.com
That looks like EMC2 G-code to me.
I am _fairly_ sure that the use of the O-word for subs and
conditionals is unique to EMC2.
It _is_ written for EMC.
--
Learn Windows Azure
Does anyone read my lines?! :D
2011/12/17 Eric Keller eekel...@psu.edu
At the link it says it's emc2 code
On Sat, Dec 17, 2011 at 7:04 AM, andy pugh bodge...@gmail.com wrote:
On 17 December 2011 10:43, Viesturs Lācis viesturs.la...@gmail.com
wrote:
The extension of the test g-code
On Saturday, December 17, 2011 12:00:41 PM Sven Wesley did opine:
Does anyone read my lines?! :D
Yippers Sven. In fact I will likely use that code on my converted
micromill next spring to make the pcb's for the opto stuff needed to do my
7x12 lathe. Thanks for the URL.s.
2011/12/17 Eric
2011/12/17 Sven Wesley svenne.d...@gmail.com:
Does anyone read my lines?! :D
I do :))
I went through the cnczone thread, this whole thing looks very
promising, thank You for bringing it in the light!
Viesturs
--
I just found this auction on eBay UK. It looks like a really good spec
for powering a CNC machine. (10A @ 50V, 12A @12V,)
http://www.ebay.co.uk/itm/260901486013
--
atp
The idea that there is no such thing as objective truth is, quite simply, wrong.
just a vague idea - it would occur to me that transforming the probe result
into a parametric model for the Z surface, feed that to the trajectory planner
and make the TP 'do the right thing' as a move is executed would greatly
simplify the usage of this method, and get rid of the G-code
2011/12/17 Michael Haberler mai...@mah.priv.at:
just a vague idea - it would occur to me that transforming the probe result
into a parametric model for the Z surface, feed that to the trajectory
planner and make the TP 'do the right thing' as a move is executed would
greatly simplify the
Viesturs Lācis wrote:
Are there any other CNC controlling applications that use this file
extension or does it mean that this solution has actually been tested
on EMC?
I also noticed that there was absolutely NO MENTION of EMC anywhere in
Poul's
document! But, the .ngc, and the way the
27 matches
Mail list logo
