For recording movements at every 1/2 second intervals is the g-code file that
you play back consist of a new line for each 1/2 second? So that every line of
g-code is a x,y,z, and feedrate value for that 1/2 second interval. If so I
think there would need to be some preprocessing of the G-code
Mark Haynes wrote:
> For recording movements at every 1/2 second intervals is the g-code file that
> you play back consist of a new line for each 1/2 second? So that every line
> of g-code is a x,y,z, and feedrate value for that 1/2 second interval. If so
> I think there would need to be some p
Il giorno dom, 12/02/2012 alle 19.48 -0600, Jon Elson ha scritto:
> Mark Haynes wrote:
> > For recording movements at every 1/2 second intervals is the g-code file
> > that you play back consist of a new line for each 1/2 second? So that every
> > line of g-code is a x,y,z, and feedrate value for
Am 13.02.2012 um 10:14 schrieb Spiderdab:
>>
> I'm sure with my pc i can have an higher recording rate. The problem is
> that when i've tried to record every 1/10 of sec doing rapid 3D
> movements (let's say 1 m/s) the curve is recorded in little lines (one
> every 1/10s..) so when i reproduce th
Il giorno lun, 13/02/2012 alle 11.48 +0100, Michael Haberler ha scritto:
> Am 13.02.2012 um 10:14 schrieb Spiderdab:
> >>
> > I'm sure with my pc i can have an higher recording rate. The problem is
> > that when i've tried to record every 1/10 of sec doing rapid 3D
> > movements (let's say 1 m/s)
hm, I just see the NURBS and quadratic splines in linuxcnc work in X/Y only, so
that isnt terribly useful for you
what you could do is - take samples, derive a spline, and approximate as
biarcs; those would work in x/y/z
a starting point: http://emergent.unpythonic.net/01171767993
-mah
Am 13.0
on my haas mashines the pin is connected to the source voltage and
the part is GND Isolated from the mashine .
http://www.google.de/url?sa=t&rct=j&q=heidenhain%20cyclen&source=web&cd=3&ved=0CDEQFjAC&url=http%3A%2F%2Fwww.heidenhain.de%2Ffileadmin%2Fpdb%2Fmedia%2Fimg%2F636_220-11.pdf&ei=7f84T__9G
Spiderdab wrote:
>
> I'm sure with my pc i can have an higher recording rate. The problem is
> that when i've tried to record every 1/10 of sec doing rapid 3D
> movements (let's say 1 m/s) the curve is recorded in little lines (one
> every 1/10s..) so when i reproduce the movement, the system canno
Il giorno lun, 13/02/2012 alle 11.12 -0600, Jon Elson ha scritto:
> Spiderdab wrote:
> >
> > I'm sure with my pc i can have an higher recording rate. The problem is
> > that when i've tried to record every 1/10 of sec doing rapid 3D
> > movements (let's say 1 m/s) the curve is recorded in little li
Spiderdab wrote:
> Yes I've tried using G64, but maybe i didn't know how this worked.
> because i didn't have the expected resulting.
>
> Just because you mentioned, is it correct to use such a line at the
> beginning of the g-code file:
> G64 Pp Qq
>
> in which p is the maximum distance admitted f
Il giorno lun, 13/02/2012 alle 20.23 -0600, Jon Elson ha scritto:
> Spiderdab wrote:
> > Yes I've tried using G64, but maybe i didn't know how this worked.
> > because i didn't have the expected resulting.
> >
> > Just because you mentioned, is it correct to use such a line at the
> > beginning of
W dniu 14.02.2012 14:28, Spiderdab pisze:
> and, I really cannot understand what Q does (naive) and if you have to
> use Q together with P or without.
>
P value is maximum deviation of path allowed.
Q value:
let say we have two parallel continous lines. we have Pointa - begin of
1 line, pointb (en
Spiderdab wrote:
>>
>> It is G64.1 Pxxx. G64 turns on some smoothing, G64.1 turns on more (and I'm
>> vastly oversimplifying it.) Well, maybe they have changed it, but I'm
>> pretty sure
>> I used G64.1 Pxxx, but looking at section 3.1.3 it doesn't say that.
>> http://www.linuxcnc.org/docs/EMC2_
Michał Geszkiewicz wrote:
> W dniu 14.02.2012 14:28, Spiderdab pisze:
>
>> and, I really cannot understand what Q does (naive) and if you have to
>> use Q together with P or without.
>>
>>
> P value is maximum deviation of path allowed.
> Q value:
> let say we have two parallel continous li
Il giorno mar, 14/02/2012 alle 12.04 -0600, Jon Elson ha scritto:
> Michał Geszkiewicz wrote:
> > W dniu 14.02.2012 14:28, Spiderdab pisze:
> >
> >> and, I really cannot understand what Q does (naive) and if you have to
> >> use Q together with P or without.
> >>
> >>
> > P value is maximum
Il giorno mar, 14/02/2012 alle 12.02 -0600, Jon Elson ha scritto:
> Spiderdab wrote:
> >>
> >> It is G64.1 Pxxx. G64 turns on some smoothing, G64.1 turns on more (and
> >> I'm
> >> vastly oversimplifying it.) Well, maybe they have changed it, but I'm
> >> pretty sure
> >> I used G64.1 Pxxx, but
2012/2/14 Spiderdab <77...@tiscali.it>:
>
> what if (talking about mm units) i exagerate the P
> value to..let's say 1000 mm, in a 'spline' made of a thousand of 1mm
> lines? What the machine is expected to do?
>
It would do the shortest and fastest move from start to finish, taking
into account t
On 14 February 2012 20:17, Viesturs Lācis wrote:
> 2012/2/14 Spiderdab <77...@tiscali.it>:
>> what if (talking about mm units) i exagerate the P
>> value to..let's say 1000 mm,
> It would do the shortest and fastest move from start to finish,
I am not sure about that, I seem to recall reading h
Spiderdab wrote:
>
> Thanks for this explaination. in my case i'll have only line-line
> movements, since the way that i'm recording movements is point-by-point.
> so i don't have need of Q.
> speaking for absurd...what if (talking about mm units) i exagerate the P
> value to..let's say 1000 mm, in
Viesturs Lācis wrote:
>
> Or even better You could ask Jon to share his test file and see a
> working implementation of G64 Pxx command.
>
My test file is at http://pico-systems.com/codes/contour.ngc
it is 10,000 G1 vectors that do a 2.0 diameter circle at a feedrate
of 30 units. You might want
andy pugh wrote:
> I am not sure about that, I seem to recall reading here that it will
> still touch at least one point of each line segment.
>
Activating the naive cam detector with Qxx (IIUC) will actually remove
some line
segments, if their deviation from a straight line is less than the
p
21 matches
Mail list logo
