> Python3 is available on all supported OS versions.
Is it safe to say just Python3, or do we need to specify a version - the python
download page offers 3.6, 3.7, 3.8, 3.9 with all different End of Support
dates. I used some software the other day that would only work with 3.8; not
3.7 and no
Doxygen can do most of this. It can generate interactive html from source
code. If you format your comments the way doxygen likes, they appear as
part of the documentation.
It's also free.
-Original Message-
From: Stuart Stevenson [mailto:stus...@gmail.com]
Sent: Thursday, 29 April 2
What I'm seeing (and I'm not a linuxcnc expert) looks correct. After the
homing sequence it jumps to HOME. This is documented here (0.1.3.7)
http://linuxcnc.org/docs/2.4/html/config_ini_homing.html
-Original Message-
From: Feral Engineer [mailto:theferalengin...@gmail.com]
Sent: Sund
> Under what sort of circumstances would you have different axis and joint
> limits?
A robot arm will have joints and axis that completely different.
___
Emc-developers mailing list
Emc-developers@lists.sourceforge.net
https://lists.sourceforge.net/
> My question:
>
> Will using the framebuffer (instead of X) make a better environment for
> LinuxCNC?
For low spec. computers, not having the overhead of X, is a bonus. You'll
also boot faster into linuxcnc.
But, you won't be able to run other applications or utilities. And you'll
need to w
John Thornton [mailto:j...@gnipsel.com]
Sent: Thursday, 27 February 2020 10:35 AM
To: emc-developers@lists.sourceforge.net
Subject: Re: [Emc-developers] Touchy Tweak?
That is a good idea! Did you get a touch screen sized buttons on your popup?
JT
On 2/26/2020 3:16 PM, Frank Tkalcevic wrote:
&g
I did that on my version of touchy. The shutdown button pops up a dialog and
gives 3 options - shutdown, restart, or close (and cancel the dialog). They
just exit touchy and return 0,1 or 2 to the calling script (to do a clean
linuxcnc exit). The script then executes the shutdown or restart.
I'm trying to get the source for version 2.6.0~pre
I tried.
git checkout tags/v2.6-pre0
but the VERSION file shows 2.5.0~pre
output.
> git checkout tags/v2.6.0-pre0
Note: checking out 'tags/v2.6.0-pre0'.
You are in 'detached HEAD' state. You can look around, make experim
I'm using the latest master.
If I get an InverseKinematics failure (I'm playing with this at the moment),
the power to the machine is cut (the power button on axis pops), but the
motors are left enabled state, ie joint.0.amp-enable-out remains TRUE.
It is a bit worse than that, because e-st
In the Updating LinuxCNC documentation, section 12.2.2 Homing says
Ø A homing api is provided by src/emc/motion/homing.ini to support users'
custom homing code in out-of-tree homing.c source files.
What is this? Where is this?
___
Emc-de
rs
Subject: Re: [Emc-developers] Homing robot arm differential joint [ was robot
arm, free to a good home]
On Mon, 25 Mar 2019 at 21:10, Frank Tkalcevic
wrote:
> I'm not sure how to home these joints. To find the 2 home positions (bend
> and rotate) requires moving both motors at
> I have a Scorbot ER-3 robot arm that I will never get around to playing
> with.� It works with LinuxCNC, though there are some bugs in the
> kinematics still.
Slightly off topic...
I'm building a robot arm that has differential joints in the elbow and wrist,
similar to the scorbot's wrist.
I have a cheap ebay $100 probe. I had doubts about the accuracy, so I
perfomed a test probing a precision disc from all orientations. (The disc
came from a 1"-2" micrometer - it is used to "zero" the micrometer).
The results were pretty poor, with the error ranging from 0.3mm to 0.75mm.
http://
so my
program works.
Can you confirm these?
regards
bkt
Il giorno lun 3 dic 2018 alle ore 20:20 Frank Tkalcevic <
fr...@franksworkshop.com.au> ha scritto:
> Just confirming, which line do you still get the error? What is the
error?
>
>
> Here are some things you can check...
>
alid or == to EMC_STAT_TYPE ... but is possible to have my
emcStatus->motion.traj.position.tran.x,
emcStatus->motion.traj.position.tran.y,
emcStatus->motion.traj.position.tran.z HEUREKA!!
actually I not know why I have these situation ... I've used your emc.nml
file without h
This is what I use on mine...
https://github.com/ftkalcevic/CNCConfig/blob/master/9x20lathe/emc.nml
I've added entries for my own process, hidcomp. The linuxcnc I'm running is a
bit old, so I don't know if the buffer sizes are correct.
-Original Message-
From: theman whosoldtheworld
The arrays are num_joints+1 in size, but the loops only initialise
num_joints members. The extra code initialises some of the last element.
That's what the code does, but doesn't explain why it does and what the
extra array element is used for.
On Sat, Sep 15, 2018 at 9:01 PM, andy pugh wrote:
> That's good. Can we also reduce the amount of depend packages determined
> by "dpkg-checkbuilddeps"? If we don't need a GUI locally, I think most
> of them are not necessary.
I don't really know what dpkg-checkbuilddeps does. If it lists the packages
required to build linuxcnc, then I'd say n
> > If you don't want a GUI do you want a G-code interpreter, or are you
> > planning on using just HAL?
> HAL, EMCMOT, EMCIO, EMCTASK, NML-SERVER should be on the small board. A
> GUI will work via network. I think that should avoid installation of
Axis can run via standard X-Windows. That's h
> > I had the same problem. A half solution was to check the size of the
> > structures eg sizeof(EMC_STAT). I only need to know if a binary has
> > changed, and stop when there was a mismatch.
> I wish there was a way to do that at runtime. As far as I know that trick
> only works at compile
I had the same problem. A half solution was to check the size of the
structures eg sizeof(EMC_STAT). I only need to know if a binary has
changed, and stop when there was a mismatch.
-Original Message-
From: Les Newell [mailto:les.new...@fastmail.co.uk]
Sent: Wednesday, 21 December 2016
> (The summary of those bugs is: the serial number method for UIs to ensure
a
> message is acted upon by task simply does not work when there is not just
> one UI. No simple modification will make it work, and no one in the
project
> in 10+ years has taken the time to understand NML with the depth
I agree. It can be really tedious. I spent a lot of time to get the model
right.
I played with OpenRave and got that to display my robot arm. You can import
CAD models and then connect it to linuxcnc with a bit of python...
http://wiki.linuxcnc.org/cgi-bin/wiki.pl?Ikfast (
You don't need to c
You may want to look at vismach in the source tree. It is a standalone window
that shows the geometry of your machine (and plot too - from memory). There is
a 5 axis example there.
> -Original Message-
> From: David Armstrong [mailto:cncbas...@gmail.com]
> Sent: Tuesday, 23 September 2
As has already been mentioned, you can whatever you want with as many
threads as you want. It is up to you to manage them and how they interface
with whatever you interface.
My VFD Modbus component is a user hal component that spawns a thread to do
the USB rs485 comms. I have to manage how my th
> Minor complexity: we have a freaky result due to
> http://www.linuxcnc.org/docs/html/gcode/overview.html#sec:Order-of-
> Execution
>
> consider:
> G20
> F5
> G1 X1
> G21 F125 X100
>
> according to NGC the F125 is taken as inches while the destination is
taken as
> mm! I hate this special case
My understanding is the size is basically the size of the shared memory
block. It needs to be as big as the largest message that is sent between
processes. Since linuxcnc works with a queue depth of 1, increasing above
that max msg size (15k I think) does nothing. Go below the max message
size,
> I guess I would first ask if the licenses are compatible. If so, then I
am sure
> you are good., If not, then you have to find a path through that
minefield.
Ouch, I never considered that. I'd hate not to be able to share back to the
linuxcnc community.
> I would say you do not understand a
infrastructure, and I’ve always had them in
the back of my mind as to how they would affect my code.
Frank
From: Michael Haberler [mailto:m...@mah.priv.at]
Sent: Wednesday, 26 March 2014 9:33 PM
To: Frank Tkalcevic
Subject: Re: [Emc-developers] GCode preview in C++ (Qt)
Hi Frank
I've just tried to add a library to one of the submakefile. The first time
I tried to make, my files compiled, but with errors. I fixed a few errors,
recompiled and NOTHING! For next 3 hours I moved my files, renamed them,
created a subdirectory, delete extraneous commands from the submakefile,
ut you can often get it down to 3x if you use numpy.
THere
> are some other tricks I have read about, but not tried that might speed it
up a
> bit more than that. Probably what Andy needs are python bindings (even if
it
> is written in another language).
>
>EBo --
>
>
e-
> From: andy pugh [mailto:bodge...@gmail.com]
> Sent: Wednesday, 26 March 2014 12:18 AM
> To: EMC developers
> Subject: Re: [Emc-developers] GCode preview in C++ (Qt)
>
> On 25 March 2014 06:00, Frank Tkalcevic
> wrote:
> > I want to add a linuxcnc gcode preview
I want to add a linuxcnc gcode preview to my C++ Qt application. So the
first question is, has anyone already done this, so I can reuse the code?
Failing that, I've been looking at implementing it myself. I can see some
examples such as, gcodemodule.cc and saicanon.cc, which seem to just
imp
just wrote a standalone server.
> -Original Message-
> From: Joachim Franek [mailto:joachim.fra...@pibf.de]
> Sent: Saturday, 8 March 2014 8:13 PM
> To: EMC developers
> Subject: Re: [Emc-developers] Board with IEEE 1588 PTP
>
>
> On 07.03.2014 23:16, Frank T
Cool. Do you have any plans for it?
> -Original Message-
> From: Joachim Franek [mailto:joachim.fra...@pibf.de]
> Sent: Saturday, 8 March 2014 5:35 AM
> To: emc-developers@lists.sourceforge.net
> Subject: [Emc-developers] Board with IEEE 1588 PTP
>
>
> ht tp:/ /www.ti.com/ ww/en/launch
I started writing a QT gui so that I could build a display for a 400x240
LCD. The display is defined by using an XML configuration file loosely
based on the available QT widgets, so I have a few different display
layouts. It's not much more than a proof of concept at the moment, with
only some o
For those that are interested, I managed to get ikfast working with
linuxcnc.
I've created an ikfast output generator to produce C code that can be
compiled as a realtime object. Closed form solutions don't link with
external libraries and are very fast. Numerical solutions use the lapack
lib
Is it possible to issue MDI or equivalent commands in joint mode? For
example, on my robot arm, I want to move joint 2 to 45 degrees.
I'm guessing no, because you have to be in MDI mode to issue MDI commands
and you can only be in MDI mode when in world mode.
If this is the case, are there
Any 3d printed file will divide up curves into short line segments. Here's
an example...
www.franksworkshop.com.au/img_bin/tux.zip
There's a few 3d printer M-Codes you may need to remove.
> -Original Message-
> From: Robert Ellenberg [mailto:rwe...@gmail.com]
> Sent: Friday, 6 Decembe
In my kinematics module, when I do a real time build, my rtapi_print
statements are not print correctly. The calls.
rtapi_print("ikfastkins kinematicsForward\n");
rtapi_print("ikfastkins::kinematicsForward(joints: %f %f %f %f %f %f - ",
joint[0],joint[1],joint[2],joint[3],joint[4],joint[5]);
A section on using Linuxcnc with GitHub would be nice. That is, if using
github is a good idea. I forked linuxcnc a year or 2 ago and have been
using it through my github. I push my changes there so they are all public,
but my fork is 2 years behind. I don't know if what I 'm doing is correct.
I managed to convert the generated C++ to C, and get that working in the
simulator (for Translate3D and TranslationXAxisAngle4D).
Then I created an openrave module to generate the C code. That worked in
the simulator.
Now I've moved from my Vmware 12.04 simulator, to a 10.04 atom box trying to
b
What strategies, branching, pushing, pulling, etc do people use for linuxcnc
development?
I have 4 linuxcnc setups, all with small tweaks to the source. These can
probably be all on the same branch because the changes are independent.
Should I create a branch to keep my changes separate, or sh
> > warning: the frame size of 117088 bytes is larger than 2560 bytes
> > [-Wframe-larger-than=]
>
> it's from src/Makefile:
> Makefile:EXTRA_CFLAGS += $(call cc-option,-Wframe-larger-than=2560)
> $(call cc-option,-Wno-declaration-after-statement)
>
> this limitation comes from the limited kerne
> ikfast got my attention recently and I looked for someone who could help
> me make custom kinematics for 6 axis articulated robot. Can I ask you to
> help find the way to do it with ikfast?
It was your discussion that got me interested in ikfast.
I don't think it is going to be simple. I've
> I think the Makefile rules will not build a RT module out of C++ code
since the kernel environment is not geared for C++ runtime support
>
> any chance you recode this into plain C ?
I've managed to get it working in the simulator by hacking some rules in the
Submakefile. I'm playing with ikfa
I'm trying to add a new C++ kinematics module. If I just put my .cc file in
the src/emc/kinematics directory, nothing happens.
I looked in src/Makefile and found these lines (I'm building the simulator
at the moment).
$(sort $(RTOBJS)) : objects/rt%.o : %.c
$(ECHO) Compiling realtime
I have a similar setup, but replace "MAC" with "Windows PC".
Any serious development is done in a virtual machine on my PC. All my
linuxcnc boxes are too slow (and remotely connected) to provide a nice
development experience. I use the simulator build of linuxcnc. Only after
I think I have fini
Isn't this just the fact that the buffers can't be shared? If you want a
command buffer don't you need to edit the linuxcnc.nml file and allocate a
buffer for you process with a unique name?
I had a problem where I was using Axis + Linuxcncrsh. After running for a
couple of hours my job stopped
> Clearly nobody is using it lately :)
I use it for my lathe and haven't noticed any problems.
> I was thinking that if this was embeddable in the screen and could access
the
> python script style wizards that it would be a very handy winner.
I was thinking the opposite. It would be nice if
> > displays GCode. It uses gremlin to embed a preview, but you can't
> > interact with it unless you use a mouse or touchscreen.
>
> In what way do you want to interact with it?
> Some things I have done: zooming, changing views, toggling the DRO on and
> off.
> I used separate GUI buttons to do
I've been building pendants that use small 3.2" LCD displays.
My first version just used a simple command based interface (fill rectangle,
draw straight line, draw text), driven by a python menu add on, which runs
in parallel with Axis.
(http://www.franksworkshop.com.au/CNC/LathePendant/LathePen
xperimental web front-end for localizing
> LinuxCNC
>
> On Mon, Mar 12, 2012 at 12:05:45PM +1100, Frank Tkalcevic wrote:
> > Is this why when I just hit www.linuxcnc_com, it came up in Russian?
>
> No. (our website
Is this why when I just hit www.linuxcnc.com, it came up in Russian?
> -Original Message-
> From: Jeff Epler [mailto:jep...@unpythonic.net]
> Sent: Sunday, 11 March 2012 1:10 PM
> To: emc-developers@lists.sourceforge.net
> Subject: [Emc-developers] Experimental web front-end for localizi
There are two ways to reparent - pass the window id to the child window so
it can embed itself, or grab the child window id and force the embedding.
I'm playing with Qt at the moment. I created a gremlin-embed.py to use
method 1. I also embedded vismach using the second method - vismach uses
tcl
> > Second question, can this "main" repository be set up on github? Given
> > that github is free, it seems like a good place to store it.
> > Any idea how? I've read about "fork" on github, but that only seems
> > to apply to repositories hosted on github.
>
> right
>
> just create an empty em
I'm new to git so I may have completely misunderstood what git can do,
but...
I have more than one linuxcnc machine now, and they all have customisations
to the source.
As I understand it, I should be able to up my own "main" repository clone of
linuxcnc. Then, each of my cnc machines can then c
I'm sure this has been described by one of the emails, but my inbox is
flooded at the moment.
>From a source code point of view, what's changing in the rebranding?
Is it just the external application names and text? The libraries, eg
libemcini.so? The internal structures, EMC_TASK_STAT?
I just
h emc.stat.tooltable (old and new tool)
>
> can you post the Python code?
> -m
> Am 06.11.2011 um 07:09 schrieb Frank Tkalcevic:
>
> > I've just updated my source to the latest in master, and I'm trying to
> > run a python GUI app, that uses the emc.stat.tooltable memb
I've just updated my source to the latest in master, and I'm trying to run a
python GUI app, that uses the emc.stat.tooltable member to list all the
tools in the tool table. However, the list is empty.
I can still run axis, and edit the tool table from the menu, so everything
seems to be in the r
I posted this in the emc-users group because I thought it was a
configuration problem, but it appears to be a software issue.
This only happens on a gantry system.
When axis starts, emc is in EMC_TRAJ_MODE_FREE, meaning all axis can be
jogged freely. axis uses the EMC_AXIS_JOG command which chec
As you suggested, most USB devices only send updates when inputs change. In
the USB spec, it is possible to query the device inputs. I'm guessing
hal_input doesn't support that.
hidcomp (http://hidcomp.sourceforge.net/) is supposed to support that, but I
can't guarantee it works because the devi
I'm trying to rebuild my application against 2.4.3. It is an external
application that uses header files in include, and links with libraries in
src/emc2/lib (emchal, emc, nml, encini)
The current version won't compile because canon.hh includes emctool.h, which
is in another directory (sorry,
> The creation of the v2.4_branch means that new features can
> go on master again, but I hope that for the time being people
> will concentrate on 2.4.
For the casual source code getter, how do I update my current git source to
that particular version?
---
I've been playing around with RTNet (realtime ethernet). I've been looking
at it as an alternative to parallel ports, and internal cards.
I've built an embedded slave using the Atmel NGW100 board. I have a test
application that runs on RTAI linux (the master), which reads the encoder
and IO stat
I'm developing a hal real time component, and I'm getting the "RTAPI: ERROR:
Unexpected realtime delay on task 2" message.
What does it mean? What is causing it? Where is the message coming from?
How can I view the symptoms (timings)? How do I find out what "task 2" is?
I've added debug to my
it in C?
> -Original Message-
> From: Frank Tkalcevic [mailto:fr...@franksworkshop.com.au]
> Sent: Sunday, 13 December 2009 1:42 PM
> To: 'EMC developers'
> Subject: Additional includes for comp
>
> Is there any way to specify additional include directorie
Is there any way to specify additional include directories when building a
.comp component, ie the -Ipath option? The realtime headers don't seem to
have all of the system header files. (at least I think that is why it can't
find the headers)
Frank
-
> Are either of you using FF2? I'm not, but it would be nice
> if that was tested too.
I am, but not by design. I just kept tweaking parameters until the servo
behaviour was adequate. Was there a significant change that I need to look
out for?
--
t index-enable to the pid component too?
> Without that the behaviour should be unchanged (might be good
> to test that too).
>
> Regards,
> Alex
>
> ------
> From: "Frank Tkalcevic"
> Sent: Tuesday, November 17, 2
I just tried the fix on my lathe, and it looks worse. I always got the
klunk the first time I homed, but it would be OK if I did subsequent homes.
I use a home switch and an index pulse. It moves to the home switch, then
reverses slowly to find the index pulse. It sets the X value to be the
cur
According the documentation
http://www.isd.mel.nist.gov/projects/rcslib/NMLcpp.html there is a
blocking_read() function that internally waits until it is signalled that
there is a message to be read. You can either look into the blocking_read
implementation to see if the events/fds are exposed, or
, but leaves me in a bit of a
pickle because, although peek() works well, it is a race condition between
me peeking and axis reading.
> -Original Message-
> From: Frank Tkalcevic [mailto:fr...@franksworkshop.com.au]
> Sent: Wednesday, 22 July 2009 1:47 PM
> To: 'EMC dev
bject: Re: [Emc-developers] Shared memory errors
>
> On Wed, Jul 22, 2009 at 08:00:14AM +1000, Frank Tkalcevic wrote:
> > I'm having another issue with reading messages from the
> error channel.
> > I've copied the code from shcom.cc where it will do a
> read(
ies to open a NML channel, it checks
> first if it's a master or not. If it is, it initializes the
> channel rather than connecting to it.
>
> Regards,
> Alex
>
>
> - Original Message -
> From: "Frank Tkalcevic"
> To: "'EMC develop
Thanks, that helped.
My problem was trying to creating the interface to nml before calling
hal_ready.
> -Original Message-
> From: Jeff Epler [mailto:jep...@unpythonic.net]
> Sent: Wednesday, 22 July 2009 6:46 AM
> To: EMC developers
> Subject: Re: [Emc-developers] Shared memory error
nect
until successful?
Thanks,
Frank
> -Original Message-
> From: Frank Tkalcevic [mailto:fr...@franksworkshop.com.au]
> Sent: Tuesday, 21 July 2009 6:50 PM
> To: 'EMC developers'
> Subject: Shared memory errors
>
> I've been writing a component, based on em
I've been writing a component, based on emclcd.cc and shcom.cc. I think the
code was working - it would start up in a halrun, but now it fails to start
with nml errors.
I'm not sure how the shared memory works. Do I need to allocate a new
key(?) for my application? I just took shcom.cc which
Oops. I think I did this. I hit refresh on the bug list in a browser and I
think it added a new one.
Sorry,
Frank
> -Original Message-
> From: SourceForge.net [mailto:nore...@sourceforge.net]
> Sent: Tuesday, 7 July 2009 3:11 PM
> To: nore...@sourceforge.net
> Subject: [Emc-developers
On May 25, 2009 04:50:25 pm Frank Tkalcevic wrote:
> My only gripe with QtDesigner is that you need to recompile the code
> afterwards.
That is not true with python, which is interpreted. Of course, you have to
regenerate the gui's python code with a command such as :
My only gripe with QtDesigner is that you need to recompile the code
afterwards. And, in order to add functionality, users need to know how to
write C++ code. This will restrict customisation of the new GUI to advanced
users.
_
From: Maxime Lemonnier [mailto:maxime.lemonn...@gmail.com
My 2 cents...
I'm new to emc. I chose emc for my lathe after using mach for my mill,
mainly for the threading support, but also because it was open source.
I was very impressed with the internal architecture of emc, and how easy it
was to build and a configure a system with components, pins an
Sorry my question was a bit vague. I was browsing the code and noticed it
opened a socket connection to emc. That was what I wanted to know. What
does it connect to, and what interface was it using?
> -Original Message-
> From: Eric H. Johnson [mailto:ejohn...@camalytics.com]
> Sent
Thanks, that looks like what I want. What interface is it using? Is that
emcrsh?
--
Register Now for Creativity and Technology (CaT), June 3rd, NYC. CaT
is a gathering of tech-side developers & brand creativity profes
I have a USB device I want to use as a pendant which has an LCD display.
I'm OK with writing the USB code to program it, I'm just unsure how I can
integrate it with emc to provide useful outputs.
I'm building yet another "joystick" type hal component. It works at a lower
level allowing access t
85 matches
Mail list logo
