On 2020-12-04 10:59, andy pugh wrote:
I suspect that it might be best to ignore the built-in homing.
Configure for immediate homing (ie zero HOME_SEARCH_VELOCITY) and then
handle to motion interlock with the motion pins.
@all, thank you for your help.
We prefer using servo drive built-in
Hello,
I have an ESTUN Ethercat servo drive with integrated homing sequence
management.
Basically, I only need to write 2 registers to start a homing sequence,
and then wait for a specific status bit to become true in a status register.
Probably easy to map to some HAL pins with a small
component:
component spindle_security "Block non G0 motions when the spindle is not
running";
author "Frederic RIBLE";
pin in s32 motion_type;
pin in bit at_speed;
pin in bit override;
pin out bit feed_hold;
variable double old;
function _;
license "GPL"; // indicat
That might be expected behaviour, I suppose.
yes, as stated in the manpage.
You could try linking to feed-hold?
Just tried: that is affecting G0 as well.
So not ok for my config because I need G0 motion with spindle stopped.
___
Emc-users
Unfortunately this solution is not solving the issue.
The motion HAL component seems ignoring spindle-at-speed when in M5 state.
On 2018-11-11 20:26, Andy Pugh wrote:
On 11 Nov 2018, at 17:41, Frederic RIBLE wrote:
Any clue for inhibiting G1 when the spindle is not activated ?
Adding logic
Just broke an endmill with a bad GCODE macro where M3 command was missing.
I have motion.spindle-at-speed properly tight to my VFD component.
But motion does not check it when the spindle is in stop condition.
Any clue for inhibiting G1 when the spindle is not activated ?
Thanks.
Frederic
On 2017-12-02 11:24, Gene Heskett wrote:
I'd have really serious doubts anybody ever considered the transmission
line termination characteristics needed to preserve a perfect square
wave anyway. And if I asked that of the guy at Lowes spooling off 100
feet of that cable for me, he'd think I was
On 2017-12-01 22:44, Gene Heskett wrote:
So 2 of those solves my diff to single ended signal problems with
bandwidth to throw away. Almost dc to light. And the duty cycle is 50%,
plus or minus 2% worst case due to the motors wibblies at nearly 8500
rpms.
Great!
Additionally, some of those
On 2017-11-30 15:13, Gene Heskett wrote:
generator. So I am going to make up a board to carry a 74ls86 today,
I am using an old 75ALS195N to convert the symmetrical encoder output to
single TTL signal.
It is working well with my spindle running at 3000rpm.
I have a RVP510 encoder configured
Integrating QtQuickVcp with LinuxCNC should be very interesting.
I have developed a DRO component for QtQuickVcp with the look and feel
of Gmoccapy.
See my CNC mill project: https://youtu.be/LnJv07yeGt0?t=3m28s
I am using also Gmoccapy for other machines, and would love having one
ported to
On 2017-10-15 19:58, John Dammeyer wrote:
I mentioned 10uS period because that's what I read is the PRU interrupt rate
and why the machine kit can't do more than 50kHz stepping pulses.Don't
know if that's true either now because that information might also be old.
10uS is the standard
On 2017-10-15 14:18, Nicklas Karlsson wrote:
Yes, 1024ppr is overkill. And my RVP510 encoder is capable of lower
resolution because it is programmable!
I use to prefer some margin, you turn down resolution because it's higher than
needed?
I am using 1024ppr, probably with lot of margin for
On 2017-10-15 11:37, andy pugh wrote:
The simple answer being to use a 500 or 250 line encoder.
(That hexagonal boring demo I showed earlier was using a home-machined
50-slot code wheel)
Yes, 1024ppr is overkill. And my RVP510 encoder is capable of lower
resolution because it is programmable!
As far as I know, PRU micro-code is using sampling for counting.
I have validate up to 100 kHz, with a PRU period of 2.5uS.
So you have 4 samples per pulse period, which is enough, but not fully
satisfying if you need to master the jitter.
See
On 2017-10-14 00:21, andy pugh wrote:
I lost track of Machinekit a while ago. Do they have any
encoder-counting facility in the PRU?
If they have then the rest is just config. (editing text files...)
Yes, Machinekit has nice PRU encoder facility with eQEP: high-speed quadrature
+ index.
On 2017-10-10 00:02, Sebastian Kuzminsky wrote:
I don't know about machinekit, but in linuxcnc incremental jogs are
safer than continuous jogs in high latency situations like these.
Same behavior with machinekit.
This is one of the motivations I have for adding physical jog wheels on
my
On 2017-10-09 23:27, andy pugh wrote:
This is why I like the approach taken by the "Touchy" interface which
basically requires a jogwheel and a physical "start" button.
You can still begin the homing sequence from the touch screen, and I
think I would prefer that to bring up a "press start to
On 2017-10-08 02:33, Chris Albertson wrote:
I don't know why people like to use these 1GHz ARM boards for machine
controllers.
I can answer, I am using both BBB and PC!
My own motivation is primary experimenting:
- 2017: EMCO-120P lathe + BBB + MK + Modified Gmoccapy GUI
- 2015: C3000
On 22/07/2017 03:58, Nicklas Karlsson wrote:
With 0MQ protocol ?
Yes, that is BeagleBone Black and ZeroMQ.
I agree, 7" screen is too small.
Good for 3D printing, but not for a more complex CNC machine.
With so small buttons, there is risk of harmful mistakes.
Also, I will add MPG in the future
On 19/07/2017 20:01, Nicklas Karlsson wrote:
GUI on one computer and real time on other machine.
I am using Machinekit exactly for that purpose.
Look at my retrofit using an Android tablet for the GUI:
https://youtu.be/LnJv07yeGt0?t=2m
I am not yet fully satisfied by this setup: this is a low
On 10/12/2016 15:34, Peter C. Wallace wrote:
> What is the magnitude of the oscillation?
>
> can you plot it and provide a link to the plot?
>
Looks like we have a scaling mismatch somewhere in the drive between
command and feedback.
However, that does not explain everything.
Will come back soon
On 09/12/2016 21:48, Peter C. Wallace wrote:
>
> There are only some 10s of usec between reading the encoder and writing the
> PWM value, assuming the read and write functions are in the proper order in
> the hal file. Also Fred said that changing the servo thread from 1 to 10 KHZ
> made little
On 09/12/2016 17:23, Peter C. Wallace wrote:
>
> Is the velocity loop gain in LinuxCNC set correctly?
>
> That is, is it set so you get ~+-10V at +-2000 RPM
>
The gain should be ok because we have tuned it to achieve very small
following error when all PID parameters are 0 except FF1=1.
In that
On 09/12/2016 02:21, Peter C. Wallace wrote:
> Are you sure the PWMGEN output mode is set to mode 2 ?
> (you will get very weird and unstable results otherwise)
> Note that changing the mode may invert the feedback so beware of runaways
>
I think so, will double check.
We have monitored the
Hello
At Electrolab hackerspace, we are working on the retrofit of a
Mori-Seiki SL0 CNC lathe.
We are are using a Mesa 6i24/7i48 combination.
The drive is a A06B-6047-H208 (double velocity control unit).
Axis motors are A06B-0641-B011 and A06B-0642-B011 (yellow cap).
The position feedback comes
.
On 17/09/2016 23:02, andy pugh wrote:
> On 17 September 2016 at 20:24, Frederic RIBLE <fri...@teaser.fr> wrote:
>
>> This is why we are thinking about using mesa to implement this function.
>
> It might be enough to "fool the drive" but there really is no information
to implement this function.
Depending on latency for this operation, we may experience additional
stability issue in the loop.
Frederic.
On 17/09/2016 15:43, andy pugh wrote:
> On 17 September 2016 at 10:20, Frederic RIBLE <fri...@teaser.fr> wrote:
>
>> We would like to repl
Yes, the diagram has been dropped by the mailing list software.
You can find it here:
https://forum.linuxcnc.org/media/kunena/attachments/16648/Mauriceservo.jpg
Frederic.
On 17/09/2016 15:32, Gene Heskett wrote:
> On Saturday 17 September 2016 05:20:09 Frederic RIBLE wrote:
>
&g
Hello,
We are working on a retrofit for a Mori Seiki SL-0 CNC lathe.
The original X/Z axis servo drives are Fanuc analog.
They work in velocity mode with external frequency to voltage conversion
embedded in the original CNC controller.
We would like to replace the controller with a Mesa
Le 10/04/2016 15:02, Andy Pugh a écrit :
> I would use "lincurve" with velocity as input and jog scale as output.
>
>
@Andy: thanks, lincurve is doing the job.
Easier than scale/limit1 combination.
# Jog wheels section
net jog-enable halui.mode.is-manual
newinst lincurve lincurve.0 pincount=2
x1 count mode
(HAL encoder x1 mode)
net jog-counts-z hpg.encoder.00.chan.01.rawcounts => axis.2.jog-counts
net jog-enable => axis.2.jog-enable
setp axis.2.jog-vel-mode 1
net jz1 hpg.encoder.00.chan.01.velocity-abs => scale.1.in
net jz2 scale.1.out => limit1.1.in
net jz3 limit1.1.out =&g
Hello,
For my lathe control panel project, I have integrated 2 MPG with 100ppr.
(see
http://blog.f1oat.org/2016/03/21/emco-turn-120p-machinekit-control-panel/)
I would like to drive the size of each jog step according to MPG
rotation velocity.
The idea is to have 0.01mm steps at low speed, an
Yes, G10L11 is what I am looking for!
Thanks.
Frederic.
Le 2015-08-01 00:03, John Thornton a écrit :
G10L11 works for me.
JT
On 7/31/2015 2:41 PM, Frederic RIBLE wrote
My machine has a 4 tools ATC
(http://blog.f1oat.org/2015/06/13/intelys-c3000-retrofit/).
I would like to implement a NGC routine for automatic measurement of the
4 tool lengths, to be called after having manually loaded the tools in
the rack.
What is the good way for updating Z offsets in the
Le 2015-06-09 08:57, Erik Christiansen a écrit :
Some serious work, to get it done in 3 months, by the look of it.
A very nifty machine, now.
Was Machinekit difficult to set up, e.g. for your custom I/O?
ISTR some pin identification ambiguities/variations in the early days,
but haven't
Hello,
After 3 months of work, my retrofit project is now almost finished.
Thank you all for your help, and especially for the tool in spindle
saving problem.
Here is a video of the machine: https://youtu.be/LnJv07yeGt0
Frederic (http://cnc.f1oat.org)
Le 2014-05-03 04:50, Jon Elson a écrit :
There is a pic from the HD5000 webcam in the Wiki at
http://wiki.linuxcnc.org/cgi-bin/wiki.pl?Adding_Digital_Zoom_To_Camview-Emc
that is not very impressive, you might be able to get it within .1mm
but that's about it.
I would like to clarify an
Le 2014-02-06 16:23, Sebastian Kuzminsky a écrit :
Frederic, Dewey, are you both happy with the state of that branch? Are
we ready to merge it into master?
There is a regression here in simulation mode (without -H flag).
I am checking the code.
Frederic.
Le 2014-02-06 19:55, Dewey Garrett a écrit :
-
Several folks have reported that the pendant stops updating.
On mine, if the pendant is idle (no buttons pressed, no wheel
movement) for approximately 40 seconds, the display stops
Marius, thank you for the patch. I have few others to integrate also.
Yes, including it in LinuxCNC is on my todo list.
I will try to boost the priority for that task.
About the display, I have the same issue here. I believe it is link to
some power saving feature.
I do not know if we can find a
Le 2013-11-30 00:44, Gene Heskett a écrit :
All 4 routines are there now. And the overly verbose comments should
explain what to do to fine tune them for your usage.
Cheers, Gene
Thanks. These ngc files look interesting.
I have uploaded my complete LinuxCNC configuration here:
Le 2013-11-29 21:43, Gene Heskett a écrit :
Found it! An almost chance comment, in the [RS274NGC] section of the
docs said I had the statement in the wrong section of the .ini file. One
would, unless instructed to do otherwise, assume that it went immediately
under the PROGRAM_PREFIX
On Wed, 27 Nov 2013 21:14:58 -0500, Gene Heskett ghesk...@wdtv.com
wrote:
So, the next question is, what file do I put that offset pair into, and
what buttons do I click to tell LinuxCNC to reference those offsets when
setting up to etch the top of a board?
Hello Gene,
You need to set
Le 2013-11-28 12:33, Gene Heskett a écrit :
On Thursday 28 November 2013 06:13:27 Frederic Rible did opine:
And just to maintain camera focus, should not restore it to the
starting height at the end of that motion? This camera does however
seem to be autofocus over about a 1.5 range
Le 2013-11-28 03:55, a k a écrit :
how to convert STL file into other useful format?
I know several tools able to do that:
- CamBam
- Deskproto
- BobCAD-CAM
- Mecsoft FreeMill
- Vectric Cut3D
CamBam is not able to handle four axis easily (you can do indexed, but
that is tricky), but the
Le 2013-11-28 22:41, Gene Heskett a écrit :
So obviously the SUBROUTINE_PATH isn't working, and the MDI
enumeration is screwed.
Yes, you have the two problems in your config.
Here are some parts of my CNC-6040.ini file
---
[RS274NGC]
PARAMETER_FILE = linuxcnc.var
SUBROUTINE_PATH =
On Thu, 28 Nov 2013 18:10:25 -0500, Gene Heskett ghesk...@wdtv.com
wrote:
On Thursday 28 November 2013 17:31:26 Frederic RIBLE did opine:
Can I specify that SUBROUTINE_PATH twice?
I do not know.
HALUI]
MDI_COMMAND=oalign-startcall
MDI_COMMAND=oalign-ycall
MDI_COMMAND=oalign-xcall
Le 2013-11-23 19:35, Marcus Bowman a écrit :
I have an optical machine scope that has that kind of gimbal adjustment,
although it is just two screws which act on a spherical arrangement inside (I
think). Works fine.
I have also cobbled together an adjustable camera mount for a USB camera,
Le 2013-11-16 00:51, Gene Heskett a écrit :
lsof |grep camunits
Here is the result
f1oat@linuxcnc2:~$ lsof | grep camunits
python1890 f1oat mem REG8,1 672081072082
/usr/lib/pyshared/python2.6/_camunits.so
python1890 f1oat mem REG8,1 157684
Le 2013-11-16 10:13, Gene Heskett a écrit :
On Saturday 16 November 2013 03:17:35 Marius Liebenberg did opine:
Gene,
Are you sure that config was not just for Gmoccaypy or Gscreen?
No I am no sure. The wiki pages does not specify. I have about 5 or
6 camviewcfgs I have collected, none of
Le 2013-11-14 21:05, Gene Heskett a écrit :
So are mine despite cooking in olive oil these days. And, it turned
out that my particular setup was not using any sum2 modules, so I
added the above, with a 4th axis for my 4 table, and it all shows up
in a hal configuration display. But still
Le 2013-11-08 13:16, Russell Brown a écrit :
...
Now I just have to make a tweakable mount and try to figure out how I
can persuade camview-emc to include the cropping slider in it's controls
section within Axis so I can twiddle when it's being used for real.
...
I have started a Wiki
Le 2013-11-07 19:04, Gene Heskett a écrit :
...
If you find a workable answer, I am all ears.
Cheers, Gene
I have built a fix mount on Y axis for my webcam with a custom ball
joint for fine orientation along the Z axis.
You can find a description here: http://blog.f1oat.org/tag/hd5000/
Le 2013-11-07 22:01, Gene Heskett a écrit :
Thats exactly what I am looking for. ultil.crop, I don't recall seeing
a mention of. Is there a man page describing it?
I have discovered that this feature was existing when analyzing camview
source tree !
Below is my camview config file. util.crop
Le 2013-07-21 19:19, Sebastian Kuzminsky a écrit :
(I'm replying to a private email from Frederic Rible, and i'm putting
emc-users@ back on the delivery list because i think others may
benefit from our conversation.)
On 07/21/2013 03:30 AM, Frederic Rible wrote:
On Thu, 18 Jul 2013 21:11
Below is the list of dev packages I have on my machine.
I do not know exactly what packages are required, but you may start by
installing
- libc6-dev
- libusb-1.0-0-dev
Frederic.
dpkg --get-selections | grep dev
autotools-devinstall
devscriptsinstall
Le 2013-06-23 07:10, andy pugh a écrit :
On 22 June 2013 02:36, f1oat fri...@teaser.fr wrote:
I have developed a HAL module for this kind of MPG pendant.
Look at
http://www.linuxcnc.org/index.php/english/forum/24-hal-components/26679-xhc-hb04-wireless-mpg-pendant-hal-module
Would you be
My experience is too short to conclude if there are issues or not with
wireless.
I see on ebay a very similar pendant with wired connection :
http://www.ebay.com/itm/CNC-Wireless-MACH3-USB-4-Axis-MPG-Pendant-HandWheel-/221236705741?pt=LH_DefaultDomain_0hash=item3382bc39cd
We have good chance
58 matches
Mail list logo