Re: [Emc-users] Debugging Python under LinuxCNC

[Fox Mulder]

That's not true for Python development in general. If you develop a
standalone Python program you can use IDEs with all their features like
breakpoints and show/edit variable contents etc. :)

The problem starts if the python program is called from another program
which is not attached to the debugger. In most cases the IDE can't
attach to this python instance and therefor can't be used. Debugging
like in the old days with print or better a logging system is still the
best way. But this is the same for C/C++ or other programs which are
executed from another program outside the IDE.

Am 14.09.21 um 19:06 schrieb John Dammeyer:

I'm more used to working with program development systems that have 
breakpoints, allow single stepping through assembly code and examining variable 
values and even processor registers.  Working with Python is like writing in 
BASIC with the Radio Shack TRS80 home computer.   I've gone back to the dark 
ages...


Ciao,
 Rainer


___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] How to change step pulse width for stepper driver

[Fox Mulder]

Thanks for the hint.

This parameter is not quite the one i would have looked into. After i
changed it from 5000 to 1 the pulse width changed from ~3,2µs to
~4µs. And with a value of 2 i get ~7µs which is enough to work with
my old stepper driver. Interestingly when i set higher values the width
doesn't go over ~7µs but i can live with this setup for the moment.

I also ordered a few new stepper driver from omc-stepperonline.com
because they are so cheap i couldn't resist. Maybe it works better when
i have replaced my old hardware. :)

Ciao,
 Rainer

Am 02.06.21 um 20:11 schrieb Peter C. Wallace:

On Wed, 2 Jun 2021, Fox Mulder wrote:


Date: Wed, 2 Jun 2021 20:06:05 +0200
From: Fox Mulder 
Reply-To: "Enhanced Machine Controller (EMC)"
    
To: emc-users@lists.sourceforge.net
Subject: Re: [Emc-users] How to change step pulse width for stepper
driver

So i tried setting stepspace = 1 but no change. Then i tried setting
both steplen and stepspace to 1 and also no change. It seems as if
LinuxCNC ignores my values.

When i open HAL Meter and show the current value it shows me 75000 for
both of them. But this would mean 75s pulse width which i don't have.
It is still around 2,3s.

I attached my old confiurations and maybe someone can see what the
problem is. I don't know if i have to change something else for the
current version of LinuxCNC. I only know that the configs worked back
then when i last used them 15 years ago. :)

Ciao,
    Rainer




"setp parport.0.reset-time 5000"

is what is setting the step pulse length

When you use reset mode its normal to set steplength and stepspace to 1


Peter Wallace
Mesa Electronics


___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users



___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] How to change step pulse width for stepper driver

[Fox Mulder]

So i tried setting stepspace = 1 but no change. Then i tried setting
both steplen and stepspace to 1 and also no change. It seems as if
LinuxCNC ignores my values.

When i open HAL Meter and show the current value it shows me 75000 for
both of them. But this would mean 75µs pulse width which i don't have.
It is still around 2,3µs.

I attached my old confiurations and maybe someone can see what the
problem is. I don't know if i have to change something else for the
current version of LinuxCNC. I only know that the configs worked back
then when i last used them 15 years ago. :)

Ciao,
 Rainer

Am 02.06.21 um 19:21 schrieb Fox Mulder:

Thanks for the hint. I will try to set stepspace at a higher value.

And i know about the newer Drivers. I already use DRV8825 for a new
project and it is far superior to the old L297/L298 combo.

At the moment i just want to get the old hardware running and i already
have planed to replace the old driver with a new one somewhere in the
future. But as long as it works for my intention it is ok. :)

Ciao,
  Rainer

Am 02.06.21 um 18:31 schrieb Chris Albertson:

That 0.5 uS spec is for the minimum time the clock needs to be low.
  STEPLEN is the time to remain high.  You have STEPSPACE=0 which
means the
low time is to be "0 nanoseconds".    I'd make STEPSPACE much larger.

BTW, that L297 is primitive at best.   You'd be amazed at how much better
the new stepper drivers are and the price has come down.  You get much
better performance (more speed and torque) and no high-pitched noise.
Something like one of these on each axis would be a great upgrade.  Buy a
24V or 35V power supply to go with it.
.omc-stepperonline.com/digital-stepper-driver/digital-stepper-driver-18~56a-20-50vdc-for-nema-23-24-34-stepper-motor-dm556t.html

<https://www.omc-stepperonline.com/digital-stepper-driver/digital-stepper-driver-18~56a-20-50vdc-for-nema-23-24-34-stepper-motor-dm556t.html>





On Wed, Jun 2, 2021 at 5:48 AM Fox Mulder  wrote:


Hi,

i try to reactivate my old self-build cnc mill with stepper motors and
come to a little problem. My 15 year old electronics seems all still to
work. It uses L297/L298 stepper drivers driven by the parallel port from
my old pc which is quite outdated but this doesn't matter at the moment.

I reinstalled Debian with LinuxCNC v2.8.1 and at first start it updated
my old config files. But then when i try to jog one of my axes it
doesn't work. I only get some jitter from the motors but no movement.

After some investigation i found out that the width of the step pulses
is too small. At the moment they are ~2,3µs which should be valid
because the datasheet of the L297 says a minimum of 0,5µs is ok. But
when i try the step generation with a simple python script which has a
wider step pulse the motors move how they should. So i think i can get
it to work again with a wider pulse.

Now to my question, how to make the pulse width for the step signal
wider. I read in the stepgen module about the steplen parameter but it
seems i'm not setting it correct.

In my old hal file the configuration for the stepgen module looks like
this:

loadrt stepgen step_type=0,0,0
addf stepgen.make-pulses base-thread
addf stepgen.capture-position servo-thread
addf stepgen.update-freq servo-thread

setp stepgen.0.position-scale [JOINT_0]SCALE
setp stepgen.0.steplen 1
setp stepgen.0.stepspace 0
setp stepgen.0.dirhold 51056
setp stepgen.0.dirsetup 21056
setp stepgen.0.maxaccel [JOINT_0]STEPGEN_MAXACCEL

Now i tried to increase steplen to something like 1 and expected a
step pulse width of ~10µs but nothing changes. The pulse width is still
~2,3µs.

Is there something i'm missing or what is the best way to change the
pulse width of the generated step signal?

Ciao,
   Rainer


___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users







___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users
loadrt [KINS]KINEMATICS
loadrt [EMCMOT]EMCMOT base_period_nsec=[EMCMOT]BASE_PERIOD 
servo_period_nsec=[EMCMOT]SERVO_PERIOD num_joints=[KINS]JOINTS 
loadrt hal_parport cfg="0xCC00 out 0xC400 in"
setp parport.0.reset-time 5000
loadrt stepgen step_type=0,0,0

addf parport.0.read base-thread
addf parport.1.read base-thread
addf stepgen.make-pulses base-thread
addf parport.0.write base-thread
addf parport.0.reset base-thread
addf parport.1.write base-thread

addf stepgen.capture-position servo-thread
addf motion-command-handler servo-thread
addf motion-controller servo-thread
addf stepgen.update-freq servo-thread

net spindle-cmd <= spindle.0.speed-out
net spindle-on <= spindle.0.on

setp parport.0.pin-01-out-invert 1
net estop-out => parport.0.pin-01-out
net estop-out => iocontrol.0.emc-enable-in
net estop-

Re: [Emc-users] How to change step pulse width for stepper driver

[Fox Mulder]

Thanks for the hint. I will try to set stepspace at a higher value.

And i know about the newer Drivers. I already use DRV8825 for a new
project and it is far superior to the old L297/L298 combo.

At the moment i just want to get the old hardware running and i already
have planed to replace the old driver with a new one somewhere in the
future. But as long as it works for my intention it is ok. :)

Ciao,
 Rainer

Am 02.06.21 um 18:31 schrieb Chris Albertson:

That 0.5 uS spec is for the minimum time the clock needs to be low.
  STEPLEN is the time to remain high.  You have STEPSPACE=0 which means the
low time is to be "0 nanoseconds".I'd make STEPSPACE much larger.

BTW, that L297 is primitive at best.   You'd be amazed at how much better
the new stepper drivers are and the price has come down.  You get much
better performance (more speed and torque) and no high-pitched noise.
Something like one of these on each axis would be a great upgrade.  Buy a
24V or 35V power supply to go with it.
.omc-stepperonline.com/digital-stepper-driver/digital-stepper-driver-18~56a-20-50vdc-for-nema-23-24-34-stepper-motor-dm556t.html
<https://www.omc-stepperonline.com/digital-stepper-driver/digital-stepper-driver-18~56a-20-50vdc-for-nema-23-24-34-stepper-motor-dm556t.html>




On Wed, Jun 2, 2021 at 5:48 AM Fox Mulder  wrote:


Hi,

i try to reactivate my old self-build cnc mill with stepper motors and
come to a little problem. My 15 year old electronics seems all still to
work. It uses L297/L298 stepper drivers driven by the parallel port from
my old pc which is quite outdated but this doesn't matter at the moment.

I reinstalled Debian with LinuxCNC v2.8.1 and at first start it updated
my old config files. But then when i try to jog one of my axes it
doesn't work. I only get some jitter from the motors but no movement.

After some investigation i found out that the width of the step pulses
is too small. At the moment they are ~2,3µs which should be valid
because the datasheet of the L297 says a minimum of 0,5µs is ok. But
when i try the step generation with a simple python script which has a
wider step pulse the motors move how they should. So i think i can get
it to work again with a wider pulse.

Now to my question, how to make the pulse width for the step signal
wider. I read in the stepgen module about the steplen parameter but it
seems i'm not setting it correct.

In my old hal file the configuration for the stepgen module looks like
this:

loadrt stepgen step_type=0,0,0
addf stepgen.make-pulses base-thread
addf stepgen.capture-position servo-thread
addf stepgen.update-freq servo-thread

setp stepgen.0.position-scale [JOINT_0]SCALE
setp stepgen.0.steplen 1
setp stepgen.0.stepspace 0
setp stepgen.0.dirhold 51056
setp stepgen.0.dirsetup 21056
setp stepgen.0.maxaccel [JOINT_0]STEPGEN_MAXACCEL

Now i tried to increase steplen to something like 1 and expected a
step pulse width of ~10µs but nothing changes. The pulse width is still
~2,3µs.

Is there something i'm missing or what is the best way to change the
pulse width of the generated step signal?

Ciao,
   Rainer


___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users







___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

[Emc-users] How to change step pulse width for stepper driver

[Fox Mulder]

Hi,

i try to reactivate my old self-build cnc mill with stepper motors and
come to a little problem. My 15 year old electronics seems all still to
work. It uses L297/L298 stepper drivers driven by the parallel port from
my old pc which is quite outdated but this doesn't matter at the moment.

I reinstalled Debian with LinuxCNC v2.8.1 and at first start it updated
my old config files. But then when i try to jog one of my axes it
doesn't work. I only get some jitter from the motors but no movement.

After some investigation i found out that the width of the step pulses
is too small. At the moment they are ~2,3µs which should be valid
because the datasheet of the L297 says a minimum of 0,5µs is ok. But
when i try the step generation with a simple python script which has a
wider step pulse the motors move how they should. So i think i can get
it to work again with a wider pulse.

Now to my question, how to make the pulse width for the step signal
wider. I read in the stepgen module about the steplen parameter but it
seems i'm not setting it correct.

In my old hal file the configuration for the stepgen module looks like this:

loadrt stepgen step_type=0,0,0
addf stepgen.make-pulses base-thread
addf stepgen.capture-position servo-thread
addf stepgen.update-freq servo-thread

setp stepgen.0.position-scale [JOINT_0]SCALE
setp stepgen.0.steplen 1
setp stepgen.0.stepspace 0
setp stepgen.0.dirhold 51056
setp stepgen.0.dirsetup 21056
setp stepgen.0.maxaccel [JOINT_0]STEPGEN_MAXACCEL

Now i tried to increase steplen to something like 1 and expected a
step pulse width of ~10µs but nothing changes. The pulse width is still
~2,3µs.

Is there something i'm missing or what is the best way to change the
pulse width of the generated step signal?

Ciao,
 Rainer


___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] New kind of stepper motor on eBay

[Fox Mulder]

Am 29.05.2017 um 10:36 schrieb Lester Caine:
> On 28/05/17 05:51, Fox Mulder wrote:
>> Looks to me just like a stepper motor with integrated motor driver without 
>> feedback. Maybe the Mechaduino is more like a closed-loop stepper-servo. I 
>> have ordered some PCBs from DirtyPcb and will assemble them when they 
>> arrive. Hope they are as good as announced.
>>
>> https://www.kickstarter.com/projects/tropicallabs/mechaduino-powerful-open-source-industrial-servo-m
> 
> Has the latest batch of boards already sold out ?
> 
Seems so. I ordered bare PCBs which i have to equip myself with the
corresponding parts.

--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] New kind of stepper motor on eBay

[Fox Mulder]

Looks to me just like a stepper motor with integrated motor driver without 
feedback. Maybe the Mechaduino is more like a closed-loop stepper-servo. I have 
ordered some PCBs from DirtyPcb and will assemble them when they arrive. Hope 
they are as good as announced.

https://www.kickstarter.com/projects/tropicallabs/mechaduino-powerful-open-source-industrial-servo-m

https://hackaday.io/project/11224-mechaduino

Ciao,
Rainer

> Am 28.05.2017 um 05:58 schrieb Erik Christiansen :
> 
>> On 27.05.17 16:58, Gene Heskett wrote:
>> I think so too.  But I am so used to thinking in in/lbs, and have never 
>> seen a formula or a chart that converts Nm to in/lbs,
> 
> But on just about any full linux distro under your roof, you have:
> 
> $ units
> You have: 3 N m
> You want: lbf in
>* 26.552237
>/ 0.03766161
> 
> The first is the forward conversion, the second is the reciprocal.
> (Use "units -1" if that's annoying.)
> 
>> so I've no clue if they might be usable for me. So what is 3Nm equ to?
> 
> Most of the advertisements seem to be in oz-in, so to "imperialise", I'd:
> 
> You have: 3 N m
> You want: ozf in
> Unknown unit 'ozf'
> You want: oz force in
>* 424.8358
>/ 0.0023538506
> 
> Ah, now, that's substantial, in the size range I tend to look at.
> 
> Erik
> 
> P.S. As a side issue, I couldn't remember the imperial g, but:
> You have: gravity
> You want: ft/s^2
>* 32.174049
>/ 0.03108095
> 
> Yes, now I remember using 32, back as a schoolkid, before we went
> metric.
> 
> --
> Check out the vibrant tech community on one of the world's most
> engaging tech sites, Slashdot.org! http://sdm.link/slashdot
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users
--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] CPU clock speed [Was: for PCW?]

[Fox Mulder]

On the Rpi you can get the current speed of the cpu with the following
command: cat /sys/devices/system/cpu/cpu0/cpufreq/scaling_cur_freq

But beware that the Rpi default uses the on-demand governor for cpu
clocking. This means when you overclock the Rpi and it is idle you get a
lower clock with this command. Only when the CPU is on heavy load it
will get overclocked and the command shows this.

You can also set the threshold when the CPU gets overclocked (in percent
CPU load) with help of the following file. System default is 50% on my
Rpi and 95% on my desktop PC:
/sys/devices/system/cpu/cpufreq/ondemand/up_threshold

Ciao,
 Rainer

Am 08.04.2017 um 16:15 schrieb Gene Heskett:
> On Saturday 08 April 2017 09:33:33 sam sokolik wrote:
> 
>> does cat /proc/cpuinfo work?
>>
> It doesn't answer the how fast is it now question Sam, only stating the 
> present bogomips as 38.something per core.
> 
> I have found a line in /boot/config.txt that seems to say its running at 
> 700 MHz as default, and could uncomment a line that tells it to run at 
> 800 MHz.
> 
> But I suspect that is assuming no stickon heat sinks, which I have on it, 
> and which aren't noticably warm. With the heat sinks on it, and forced 
> fan cooling 1200 MHz. I may clone that u-sd card to use as a recovery, 
> then raise that to 800 MHz.  Wash, rinse & repeat until its heating 
> according to my ir thermometer, to an unhealthy temp. I don't have a fan 
> on it. But if it will do 1 GHz, that would be a noticable improvement.
> 
> Unfortunately, the spi buss seems to be limited by the slew rate of a 
> gpio pin to about 32 MHz. A single spi clock pulse, on a gigahertz 
> scope, is a quite good approximation of a classic sine-squared pulse, 
> very little evidence of an extended logic 1 to trigger the 7i90, and so 
> fussy that the 7i90 will not talk to it if the 10x probe is on the 
> signal at lcnc startup. I do not know how, but it might be helped if the 
> pi could drive two gpio's for the clocking signal, thereby increasing 
> the slew rate by paralleling two gpio's for that function.  That of 
> course would need a modified adaptor board to combine them. Better yet 
> would be a schmidt trigger buffer to steepen up the sides of the pulse, 
> and by adjusting the trigger level, widen the pulse by a couple 
> nanoseconds.  All this of coarse would be before the src termination on 
> the adaptor board.
> 
> Thank you Sam.
> 
> Cheers, Gene Heskett
> 

--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] I think its time to issue a warning about sandisk 32GB supposedly 2x speed of a samsung sd card

[Fox Mulder]

Am 20.12.2016 um 01:35 schrieb Bertho Stultiens:
> On 12/20/2016 01:10 AM, Gene Heskett wrote:
>> What has been other raspi users experience?  Is a sandisk crap?
> The *biggest* problem is a read/write file-system on an SD card. That
> kills about any cards within 1..3 months for me (especially when local
> logging is enabled). I normally have a R/O fs and remote logging.
> Everything else is in ramfs.

That's true. The same problem goes or ssd hard drives on the pc as well.
So some of the solutions for ssd's are also good for sd cards on the
rpi. Use of a ramdisk for directories like /tmp or /var/log is a good
idea because these are normally the most frequent written directories on
linux.

For the /tmp dir there is already a prepared systemd script which also
works on the rpi. Just execute the following commands which i found here
[1]:
sudo cp /usr/share/systemd/tmp.mount /etc/systemd/system/
sudo systemctl enable tmp.mount

For /var/log i didn't found a ready solution and wrote my own script. It
makes /var/log a ramdisk and copies the content at boot/shutdown from/to
a permanent directory /var/permanentlog so i don't loos all the logs. A
Cronjob also synchronizes the content every hour. If you are interested
you can try it out but the description is only in german on my wiki [2].
I use the script on my main PC and on my 24/7 RPI camera and it works
quite good. :)

Ciao,
 Rainer

[1]
https://wiki.debian.org/SSDOptimization#Reduction_of_SSD_write_frequency_via_RAMDISK

[2]
https://quakeman.mooo.com/wiki/doku.php?id=wissen:linux:ramdisk#var_log_in_eine_ramdisk_verlegen

--
Developer Access Program for Intel Xeon Phi Processors
Access to Intel Xeon Phi processor-based developer platforms.
With one year of Intel Parallel Studio XE.
Training and support from Colfax.
Order your platform today.http://sdm.link/intel
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] X on pi-3?

[Fox Mulder]

Am 06.11.2016 um 00:20 schrieb Gene Heskett:
> My instant problem is that X is not running on the r-pi-3, so while I 
> have beautiful hires text screens, at least on the login screen, no x, 
> so linuxcnc cannot run.
> 
> Looking in /var/cache/apt/archives it does look like quite a bit of x is 
> installed.
> 
> Does anyone have some hints on how to bring up a working x on the r-pi-3 
> running linux?

In the tool "raspi-config" you can select the "Boot Options" and change
if it should start to console or the X11 gui after boot.

This tool has some more useful options specific for the rpi so just take
a look at the different menu entries. :)

Ciao,
 Rainer

--
Developer Access Program for Intel Xeon Phi Processors
Access to Intel Xeon Phi processor-based developer platforms.
With one year of Intel Parallel Studio XE.
Training and support from Colfax.
Order your platform today. http://sdm.link/xeonphi
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] BTDT re BACKGROUND,FOREGROUND not defined.

[Fox Mulder]

Am 06.02.2016 um 02:40 schrieb Gene Heskett:
> On Friday 05 February 2016 13:30:54 Fox Mulder wrote:
> 
>> Am 05.02.2016 um 17:42 schrieb Gene Heskett:
>>> On Friday 05 February 2016 08:15:32 Dr. Nikolaus Klepp wrote:
>>>> Am Freitag, 5. Februar 2016 schrieb Gene Heskett:
>> The second problem with the missing lower part of the video is a bit
>> strange. The video in your screenshot has a resolution of 1280x720
>> (16:9). Could it be that your camera maybe is in widescreen format and
>> 1280x720 is the native resolution?
> 
> I did another lsusb -vv|less and scanned thru the report, I I believe you 
> are 101% correct, in the 1280x1024 stanza's is "still image not 
> supported"
> 
> So just for grins, I changed the vertical in both places to 720, and its 
> working quite nicely in 16x9 format.  It also has a 320x240 mode but I 
> don't need that at all.  It also shows a 1600x1200, but again "still 
> image not supported". In that event, I get the 1280x720 image with right 
> and lower borders.
> 
> So this problem is solved, and I thank you very much.

Ok that explains the missing lower parts. :)

> Thanks again Rainer, with your considerable help I think I have 
> it "whupped".  I think the only other feature I'd like is a crop 
> function so I am only seeing the center 200x200 out of the 1280x720 
> mode.  A poor mans telephoto lens IOW.

I updated my script at http://pastebin.com/YaCqLNzz to implement
optional image cropping to the center 200x200, 300x300 or 400x400 pixels
of the image (look in the "Resolution" menu of the tk window for
possible options).

Ciao,
 Rainer

--
Site24x7 APM Insight: Get Deep Visibility into Application Performance
APM + Mobile APM + RUM: Monitor 3 App instances at just $35/Month
Monitor end-to-end web transactions and take corrective actions now
Troubleshoot faster and improve end-user experience. Signup Now!
http://pubads.g.doubleclick.net/gampad/clk?id=272487151&iu=/4140
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] BTDT re BACKGROUND,FOREGROUND not defined.

[Fox Mulder]

Am 05.02.2016 um 17:42 schrieb Gene Heskett:
> On Friday 05 February 2016 08:15:32 Dr. Nikolaus Klepp wrote:
> 
>> Am Freitag, 5. Februar 2016 schrieb Gene Heskett:
> Then opencv has a what looks like an input buffer size problem. two 
> screen snapshots attached to show that.  The camera only has 2 output 
> modes according to an lsusb -vv, a jpg in 640x480, which works well, but 
> is about half resolution.  Switching to  1280x1024, its native raw video 
> format, and while the tcl/tk window is the correct size, about 300 lines 
> of the bottom of the pix are dropped. I /think/ the target crosshairs 
> and circle are still correct, in which case I get the impression a few 
> pixels are missing on the right.  But since the horizontal scale is 
> missing I can't be sure.

One thing i see in your attached screenshot is that the video area in
the tk window has not the correct resolution. In my code the resolution
was set to 1280x960 for the video. In your screenshot the area for the
video is still 1280x960 and not 1280x1024. This leads to the question if
all occurrences for the resolution were changed correctly in the python
code.

The second problem with the missing lower part of the video is a bit
strange. The video in your screenshot has a resolution of 1280x720
(16:9). Could it be that your camera maybe is in widescreen format and
1280x720 is the native resolution?

This could explain why the grid is cropped at the lower end because the
program thinks the resolution is 1280x1024 when it in reality is
1280x720. The calculation for the position of the grid would then be
starting below the video frame. But this is only a wild guess.

What native resolution does other programs show for your camera like vlc
or guvc-view?

Ciao,
 Rainer

--
Site24x7 APM Insight: Get Deep Visibility into Application Performance
APM + Mobile APM + RUM: Monitor 3 App instances at just $35/Month
Monitor end-to-end web transactions and take corrective actions now
Troubleshoot faster and improve end-user experience. Signup Now!
http://pubads.g.doubleclick.net/gampad/clk?id=272487151&iu=/4140
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Camview-emc workalike?

[Fox Mulder]

Am 01.02.2016 um 18:32 schrieb Gene Heskett:
> On Monday 01 February 2016 05:46:16 Fox Mulder wrote:
> 
>> Am 31.01.2016 um 23:49 schrieb Gene Heskett:
>>> On Sunday 31 January 2016 11:34:46 Fox Mulder wrote:
>>>> Am 31.01.2016 um 16:19 schrieb Gene Heskett:
>>>>> On Sunday 31 January 2016 07:48:43 Fox Mulder wrote:
>>>>>> Am 30.01.2016 um 02:10 schrieb Gene Heskett:
>>>>>>> On Friday 29 January 2016 16:52:58 Gene Heskett wrote:
>>>>>>>> On Friday 29 January 2016 15:01:16 Fox Mulder wrote:
>>>>>>>>> Am 29.01.2016 um 04:49 schrieb Gene Heskett:
>>>>>>>>>> On Tuesday 26 January 2016 14:16:03 Gene Heskett wrote:
>>>>>>>>>>> On Tuesday 26 January 2016 12:41:26 Fox Mulder wrote:
>>>>>>>>>>>> Am 26.01.2016 um 18:00 schrieb Gene Heskett:
>>>>>>>>>>>>> On Tuesday 26 January 2016 09:55:33 Fox Mulder wrote:
>>>>>>>>>>>>>> Am 26.01.2016 um 13:20 schrieb Gene Heskett:
>> You can try the relevant lines in an interactive python shell to see
>> if they work. Just start "python" (for python 2) or "python3" and try
>> the following lines (change the picture and path to an existing one):
>>
>> from PIL import Image as pilimg
>> from PIL import ImageTk
>> from tkinter import *
> Fails here
> gene@GO704:~/linuxcnc/configs/GO704fast$ python
> Python 2.7.3 (default, Mar 14 2014, 11:57:14) 
> [GCC 4.7.2] on linux2
> Type "help", "copyright", "credits" or "license" for more information.
>>>> from PIL import Image as pilimg
>>>> from PIL import ImageTk
>>>> from tkinter import *
> Traceback (most recent call last):
>   File "", line 1, in 
> ImportError: No module named tkinter
> 
> But 
> from Tkinter import *
> works
Sorry i forgot that it is "import Tkinter" for your version.

>> root = Tk()
> But fails here:
>>>> root Tk()
>   File "", line 1
> root Tk()
>   ^
> SyntaxError: invalid syntax

Sounds like the Python Tkinter package is missing (python-tk or
python3-tk). Also you must have the Tk package installed for it to work.
But when installing python-tk the dependency for tk should automatically
be selected.

> No use continueing, try python3
>> img = pilimg.open('/temp/picture.png')
>> imgtk = ImageTk.PhotoImage(img)
> 
> When running python3 shell:
> gene@GO704:~/linuxcnc/configs/GO704fast$ python3
> Python 3.2.3 (default, Feb 20 2013, 17:02:41) 
> [GCC 4.7.2] on linux2
> Type "help", "copyright", "credits" or "license" for more information.
>>>> from PIL import Image as pilimg
> Traceback (most recent call last):
>   File "", line 1, in 
> ImportError: No module named PIL
> 
> I THOUGHT pillow built and installed itself in both paths.  But from 
> locates response, only for 2.6 and 2.7.  I'll google for it again, and 
> see if there is a pip3 utility.  Humm, but first, need pip3. But its 
> pip-3.2, not pip3, so its munching its way thru a python3 install of 
> pillow.

Pip only installs python2 packages, for python3 you need pip3.2.
Just do "pip3 install Pillow" and you should have the python3 version of
it. But with the error from python2 i don't think you will get any
further. You need the line root = Tk() to initialize the Tk Framework
before the other commands can be executed.

> Now it should be there, but again the install bitched about a missing 
> tkinter, needs Tcl/Tk-8.5 libraries and wheezy only has 8.4. This is 
> wearing thin... 

Hmm that's no good news. One thing you can try is to install the wheezy
version of the python pil packages (apt-get install python-imaging
python-imaging-tk) which should work with the local Tk version. But
before you have to uninstall pillow (pip uninstall Pillow) because they
don't work together.

Ciao,
 Rainer

--
Site24x7 APM Insight: Get Deep Visibility into Application Performance
APM + Mobile APM + RUM: Monitor 3 App instances at just $35/Month
Monitor end-to-end web transactions and take corrective actions now
Troubleshoot faster and improve end-user experience. Signup Now!
http://pubads.g.doubleclick.net/gampad/clk?id=267308311&iu=/4140
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Camview-emc workalike?

[Fox Mulder]

Am 31.01.2016 um 23:49 schrieb Gene Heskett:
> On Sunday 31 January 2016 11:34:46 Fox Mulder wrote:
> 
>> Am 31.01.2016 um 16:19 schrieb Gene Heskett:
>>> On Sunday 31 January 2016 07:48:43 Fox Mulder wrote:
>>>> Am 30.01.2016 um 02:10 schrieb Gene Heskett:
>>>>> On Friday 29 January 2016 16:52:58 Gene Heskett wrote:
>>>>>> On Friday 29 January 2016 15:01:16 Fox Mulder wrote:
>>>>>>> Am 29.01.2016 um 04:49 schrieb Gene Heskett:
>>>>>>>> On Tuesday 26 January 2016 14:16:03 Gene Heskett wrote:
>>>>>>>>> On Tuesday 26 January 2016 12:41:26 Fox Mulder wrote:
>>>>>>>>>> Am 26.01.2016 um 18:00 schrieb Gene Heskett:
>>>>>>>>>>> On Tuesday 26 January 2016 09:55:33 Fox Mulder wrote:
>>>>>>>>>>>> Am 26.01.2016 um 13:20 schrieb Gene Heskett:
>>>>
>>>> Here is the new script which needs a few new python packages
>>>> installed. For me in debian it was python3-pil,
>>>> python3-pil.imagetk, python3-tk.
>>>
>>> I hope they are available in wheezy.  Checking now.  The first 2 are
>>> not. 3.2.3 Compatible deb's?  URL IOW. python3-tk now installed.
>>
>> Best way for you would be to use python2 where these packages would be
>> called python-imaging and python-imaging-tk. I think wheezy doesn't
>> have these packages for python3 included.
>>
>> Alternatively you could try and use python pip to install these
>> packages for python3.
>>
>>From the pypi sites web server, pillow (a fork of PIL) is recommended to 
> replace PIL.
I replaced PIL with latest pillow 3.1.0 with pip3 and it works without
changing anything in the code.

> So I have installed that and now get to line 6 with python2, can't find 
> tkinter *.  It is not installed & not available.
> 
> Uppercased the name to Tkinter, and I did get it to run, and it opens a 
> white screen with the menu stuff but no video.
That's a problem with package renaming (Tkinter to tkinter) between some
older and newer packages. Don't know who thought that would be a good
idea. :/

> Trace from the launching screen:
> 
> gene@GO704:~/linuxcnc$ python python_opencv_camera_example_with_gui.py
> 
> Exception in Tkinter callback
> Traceback (most recent call last):
>   File "/usr/lib/python2.7/lib-tk/Tkinter.py", line 1437, in __call__
> return self.func(*args)
>   File "/usr/lib/python2.7/lib-tk/Tkinter.py", line 498, in callit
> func(*args)
>   File "python_opencv_camera_example_with_gui.py", line 92, in show_video
> imgtk = ImageTk.PhotoImage(image=img)
>   File "/usr/local/lib/python2.7/dist-packages/PIL/ImageTk.py", line 115, 
> in __init__
> self.paste(image)
>   File "/usr/local/lib/python2.7/dist-packages/PIL/ImageTk.py", line 180, 
> in paste
> from PIL import _imagingtk
> ImportError: cannot import name _imagingtk
You can try the relevant lines in an interactive python shell to see if
they work. Just start "python" (for python 2) or "python3" and try the
following lines (change the picture and path to an existing one):

from PIL import Image as pilimg
from PIL import ImageTk
from tkinter import *
root = Tk()
img = pilimg.open('/temp/picture.png')
imgtk = ImageTk.PhotoImage(img)

If you got an error you can see after which line it comes. This maybe
help to find the problem.

> And with my lack of python knowledge, no clue how it got renamed with the 
> leading _.
> 
> With pillow installed, locate says:
> 
> gene@GO704:~/linuxcnc$ locate imagingtk
> /usr/lib/pyshared/python2.6/PIL/_imagingtk.so
> /usr/lib/pyshared/python2.7/PIL/_imagingtk.so
> /usr/lib/python2.6/dist-packages/PIL/_imagingtk.so
> /usr/lib/python2.7/dist-packages/PIL/_imagingtk.so
> 
> Subtle syntax error? ENV path? Obviously IDK.
The names of the .so libraries on the harddisk doesn't matter. It's
normal that they have an underscore in front, same here on my installation.

Ciao,
 Rainer

--
Site24x7 APM Insight: Get Deep Visibility into Application Performance
APM + Mobile APM + RUM: Monitor 3 App instances at just $35/Month
Monitor end-to-end web transactions and take corrective actions now
Troubleshoot faster and improve end-user experience. Signup Now!
http://pubads.g.doubleclick.net/gampad/clk?id=267308311&iu=/4140
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Camview-emc workalike?

[Fox Mulder]

Am 31.01.2016 um 16:19 schrieb Gene Heskett:
> On Sunday 31 January 2016 07:48:43 Fox Mulder wrote:
> 
>> Am 30.01.2016 um 02:10 schrieb Gene Heskett:
>>> On Friday 29 January 2016 16:52:58 Gene Heskett wrote:
>>>> On Friday 29 January 2016 15:01:16 Fox Mulder wrote:
>>>>> Am 29.01.2016 um 04:49 schrieb Gene Heskett:
>>>>>> On Tuesday 26 January 2016 14:16:03 Gene Heskett wrote:
>>>>>>> On Tuesday 26 January 2016 12:41:26 Fox Mulder wrote:
>>>>>>>> Am 26.01.2016 um 18:00 schrieb Gene Heskett:
>>>>>>>>> On Tuesday 26 January 2016 09:55:33 Fox Mulder wrote:
>>>>>>>>>> Am 26.01.2016 um 13:20 schrieb Gene Heskett:
>> Here is the new script which needs a few new python packages
>> installed. For me in debian it was python3-pil, python3-pil.imagetk,
>> python3-tk.
> 
> I hope they are available in wheezy.  Checking now.  The first 2 are not. 
> 3.2.3 Compatible deb's?  URL IOW. python3-tk now installed.

Best way for you would be to use python2 where these packages would be
called python-imaging and python-imaging-tk. I think wheezy doesn't have
these packages for python3 included.

Alternatively you could try and use python pip to install these packages
for python3.

>> http://pastebin.com/YaCqLNzz
> 
> Got it, thank you. In the previous script this new camera does 1280x1024 
> natively so I updated that in the two places shown as 960.  That worked 
> well except the bottom of the display was clipped off. So I need to find 
> where I can expand the display window for that setting.

Normally the size of the tk window should resize to the given
resolution. The window itself is larger than the video resolution
because of the toolbar and the menu around it.
This is around ~86 pixels more in height which gives 1024 + 86 = 1110
pixels in height. So if your monitor doesn't have a high resolution the
window is at it's maximum and the lower end of the video get's cropped
off. Even Full-HD isn't enough for that resolution because 30 pixels are
missing in height (1110 - 1080 = 30 pixels too large). I didn't
implement any video resizing to get around this problem.

Ciao,
 Rainer

--
Site24x7 APM Insight: Get Deep Visibility into Application Performance
APM + Mobile APM + RUM: Monitor 3 App instances at just $35/Month
Monitor end-to-end web transactions and take corrective actions now
Troubleshoot faster and improve end-user experience. Signup Now!
http://pubads.g.doubleclick.net/gampad/clk?id=267308311&iu=/4140
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Camview-emc workalike?

[Fox Mulder]

Am 29.01.2016 um 04:49 schrieb Gene Heskett:
> On Tuesday 26 January 2016 14:16:03 Gene Heskett wrote:
> 
>> On Tuesday 26 January 2016 12:41:26 Fox Mulder wrote:
>>> Am 26.01.2016 um 18:00 schrieb Gene Heskett:
>>>> On Tuesday 26 January 2016 09:55:33 Fox Mulder wrote:
>>>>> Am 26.01.2016 um 13:20 schrieb Gene Heskett:
> Rainer:  Any progress on a camview-emc workalike?  I should know by 
> tomorrow late if the camera I bought will work with v4l2 & the opencv 
> script I have.
I don't have much time at the moment to get the script working together
with linuxcnc. For me it only was a simple experiment if it was possible
to overlay an image on a live camera stream with means of simple
programming.

> In the FWIW category, good lighting jumps the frame rate on this 
> el-cheapo camera to something around 35 FPS.  Totally realtime, but 
> slightly overexposed.  See's a piece of pcb copper just fine though.
> Blob detection works but slows the frame rate to half.  That is to be 
> expected with the frame comparison method chosen.
Most cameras only have their maximum frame rate with enough lighting.
Else they have longer exposure times to get a working picture which
limits the fps. But the limiting factor for the fps in this case is the
processing of the video. The more complex it is the lower the fps will
get. Blob detection is a simple example which shows how much the
processing power decides the final frame rate.

Ciao,
 Rainer

--
Site24x7 APM Insight: Get Deep Visibility into Application Performance
APM + Mobile APM + RUM: Monitor 3 App instances at just $35/Month
Monitor end-to-end web transactions and take corrective actions now
Troubleshoot faster and improve end-user experience. Signup Now!
http://pubads.g.doubleclick.net/gampad/clk?id=267308311&iu=/4140
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Camview-emc update

[Fox Mulder]

Am 26.01.2016 um 18:00 schrieb Gene Heskett:
> On Tuesday 26 January 2016 09:55:33 Fox Mulder wrote:
> 
>> Am 26.01.2016 um 13:20 schrieb Gene Heskett:
>>> All my machines are on the wheezy based install. Latest installed
>>> python is 2.7.3.  But I do see that 3.2.3 is available.  Can they
>>> co-exist?
>>
>> Pyhton3 and Python2 can coexist on the same machine for a long time
>> since many programs still use Python2.
>> My examples should also work with Python2 and there is no real need
>> for Python3. I only wanted to use newer Python because i program
>> mostly with it.
>>
>> Ciao,
>> Rainer
> 
> Ahh, so. That works, but is there a menu to control options?  All I see 
> is the out of focus camera output, noisy enough that the frame rate 
> might be usable.  With what looks like a 10 cm scale overlaid on the 
> left but it doesn't start at 0, like the bottom of the image is clipped 
> of by the fixed size window border.  The scale starts at 1.4 at the 
> bottom of the image.  Is that what I should be seeing?
> 
> I enabled the help, but I can't read your native language. :-[  A 
> translation I can edit in would be super. But I figured that out I 
> think.
> 
> One thing I note is that this camera can do 640x480 natively, a entering 
> a 1 for that is the only mode that shows the frame rate.
> 
> Making 5.8 FPS but low light slows that camera, its a $6 camera. I need 
> to locate something with a much longer lens as the min distance where I 
> have it mounted is about 4" from the workpiece unless I remove the tool.  
> With its fairly wide angle utility lens, the FOV at that range is 
> measured in multiple inches.

There are not much control options because it is only a quick test
script for opencv with a camera i hacked in a few hours. ;)

I updated the script on pastebin with english help for the key bindings.
You could just select the color of the grid, some additional grid lines
and the blob detection.

The scaling just doesn't show the 0 in the left down corner. Also a
quick hack and it doesn't work very well with different resolutions. The
hard coded resolutions on keys 1,2,3,4 are for my camera model. Maybe
you have to change them for yours in the code.

Because my usb microscope camera has a manual focus ring i didn't care
about controlling the focus. So i think if the camera doesn't do it
natively the program will not change it at all. ;)

Ciao,
 Rainer

--
Site24x7 APM Insight: Get Deep Visibility into Application Performance
APM + Mobile APM + RUM: Monitor 3 App instances at just $35/Month
Monitor end-to-end web transactions and take corrective actions now
Troubleshoot faster and improve end-user experience. Signup Now!
http://pubads.g.doubleclick.net/gampad/clk?id=267308311&iu=/4140
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Camview-emc update

[Fox Mulder]

Am 26.01.2016 um 16:49 schrieb Gene Heskett:
> On Tuesday 26 January 2016 06:44:48 Fox Mulder wrote:
> 
>> Am 26.01.2016 um 08:16 schrieb Gene Heskett:
>>>> All the cool kids are using OpenCV
>>>> http://opencv.org/
>>>
>>> Yee gawds, the git clone is north of 400 MiB!
>>> Nevertheless, it has been built and installed according to the web
>>> page directions, and I have scanned thru all the tuts, but haven't
>>> found any tuts on integrating it with LCNC. And 98% of the tuts
>>> discuss fixed still images. Not moving video which this LCNC
>>> scenario is. I have executed the 3 files it installed in
>>> /usr/local/bin, getting the help screens, none of which mention a
>>> v4l2 video source.
>>
>> Just a few days ago i needed a little program which could overlay a
>> grid over my cheap usb microscope camera. Because i couldn't find any
>> simple program i tried opencv for that purpose.
>>
>> And to be honest it is quite simple to do. Therefor i cloned the
>> latest opencv 3.1.0 git repo and compiled it to use the newly python3
>> bindings. It works nicely with v4l2 camera devices (like my usb
>> camera) and can process the video.
>>
>> But first i had to realize that opencv works with video like with all
>> other normal picture formats. It gets one frame, processes it and
>> shows the result. Than in a loop it goes to the next frame and so on.
>> So the resulting framerate depends on the processing power of the cpu.
>>
>> My simple test Script for Python 3.x is available at pastebin
>> (http://pastebin.com/jfnpDs72) which should work with most usb
>> cameras.
>>
> I got the script, installed some python3, but it dies on the import cv2 
> line.  The only cv2 is cv2.so in the 2.7 chain.
> 
> /usr/local/lib/python2.7/dist-packages/cv2.so
> 
> And I cannot find a python3 equivalent. So what package is it in?  Or is 
> it something that could be handled by python-six? But it claims to be 
> the python2 interface.
> 
> So I need some guidance.
> 
> Thanks Rainer.

Python3 bindings only exist in latest opencv 3.x version. Therefor i had
to compile the git version myself because the repository only had opencv
2.x available. For Python3 support you have to install the python3-dev
package before compiling opencv.

But the simpler solution is to use python2 with opencv like i mentioned
in my last email. It gives the same functionality and should also work
with my example. I tested it and the example script needs no changes to
work with python2 out of the box with opencv 3.x.

I haven't installed opencv 2.4 from the repository to try this version
but i think the python API hasn't changed since then.

Ciao,
 Rainer


--
Site24x7 APM Insight: Get Deep Visibility into Application Performance
APM + Mobile APM + RUM: Monitor 3 App instances at just $35/Month
Monitor end-to-end web transactions and take corrective actions now
Troubleshoot faster and improve end-user experience. Signup Now!
http://pubads.g.doubleclick.net/gampad/clk?id=267308311&iu=/4140
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Camview-emc update

[Fox Mulder]

Am 26.01.2016 um 13:20 schrieb Gene Heskett:
> All my machines are on the wheezy based install. Latest installed python 
> is 2.7.3.  But I do see that 3.2.3 is available.  Can they co-exist?

Pyhton3 and Python2 can coexist on the same machine for a long time
since many programs still use Python2.
My examples should also work with Python2 and there is no real need for
Python3. I only wanted to use newer Python because i program mostly with it.

Ciao,
Rainer


--
Site24x7 APM Insight: Get Deep Visibility into Application Performance
APM + Mobile APM + RUM: Monitor 3 App instances at just $35/Month
Monitor end-to-end web transactions and take corrective actions now
Troubleshoot faster and improve end-user experience. Signup Now!
http://pubads.g.doubleclick.net/gampad/clk?id=267308311&iu=/4140
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Camview-emc update

[Fox Mulder]

Am 26.01.2016 um 08:16 schrieb Gene Heskett:
>> All the cool kids are using OpenCV
>> http://opencv.org/
> 
> Yee gawds, the git clone is north of 400 MiB!
> Nevertheless, it has been built and installed according to the web page 
> directions, and I have scanned thru all the tuts, but haven't found any 
> tuts on integrating it with LCNC. And 98% of the tuts discuss fixed 
> still images. Not moving video which this LCNC scenario is.
> I have executed the 3 files it installed in /usr/local/bin, getting the 
> help screens, none of which mention a v4l2 video source.

Just a few days ago i needed a little program which could overlay a grid
over my cheap usb microscope camera. Because i couldn't find any simple
program i tried opencv for that purpose.

And to be honest it is quite simple to do. Therefor i cloned the latest
opencv 3.1.0 git repo and compiled it to use the newly python3 bindings.
It works nicely with v4l2 camera devices (like my usb camera) and can
process the video.

But first i had to realize that opencv works with video like with all
other normal picture formats. It gets one frame, processes it and shows
the result. Than in a loop it goes to the next frame and so on. So the
resulting framerate depends on the processing power of the cpu.

My simple test Script for Python 3.x is available at pastebin
(http://pastebin.com/jfnpDs72) which should work with most usb cameras.

I implemented different things to try out. Even some simple blob
detection filter could be de-/activated (see key bindings in code or
press h in live video preview).

About the integration into linuxcnc i have no clue but i think it
shouldn't be too hard to do somehow.

Ciao,
 Rainer

--
Site24x7 APM Insight: Get Deep Visibility into Application Performance
APM + Mobile APM + RUM: Monitor 3 App instances at just $35/Month
Monitor end-to-end web transactions and take corrective actions now
Troubleshoot faster and improve end-user experience. Signup Now!
http://pubads.g.doubleclick.net/gampad/clk?id=267308311&iu=/4140
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Little OT, but...

[Fox Mulder]

My statement for this OT discussion about security of pc's internet
connection is to stop it because it leads to nowhere. Nobody will change
his own position and it is unnecessary work for me to delete all these
useless messages in my email client. ;)

Ciao,
 Rainer

--
Subversion Kills Productivity. Get off Subversion & Make the Move to Perforce.
With Perforce, you get hassle-free workflows. Merge that actually works. 
Faster operations. Version large binaries.  Built-in WAN optimization and the
freedom to use Git, Perforce or both. Make the move to Perforce.
http://pubads.g.doubleclick.net/gampad/clk?id=122218951&iu=/4140/ostg.clktrk
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] "Cubieboard: available at 49$" (was: LinuxCNC would be very nice on the UDOO board.)

[Fox Mulder]

Am 23.04.2013 00:07, schrieb Kasey Matejcek:
> Has anyone tried to play a video file from something like a Gopro camera is
> there a enough power to play these type of files ?
> Or isn't there enough video or cpu power to do this
> I have costumer that is asking if I could put something together that would
> play the files after their done recording and they don't want to spend the
> money on a laptop

There are some android xbmc versions which "should" work with hardware
accelerated video decoding. But it seems not to be ready for the masses.

So if you want just an little arm box for video output with hardware
full-hd support i would suggest you buy a rasperry-pi modell A (25$) or
B (35$), depending on the need for ethernet support. For these exist
multiple xbmc versions (openelec, raspbmc, ...) with hardware
accelerated video output. I own one myself and use it as standalone
media player.

Newer gopro cameras use h264 encoded video format which is supported by
the rpi gpu. But you should first be sure that the video format is
supported before you order one. ;)

Ciao,
 Rainer

--
Try New Relic Now & We'll Send You this Cool Shirt
New Relic is the only SaaS-based application performance monitoring service 
that delivers powerful full stack analytics. Optimize and monitor your
browser, app, & servers with just a few lines of code. Try New Relic
and get this awesome Nerd Life shirt! http://p.sf.net/sfu/newrelic_d2d_apr
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] "Cubieboard: available at 49$" (was: LinuxCNC would be very nice on the UDOO board.)

[Fox Mulder]

Am 22.04.2013 23:11, schrieb Matt Shaver:
> On Mon, 22 Apr 2013 21:47:12 +0200
> Fox Mulder  wrote:
> 
>> How about the Cubieboard for 49$ from http://cubieboard.org/ ?
> 
> AFAIK, these are hard to buy in the U.S. There's no stocking
> distributor here :(
> 
> These guys are in the U.S., but don't have stock:
> https://store.iotllc.com/product.php?productid=2&cat=0&bestseller=Y
> 
> I like the board though! When they are available, I'll try to get one.
> If anyone coming to Wichita has a Cubieboard, bring it!

On their homepage are some world wide shipping shops listed. My board
was shipped from china and it worked without problems. :)

Ciao,
 Rainer

--
Precog is a next-generation analytics platform capable of advanced
analytics on semi-structured data. The platform includes APIs for building
apps and a phenomenal toolset for data science. Developers can use
our toolset for easy data analysis & visualization. Get a free account!
http://www2.precog.com/precogplatform/slashdotnewsletter
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] "Cubieboard: available at 49$" (was: LinuxCNC would be very nice on the UDOO board.)

[Fox Mulder]

How about the Cubieboard for 49$ from http://cubieboard.org/ ?

It uses a Cortex-A8@1GHz (Allwinner A10 SoC with FPU and SIMD) with 1GB
DDR3@480MHz and 4GB onboard flash memory. The other onboard goodies are
HDMI 1080p, 100MBit network, USB host, 96 GPIO (with I2C, SPI, RS232,
ADC, ...), uSD, OpenGL ES GPU, Sata II port, IR receiver.
It runs Android and other linux distros without problems.

I own one from the original indiegogo campaign (like kickstarter) but
until now i had not much time to do something useful with it. It would
be nice if it also could be used for linuxcnc. :)

Ciao,
 Rainer

Am 19.04.2013 06:26, schrieb Gregg Eshelman:
> The UDOO is a single board computer with a 1ghz quad core CPU plus an Arduino 
> DUE. Same CPU as the Raspberry Pi but with 4 cores.
> 
> Runs Linux and Android.
> 
> http://www.kickstarter.com/projects/435742530/udoo-android-linux-arduino-in-a-tiny-single-board
> 
> Note the much saner port arrangement than what's on the RasPi with its ports 
> on all sides and sticking out at apparently random distances. Except for 
> power input and Micro SD slot, UDOO's ports are all on one side and flush 
> with each other, which will make case design much easier.
> 
> Port LinuxCNC to the ARM i.MX6 Freescale CPU and have it communicate 
> to/through the Arduino and use a motor controller plugged into the Arduino 
> pin header sockets.
> 
> At 4.33 inch x 3.35 inch (11 cm x 8.5 cm) this would make a very compact 
> control system for CNC. The motor drivers and power supply would dwarf it.

--
Precog is a next-generation analytics platform capable of advanced
analytics on semi-structured data. The platform includes APIs for building
apps and a phenomenal toolset for data science. Developers can use
our toolset for easy data analysis & visualization. Get a free account!
http://www2.precog.com/precogplatform/slashdotnewsletter
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] A nice unusual LCNC video.

[Fox Mulder]

Youtube tells me that this video is not available in my country
(germany). But it doesn't say why. :(

Am 08.02.2013 11:00, schrieb Spiderdab:
> Hi all, it's a lot of time i'm not writing here (and too little time to
> put my hands on LCNC again..)
> but I have an unusual LCNC video for you.
> 
> http://www.youtube.com/watch?v=22XJMw-28DU
> 
> Some of you already know my "inverted tetrapod". Here is a little
> demonstration I did some time ago on a real stage during a pause.
> I just wish it may be be nice to watch.
> 
> Greetings, 
> Davide D'Angelo.
> 
> 

--
Free Next-Gen Firewall Hardware Offer
Buy your Sophos next-gen firewall before the end March 2013 
and get the hardware for free! Learn more.
http://p.sf.net/sfu/sophos-d2d-feb
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Z-probing derived kinematics

[Fox Mulder]

Hi,

at the moment i'm building myself a simple probe and do some tests with
gridprobe.ngc. But i'm still at the problem of using the results as
feedback for emc to use it for correcting the moves.

So is there anything new on this front? :)

Ciao,
 Rainer

Am 21.03.2012 21:53, schrieb Michael Haberler:
> Rainer,
> 
> I'm not sure if I fully understand your question. 
> 
> The way this works is through forward and reverse kinematics (see 
> http://wiki.linuxcnc.org/uploads/EMC_Control_LG.gif)
> 
> when the trajectory planner outputs a new target position it passes it 
> through reverse kins to determine the corresponding axis positions. So thats 
> where the z correction happens - tp runs about 1000 times a sec, so the 
> correction is applied at the same rate
> 
> the probekins module currently searches sequentially through the list of 
> triangles searching for a hit. On a hit, the z value is used to correct the 
> output to the axis control.
> 
> The problem:
> 
> Assume you have two triangles which share an edge. Now program a move which 
> in x,y follows this shared edge. I *think* this border case is responsible 
> for the following error. if you move across triangles in an arbitrary fashion 
> (not tracking edges) it works just fine. Maybe somebody else can dig deeper 
> into this, it might be just a trivial thing.
> 
> The ray/triangle intersection test itself is very fast - its a favorite 
> CG/raytracing community problem. Sequentially testing many triangles might 
> become an issue, but there are standard ways to deal with this, like KD trees.
> 
> -m
> 
> Am 21.03.2012 um 20:06 schrieb Fox Mulder:
> 
>> Today i wrote a little program which calculates the absolute z distance
>> between a distorted plane (stl file) at a given xy location and zero. I
>> used simple triangles and no higher order interpolation. The result was
>> calculated very fast because of the simple maths used (only + - * /) and
>> always correct compared to a cad program.
>>
>> So i think the problems lay beyond the z-depth calculation from the stl
>> file. I wonder how you process a simple move over a distance which may
>> span over 3 triangles.
>>
>> Do you separate this move into three single moves with different start
>> and end z-depths?
>>
>> Or do you check before each step the z-depth of the stl file and correct
>> the commanded z-depth (would be very cpu intensive if done for each
>> single step)?
>>
>> Ciao,
>> Rainer
>>
>> Am 20.03.2012 00:00, schrieb Michael Haberler:
>>> I ran into a problem with following errors if a move is exactly along a 
>>> triangle edge
>>>
>>> I havent investigated it further but I guess it could be a 
>>> discontinuity/numerical accuracy/stability problem along the edges
>>>
>>> If that is the case, the approach using discrete-triangle based correction 
>>> could be fundamentally flawed 
>>>
>>> one way I guess would work is to create a continuous surface correction 
>>> function, like a cubic spline
>>>
>>>
>>> - Michael
>>>
>>>
>>> Am 19.03.2012 um 21:03 schrieb Fox Mulder:
>>>
>>>> Since the last mail from december it has gotten really quite about this
>>>> very handy feature. But i can see in the git logs that it was further
>>>> developed.
>>>>
>>>> So what is the state of it right now and how could it be used?
>>>>
>>>> Sadly i can't find any info at all on the linuxcnc website.
>>>> I want to do some pcb etching tests in the next few weeks and i think
>>>> this feature would be a great benefit for this purpose. :)
>>>>
>>>> Ciao,
>>>>Rainer
>>>>
>>>> Am 21.12.2011 08:31, schrieb Michael Haberler:
>>>>> I found the idea quite intriguing, and thought about integrating the 
>>>>> surface correction idea better. After bouncing a few ideas with Andy, 
>>>>> this is what I have so far:
>>>>>
>>>>> probekins: a kinematics module which be default behaves like trivkins
>>>>> it accepts a mesh of triangles which define Z correction values (actually 
>>>>> an approximation of the workpiece surface, or machine surface for that 
>>>>> matter)
>>>>>
>>>>> this mesh might eventually generated by probing, but also used as a 
>>>>> general machine Z correction 
>>>>>
>>>>>
>>>>> this isnt finished or polished, but I&

Re: [Emc-users] Hey Gene! Hand held O'scope for ya.

[Fox Mulder]

I also got the DS1052E and for the very low price (~280€ without vat) it
has quite much features. The software hack to 100 MHz is also very nice. :)

Ciao,
 Rainer

Am 15.08.2012 23:09, schrieb Peter Loron:
> The Nano is a decent little scope. Very handy and can be easily dropped in 
> your gear bag or used on your bench as an extra scope to supplement your 
> bench unit. 
> 
> Far better than no scope, although I would recommend having something bigger 
> if possible. I've also got one of the DS1052E scopes and it is pretty good 
> for the money. It's also quite portable, compared to my old boatanchor scopes.
> 
> -Pete
>  
> On Aug 14, 2012, at 8:12 PM, Gene Heskett wrote:
> 
>> On Tuesday 14 August 2012 23:09:35 Jason Burton did opine:
>>
>>> On Aug 14, 2012 9:08 AM, "Gene Heskett"  wrote:
 On Tuesday 14 August 2012 10:00:30 Mark Wendt did opine:
> Ebay auction #160861790127
>
> Mark

 A bit rich ($330) for my budget.  I bought one of the DSO NANO's.  For
>>>
>>> what
>>>
 I want to use it for its adequate.  Unplug the usb charging cable and
 its fully isolated.

 Cheers, Gene
 --
 "There are four boxes...
>>>
>>> How do you like the DSO Nano? I've been eyeing them for a while.
>>>
>> About the only thing I haven't mastered is getting the sweep trigger set so 
>> I can see the actual trigger.  Other than that, and a miss-match between 
>> the probe and its frequency comp that distorts a square wave somewhat, it 
>> Just Works(TM).
>>
>> I didn't expect it to match my Hitachi V-1065, but at 5% of the cost ...
>>
>> Cheers, Gene
>> -- 
>> "There are four boxes to be used in defense of liberty:
>> soap, ballot, jury, and ammo. Please use in that order."
>> -Ed Howdershelt (Author)
>> My web page:  is up!
>> The less a statesman amounts to, the more he loves the flag.
>>  -- Kin Hubbard
>>
>> --
>> Live Security Virtual Conference
>> Exclusive live event will cover all the ways today's security and 
>> threat landscape has changed and how IT managers can respond. Discussions 
>> will include endpoint security, mobile security and the latest in malware 
>> threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
>> ___
>> Emc-users mailing list
>> Emc-users@lists.sourceforge.net
>> https://lists.sourceforge.net/lists/listinfo/emc-users
> 
> 
> --
> Live Security Virtual Conference
> Exclusive live event will cover all the ways today's security and 
> threat landscape has changed and how IT managers can respond. Discussions 
> will include endpoint security, mobile security and the latest in malware 
> threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users
> 

--
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and 
threat landscape has changed and how IT managers can respond. Discussions 
will include endpoint security, mobile security and the latest in malware 
threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Brushless Servo Selection?

[Fox Mulder]

Am 15.06.2012 19:58, schrieb Viesturs Lācis:
> 2012/6/15 Kirk Wallace :
>>
>> Building an Antek supply shouldn't be too expensive, but I haven't
>> checked prices recently.
>> http://www.antekinc.com/index.php
>> http://www.antekinc.com/gview.php
>>
> 
> Thanks for the links! 155$ is little more that I would like,
> especially if it is a shipment from US - add 22% VAT and also shipping
> cost...
> 
> But I still got interested to find out, so I looked at this one:
> http://www.antekinc.com/details.php?p=146
> 
> It is rated for 23,4 A and has 4 outputs. A noob question:
> Is it meant to be able to handle 23,4 A on each of the 4 outputs at
> the same time?
> 

The datasheet shows that it has two 32V and one 18V and one 12V output
(Makes in total 4).
On each of the 32V lines you can have up to 23.4A and both other 2A.

Ciao,
 Rainer

--
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and 
threat landscape has changed and how IT managers can respond. Discussions 
will include endpoint security, mobile security and the latest in malware 
threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] 3D Printer Parts

[Fox Mulder]

Am 11.06.2012 19:32, schrieb Joachim Franek:
> On Monday 11 June 2012 02:09:23 Jeshua Lacock wrote:
>> Oh, can anyone recommend a decent RS485 card that will just work with Ubuntu?
> http://www.ftdichip.com/Products/Cables/USBRS485.htm
> 
> But I have not tested it.
> 
> Joachim 

This board should work without problems IF the ftdi_sio module for the
kernel is avaiable. But i don't know if it is compiled for the realtime
kernel.

Ciao,
 Rainer

--
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and 
threat landscape has changed and how IT managers can respond. Discussions 
will include endpoint security, mobile security and the latest in malware 
threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Visolate

[Fox Mulder]

Am 25.03.2012 13:03, schrieb Mark Wendt (Contractor):
> Anybody successfully get visolate to successfully run on a 10.04 system, 
> and if so, what did you do?

There were an older thread about visolate not running from gene (middle
of january). The problem was that visolate couldn't find the java3d
library and therefore exited with an exception.
As an solution i compiled visolate including the java3d libs for him and
than it worked. So firstly try to install java3d and set classpaths
correctly. Just read the older mails where some hints were given.

If it doesn't work i can compile it again with java3d if you want.

Ciao,
 Rainer

--
This SF email is sponsosred by:
Try Windows Azure free for 90 days Click Here 
http://p.sf.net/sfu/sfd2d-msazure
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Z-probing derived kinematics

[Fox Mulder]

Today i wrote a little program which calculates the absolute z distance
between a distorted plane (stl file) at a given xy location and zero. I
used simple triangles and no higher order interpolation. The result was
calculated very fast because of the simple maths used (only + - * /) and
always correct compared to a cad program.

So i think the problems lay beyond the z-depth calculation from the stl
file. I wonder how you process a simple move over a distance which may
span over 3 triangles.

Do you separate this move into three single moves with different start
and end z-depths?

Or do you check before each step the z-depth of the stl file and correct
the commanded z-depth (would be very cpu intensive if done for each
single step)?

Ciao,
 Rainer

Am 20.03.2012 00:00, schrieb Michael Haberler:
> I ran into a problem with following errors if a move is exactly along a 
> triangle edge
> 
> I havent investigated it further but I guess it could be a 
> discontinuity/numerical accuracy/stability problem along the edges
> 
> If that is the case, the approach using discrete-triangle based correction 
> could be fundamentally flawed 
> 
> one way I guess would work is to create a continuous surface correction 
> function, like a cubic spline
> 
> 
> - Michael
> 
> 
> Am 19.03.2012 um 21:03 schrieb Fox Mulder:
> 
>> Since the last mail from december it has gotten really quite about this
>> very handy feature. But i can see in the git logs that it was further
>> developed.
>>
>> So what is the state of it right now and how could it be used?
>>
>> Sadly i can't find any info at all on the linuxcnc website.
>> I want to do some pcb etching tests in the next few weeks and i think
>> this feature would be a great benefit for this purpose. :)
>>
>> Ciao,
>> Rainer
>>
>> Am 21.12.2011 08:31, schrieb Michael Haberler:
>>> I found the idea quite intriguing, and thought about integrating the 
>>> surface correction idea better. After bouncing a few ideas with Andy, this 
>>> is what I have so far:
>>>
>>> probekins: a kinematics module which be default behaves like trivkins
>>> it accepts a mesh of triangles which define Z correction values (actually 
>>> an approximation of the workpiece surface, or machine surface for that 
>>> matter)
>>>
>>> this mesh might eventually generated by probing, but also used as a general 
>>> machine Z correction 
>>>
>>>
>>> this isnt finished or polished, but I'm posting it anyway to see what folks 
>>> think about it.
>>>
>>> see http://git.mah.priv.at/gitweb/emc2-dev.git/shortlog/refs/heads/probekins
>>>
>>> - Michael
>>>
>>> 
>>>
>>>
>>>
>>> v0.1 README:
>>>
>>> Z correction by triangular mesh
>>> ---
>>>
>>> The probekins kins module is a trivial kinematics module unless loaded with 
>>> a Z correction mesh.
>>>
>>> It helps dealing with warped workpiece surfaces (or machines), and was 
>>> inspired by a recent thread on emc-users about PCB milling.
>>>
>>> the correction mesh is a set of triangles specified by an STL file. This 
>>> file might eventually be generated by probing a few points, 
>>> and creating a triangular mesh from it, for instance by a Delauney 
>>> triangulation,
>>>
>>> If a point (x,y) lies within the one of the triangles, it's z value is 
>>> adjusted by the interpolated value on the enclosing triangle.
>>> Points outside any triangle are not corrected for. This means the 
>>> correction value currently 'falls off the cliff' at the convex hull of the 
>>> triangles (the enclosing polygon if you will).
>>>
>>> The triangles are applied in the order specified in the file, so there 
>>> could be a 'later' larger triangle enclosing or overlapping
>>> an earlier one.
>>>
>>> The algorithm currently is based on a brute-force test of all triangles 
>>> until first match. 
>>> The ray/triangle intersection test is very fast, nevertheless this is an 
>>> O(number of triangles) implementation.
>>> There are numerous ways to optimize this which are left as an exercise for 
>>> the reader;).
>>>
>>> To load an STL Z correction file:
>>> -
>>> python stlcorr.py 
>>>
>>> To inspect the current correction:
>>> --
>>> python stlcorr.

Re: [Emc-users] Z-probing derived kinematics

[Fox Mulder]

Since the last mail from december it has gotten really quite about this
very handy feature. But i can see in the git logs that it was further
developed.

So what is the state of it right now and how could it be used?

Sadly i can't find any info at all on the linuxcnc website.
I want to do some pcb etching tests in the next few weeks and i think
this feature would be a great benefit for this purpose. :)

Ciao,
 Rainer

Am 21.12.2011 08:31, schrieb Michael Haberler:
> I found the idea quite intriguing, and thought about integrating the surface 
> correction idea better. After bouncing a few ideas with Andy, this is what I 
> have so far:
> 
> probekins: a kinematics module which be default behaves like trivkins
> it accepts a mesh of triangles which define Z correction values (actually an 
> approximation of the workpiece surface, or machine surface for that matter)
> 
> this mesh might eventually generated by probing, but also used as a general 
> machine Z correction 
> 
> 
> this isnt finished or polished, but I'm posting it anyway to see what folks 
> think about it.
> 
> see http://git.mah.priv.at/gitweb/emc2-dev.git/shortlog/refs/heads/probekins
> 
> - Michael
> 
> 
> 
> 
> 
> v0.1 README:
> 
> Z correction by triangular mesh
> ---
> 
> The probekins kins module is a trivial kinematics module unless loaded with a 
> Z correction mesh.
> 
> It helps dealing with warped workpiece surfaces (or machines), and was 
> inspired by a recent thread on emc-users about PCB milling.
> 
> the correction mesh is a set of triangles specified by an STL file. This file 
> might eventually be generated by probing a few points, 
> and creating a triangular mesh from it, for instance by a Delauney 
> triangulation,
> 
> If a point (x,y) lies within the one of the triangles, it's z value is 
> adjusted by the interpolated value on the enclosing triangle.
> Points outside any triangle are not corrected for. This means the correction 
> value currently 'falls off the cliff' at the convex hull of the 
> triangles (the enclosing polygon if you will).
> 
> The triangles are applied in the order specified in the file, so there could 
> be a 'later' larger triangle enclosing or overlapping
> an earlier one.
> 
> The algorithm currently is based on a brute-force test of all triangles until 
> first match. 
> The ray/triangle intersection test is very fast, nevertheless this is an 
> O(number of triangles) implementation.
> There are numerous ways to optimize this which are left as an exercise for 
> the reader;).
> 
> To load an STL Z correction file:
> -
> python stlcorr.py 
> 
> To inspect the current correction:
> --
> python stlcorr.py
> 
> To clear the correction:
> 
> python stlcorr.py -c
> 
> 
> 
> 
> Example STL file containing two triangles:
> 
> solid
> facet normal 0 1 1 
>   outer loop
> vertex 0 0 0 
> vertex 0 5 0
> vertex 5 0 2
>   endloop
> endfacet
> facet normal 2 3 4
>   outer loop
> vertex 1 1 1 
> vertex -1 -1 1
> vertex 1 -1 2
>   endloop
> endfacet
> endsolid
> -
> The normal vectors are ignored and can be left out for the purpose of 
> describing a correction mesh.

--
This SF email is sponsosred by:
Try Windows Azure free for 90 days Click Here 
http://p.sf.net/sfu/sfd2d-msazure
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] gEDA and PCB-Gcode

[Fox Mulder]

PCB-Gcode is written in ULP Language. As long as gEDA doesn't support
this programming language you can't use it. I think it was originally
written only for eagle.

Ciao,
 Rainer

Am 27.02.2012 01:51, schrieb Gary Fiber:
> Any one on this list successfully using gEDA and PCB with PCB-Gcode to 
> mill printed Circuit Boards?
> 
> Gary Fiber K8IZ
> 
> --
> Try before you buy = See our experts in action!
> The most comprehensive online learning library for Microsoft developers
> is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3,
> Metro Style Apps, more. Free future releases when you subscribe now!
> http://p.sf.net/sfu/learndevnow-dev2
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users
> 


--
Try before you buy = See our experts in action!
The most comprehensive online learning library for Microsoft developers
is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3,
Metro Style Apps, more. Free future releases when you subscribe now!
http://p.sf.net/sfu/learndevnow-dev2
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] More pcb-gcode questions

[Fox Mulder]

Am 04.02.2012 18:22, schrieb gene heskett:
> On Saturday, February 04, 2012 12:04:30 PM Fox Mulder did opine:
> 
>> Am 04.02.2012 13:02, schrieb gene heskett:
>>> On Saturday, February 04, 2012 06:46:40 AM Fox Mulder did opine:
>>>> Am 04.02.2012 08:25, schrieb gene heskett:
>>>>> I have moved these labels one at a time to other layers so I can try
>>>>> 5 layers per run of pbc-gcode, but I'm having zero luck finding the
>>>>> magic tname label that works.
>>>>>
>>>>> So, what layer in eagle do I move these text labels to so they come
>>>>> out in the $basename.top.txt.ngc file?
>>>>>
>>>>> And is there something else I need to do, the pcb-gcode previewer
>>>>> shows total blanks where they should be at all times.
>>>>
>>>> To show the text in the top ngc file you have to put it into the top
>>>> layer. This way it will get milled in the same step as the top traces
>>>> and will be shown in the preview window of pcb-gcode. The TNAME layer
>>>> is only for reference and will never be in the generated gcode.
>>>
>>> But then it carves it as an outline, and pcb-gcode alludes to just
>>> carving the centerlines of the character somehow.  It is quite
>>> readable on the first pass but later passes as the isolation is
>>> increased do a massive job of making the characters illegible. 
>>> pcb-gcode does generate the files but they show as blank in the
>>> preview viewer, and contain no code that touches the board for either
>>> .top or .bot. files.  The font is set to vector.
>>>
>>> I there something I can set in the eagle 'info' screen for these
>>> characters that will tell pcb-gcode that this is text, treat it
>>> accordingly?
>>
>> Ah ok now i know what you wanted. For that you have to put the text into
>> the milling layer (46). If the text is normal it will be on the top side
>> and when mirrored it appears on the bottom side. In pcb-gcode you have
>> to activate "generate text" in the "generation options" tab. Than after
>> generation you will see a preview of the text and the output for the
>> text is in a file called $basename.top.text.ngc and
>> $basename.bot.text.ngc.
>>
>> I tried it myself and it works as expected for engraving result. :)
>>
>> I doesn't like the way that pcb-gcode normally will mill away all copper
>> which is not used. Therefore i put a polygon over the whole board with a
>> gap of ~0,2mm between all traces. Than pcb-gcode will only mill away
>> these small 0.2mm width lines around the traces which is much faster and
>> preserves your precious router bit. :)
> 
> Which is what it is doing now, so I wonder if we have the same version of 
> pcb-gcode.  My version only mills away to the maximum isolation setting & 
> anything outside of that is left. However, by golly that worked, and even 
> has the right polarity, showing the white characters on a black background.  
> On this point I am a happy camper again, thank you very much Rainer.

Hmm ok, maybe i forgot that pcb-gcode already does it this way. :)
I've started using the technique with polygons for GND layer over the
whole board since many years back from the etching time where it saved
time and acid.

To be honest i never really used pcb-gcode for milling a board because i
never found perfect settings. And my hobby mill doesn't have such a
constant z-depth that the layouts would be good enough over a longer
track. Maybe with help of the automatic depth-probing which was
discussed lately i can start experimenting again. :)

Ciao,
 Rainer

--
Try before you buy = See our experts in action!
The most comprehensive online learning library for Microsoft developers
is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3,
Metro Style Apps, more. Free future releases when you subscribe now!
http://p.sf.net/sfu/learndevnow-dev2
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] More pcb-gcode questions

[Fox Mulder]

Am 04.02.2012 13:02, schrieb gene heskett:
> On Saturday, February 04, 2012 06:46:40 AM Fox Mulder did opine:
> 
>> Am 04.02.2012 08:25, schrieb gene heskett:
>>> I have moved these labels one at a time to other layers so I can try 5
>>> layers per run of pbc-gcode, but I'm having zero luck finding the
>>> magic tname label that works.
>>>
>>> So, what layer in eagle do I move these text labels to so they come
>>> out in the $basename.top.txt.ngc file?
>>>
>>> And is there something else I need to do, the pcb-gcode previewer
>>> shows total blanks where they should be at all times.
>>
>> To show the text in the top ngc file you have to put it into the top
>> layer. This way it will get milled in the same step as the top traces
>> and will be shown in the preview window of pcb-gcode. The TNAME layer is
>> only for reference and will never be in the generated gcode.
> 
> But then it carves it as an outline, and pcb-gcode alludes to just carving 
> the centerlines of the character somehow.  It is quite readable on the 
> first pass but later passes as the isolation is increased do a massive job 
> of making the characters illegible.  pcb-gcode does generate the files but 
> they show as blank in the preview viewer, and contain no code that touches 
> the board for either .top or .bot. files.  The font is set to vector.
> 
> I there something I can set in the eagle 'info' screen for these characters 
> that will tell pcb-gcode that this is text, treat it accordingly?

Ah ok now i know what you wanted. For that you have to put the text into
the milling layer (46). If the text is normal it will be on the top side
and when mirrored it appears on the bottom side. In pcb-gcode you have
to activate "generate text" in the "generation options" tab. Than after
generation you will see a preview of the text and the output for the
text is in a file called $basename.top.text.ngc and $basename.bot.text.ngc.

I tried it myself and it works as expected for engraving result. :)

I doesn't like the way that pcb-gcode normally will mill away all copper
which is not used. Therefore i put a polygon over the whole board with a
gap of ~0,2mm between all traces. Than pcb-gcode will only mill away
these small 0.2mm width lines around the traces which is much faster and
preserves your precious router bit. :)

Ciao,
 Rainer

--
Try before you buy = See our experts in action!
The most comprehensive online learning library for Microsoft developers
is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3,
Metro Style Apps, more. Free future releases when you subscribe now!
http://p.sf.net/sfu/learndevnow-dev2
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] More pcb-gcode questions

[Fox Mulder]

Am 04.02.2012 08:25, schrieb gene heskett:
> I have moved these labels one at a time to other layers so I can try 5 
> layers per run of pbc-gcode, but I'm having zero luck finding the magic 
> tname label that works.
> 
> So, what layer in eagle do I move these text labels to so they come out in 
> the $basename.top.txt.ngc file?  
> 
> And is there something else I need to do, the pcb-gcode previewer shows 
> total blanks where they should be at all times.

To show the text in the top ngc file you have to put it into the top
layer. This way it will get milled in the same step as the top traces
and will be shown in the preview window of pcb-gcode. The TNAME layer is
only for reference and will never be in the generated gcode.

Ciao,
 Rainer

--
Try before you buy = See our experts in action!
The most comprehensive online learning library for Microsoft developers
is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3,
Metro Style Apps, more. Free future releases when you subscribe now!
http://p.sf.net/sfu/learndevnow-dev2
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] [OT] need advice from old hands

[Fox Mulder]

Am 22.01.2012 05:24, schrieb gene heskett:
> On Saturday, January 21, 2012 11:14:36 PM Fox Mulder did opine:
> 
>> Am 21.01.2012 20:41, schrieb gene heskett:
>>> Stripping it down to the holes, via, and dimensions and running
>>> pcb-gcode- setup.ulp gets me an instant can't open source/drill.h. 
>>> So I'll have to generate that with gedit I suppose.  The source
>>> directory doesn't exist yet.
>>
>> That is strange because within the original pcb-gcode zip file the
>> source folder exists, including drill.h. Seems to me that you have some
>> problems with the directory structure for pcb-gcode. Did you extract ALL
>> the files from within the zip file into some directory and ran the ulp
>> file from within it?
>>
>> Ciao,
>>  Rainer
>>
> That is what I thought I was doing.  Somewhere in one of the readme's it 
> said to put it in ones /home/gene/eagle tree, so currently when I an cd'd 
> to where I can see the drill.h file, a pwd says:
> 
> /home/gene/eagle/ulp/source
> 
> The two installs aren't exactly identical, eagle is installed on the shop 
> box as /opt/eagle-6.1.0, where on this box it is in /opt/eagle/eagle-6.1.0.
> 
> Do I need to set another env variable?
> 
> eagle has no problem seeing /home/gene/eagle/ulp/pcb-gcode-setup.ulp, but 
> throws the error 50ms after the dbl click to run it.
> 
> Clues gleefully checked out.  :)

On windows i just extracted all the content of pcb-gcode.zip to a
directory like c:\pcb-gcode. Than set the ulp and script path (don't
know if script path is really needed) in eagle to c:\pcb-gcode as first
entry. That's it and it works without problems. :)

Ciao,
 Rainer

--
Try before you buy = See our experts in action!
The most comprehensive online learning library for Microsoft developers
is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3,
Metro Style Apps, more. Free future releases when you subscribe now!
http://p.sf.net/sfu/learndevnow-dev2
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] [OT] need advice from old hands

[Fox Mulder]

Am 21.01.2012 20:41, schrieb gene heskett:
> Stripping it down to the holes, via, and dimensions and running pcb-gcode-
> setup.ulp gets me an instant can't open source/drill.h.  So I'll have to 
> generate that with gedit I suppose.  The source directory doesn't exist 
> yet.

That is strange because within the original pcb-gcode zip file the
source folder exists, including drill.h. Seems to me that you have some
problems with the directory structure for pcb-gcode. Did you extract ALL
the files from within the zip file into some directory and ran the ulp
file from within it?

Ciao,
 Rainer

--
Try before you buy = See our experts in action!
The most comprehensive online learning library for Microsoft developers
is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3,
Metro Style Apps, more. Free future releases when you subscribe now!
http://p.sf.net/sfu/learndevnow-dev2
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] need advice from old hands

[Fox Mulder]

Am 21.01.2012 14:40, schrieb gene heskett:
> On Saturday, January 21, 2012 07:15:30 AM Fox Mulder did opine:
> 
>> Am 21.01.2012 00:22, schrieb gene heskett:
>>> Greetings all;
>>>
>>> I have thrown a schematic together, but before I commit to the
>>> connector for the cable, which will need at least 5 conductors, GND,
>>> Vcc5, A, B, Z signals and needs to be fairly compact, I am drawn to
>>> the pcb mounted version of the RJ-11 connector.  But sitting here, I
>>> realized that although it is supposed to be a 6 pin connector, I have
>>> never seen one with more than 4 positions populated, nor have I seen
>>> The matching 6 wire cable. I considered the RJ-45 too, but have not
>>> seen any pcb mount jacks in my neck of these woods.  A db9 is just
>>> plain too darned big unless the backshell is thrown in the corner.
>>
>> If you don't need too much current over the cable than RJ45 is a good
>> connector. The pcb mounted connector should be available at most
>> electronic shops and you can just use finished network cables to connect
>> the device.
>> But be sure to use network cables which are specified for 1000MBit
>> because 10/100MBit only needs 4 wires and therefore many old cables
>> won't have all the 8 wires. ;)
>>
>> Ciao,
>>  Rainer
> 
> Any cat5 cable I've ever seen has 4 twisted pairs on this side of the pond.  
> This place has about 400 feet of cat5 strung about and its all just 
> 100mbit.  I think the router I bought last summer can do a gigabit but I 
> don't believe either side of it is running that fast.
> 
> I did find some crimp on 6 wire plugs at the shack last night, and some 25 
> foot wall to phone cables but the only females were either in the usual 2" 
> square boxes or in line couplings, half of which were still too big if they 
> had been pried apart.  So I settled on a 5 way inline with setscrews to 
> retain the wires.  A lot easier to breadboard for testing too.  Later last 
> night I fine tuned the .brd file a bit, finally figuring out how adjust a 
> hole so it didn't snap to the grid which because of that parts foot print I 
> had to set to 0.05 inches.
> 
> But when I tried to run pcbgcode --config I think it was, I got a missing 
> drill.cfg (or something similar) so when I am awake next, I'll need to read 
> up on that.  What I did read seems to indicate all drill holes are in the 
> same file regardless of size, and there is only one 'layer' for unplated 
> holes.  I would have thought it would put each size of drill in its own 
> layer so that one could mount a given size drill, and run the output file 
> for that size of drill.  Probably something I don't understand yet.  The 
> only real problem I had with the board layout is that I had to repeatedly 
> ripup the individual nets by hand, all of them, before the auto function 
> would even try to fix things when I moved parts to actually fit the board 
> real estate I can use.  Seems to me if there is a Drc error, it should do 
> an auto-ripup and start all over, but I didn't find that magic twanger to 
> do that.  But I got it done, so its not a show stopper by any means.

I only wanted to say that the standard for 10/100MBit only uses 2 pairs
and 1000MBit was the first to use all 4 pairs. But you are right that
most cat5 cables used all 4 pairs way before 1000MBit over copper was
available. But i have some old cables that still uses 2 pairs which
could lead to an error for your case. :)

It was pcb-gcode-setup.ulp but maybe it wraps around with --config.
The drills are all in the same drill file. But for each new drill size a
tool change code is inserted. So you could extract the code for each
drill size into separate files or use the tool change command.

To be honest i never liked the output of the auto-router and route all
layouts by hand. The auto-router could be useful for complex layouts but
not for simpler boards.

Ciao,
 Rainer

--
Try before you buy = See our experts in action!
The most comprehensive online learning library for Microsoft developers
is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3,
Metro Style Apps, more. Free future releases when you subscribe now!
http://p.sf.net/sfu/learndevnow-dev2
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] need advice from old hands

[Fox Mulder]

Am 21.01.2012 00:22, schrieb gene heskett:
> Greetings all;
> 
> I have thrown a schematic together, but before I commit to the connector 
> for the cable, which will need at least 5 conductors, GND, Vcc5, A, B, Z 
> signals and needs to be fairly compact, I am drawn to the pcb mounted 
> version of the RJ-11 connector.  But sitting here, I realized that although 
> it is supposed to be a 6 pin connector, I have never seen one with more 
> than 4 positions populated, nor have I seen The matching 6 wire cable. I 
> considered the RJ-45 too, but have not seen any pcb mount jacks in my neck 
> of these woods.  A db9 is just plain too darned big unless the backshell is 
> thrown in the corner.

If you don't need too much current over the cable than RJ45 is a good
connector. The pcb mounted connector should be available at most
electronic shops and you can just use finished network cables to connect
the device.
But be sure to use network cables which are specified for 1000MBit
because 10/100MBit only needs 4 wires and therefore many old cables
won't have all the 8 wires. ;)

Ciao,
 Rainer

--
Try before you buy = See our experts in action!
The most comprehensive online learning library for Microsoft developers
is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3,
Metro Style Apps, more. Free future releases when you subscribe now!
http://p.sf.net/sfu/learndevnow-dev2
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] eagle-6.1.0 (again)

[Fox Mulder]

Am 17.01.2012 21:47, schrieb gene heskett:
> Greets all;
> 
> Going thru the docs for eagle, in this case following the tutorial_en.pdf,
> resetting the directories to where they exist on that machine seems to 
> stick over an exit and restart.  Checking requirements, I see that 
> libpng14.so.14 is required so you helped me build a script:
> gene@shop:/opt/eagle-6.1.0$ cat bin/eagle
> LD_LIBRARY_PATH="$(dirname "$0")" "$(dirname "$0")/eagle.orig" $@
> 
> It doesn't complain about a missing library so I assume that is working, 
> but for a universal launcher, should I not have a /usr/bin bash script that 
> first does a cd to /opt/eagle-6.1.0/ in order to setup the pwd, and then do 
> a 'bin/eagle'?  In any event, it seems to work.
> 
> I have eagle open and showing my project(empty of course) highlighted with 
> a green indicator button dot in the left pane of the window.
> 
> Reading along in the tut and seeing a few things as I go along in its 
> description, I get to pages 12-13-14 where it says
>  'Option->Menu etc' or Options/Set, Color tab,
> 
> But when I open the option menu, I only have 4 items:
> 
> Directories
> Backup
> User Interface
> Window positions

These settings are only available when you are in the schematic or board
window, not main window.

> I've been thru the user interface panel & changed a few things, like 
> inverting the foreground/background colors used, but with no effects seen.

These settings only apply for the schematics and board window, not main
window.

> The settings do seem to stick over a quit & restart ok.
> 
> I don't seem to be able to create an empty schematic page, only ever deeper 
> directories named for the page.sch I want.  This under the 'file' menu.

Normally you just have to click file->new->schematic to create an empty
schematic window. You can also right click on the selected project
folder (green dot at the right of it) and select new->schematic to start.

> 
> I can open the libraries and select the device, which then opens in its own 
> window showing it both schematically and a footprint outline but despite 
> reading how to select and use the device, I have no 'schematic' window to 
> copy it into, so none of those commands do anything, not even blinking the 
> screen to indicate an error.  The claim is made that a device should be 
> hanging from the cursor when selected, but its never happened.  It is also 
> single threaded making it impossible to do a drag & drop because as long as 
> the library window is open, all others appear to be unfocusable, greyed 
> out.  I get the impression that when I am looking at the selected device in 
> the library's window, that I am actually looking at the device in its 
> editor window, presumably not the select and use window.  And no clue how 
> to get there except I have now gotten there probably 20 times by opening 
> the library, then opening the library in that window and clicking on the 
> now visible devices name which appears to open an editor to edit the 
> device.
> 
> Do I have a defective install?  or am I just too damned dumb??

This is the wrong way of working with eagle. When you open a library
than only if you want to create/edit/delete parts in the library. For
using it you go through the schematic/board window and use the "ADD"
function which uses these libraries.

> 
> I have also downloaded & unpacked pcb-gcode-3.5.2.11.zip, which seems to be 
> a fairly major suite of tools in its own right, but haven't attempted to 
> sort that basket of rattlesnakes just yet.  Looks simple enough, unzip it 
> to a 'ulp' directory & add that to the ulp directory path, so now I have 2 
> ulp directories. One with the contents of that zip in it.  Kewl.  But in 
> eagle, I have yet to find that command line box talked about in the tuts so 
> I can't run the setup utility it talks about in its install.html.  And I 
> just noted that this added ulp directory, built where the docs said to 
> build it, also shows up in the projects menu.  No clue if that is good or 
> bad, just commenting.

Instead of using the commandline in eagle to run it you can (in the
schematic/board window) use file->execute ULP and select the
corresponding ulp file like pcb-gcode-setup.ulp which does the same.

Ciao,
 Rainer

--
Keep Your Developer Skills Current with LearnDevNow!
The most comprehensive online learning library for Microsoft developers
is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3,
Metro Style Apps, more. Free future releases when you subscribe now!
http://p.sf.net/sfu/learndevnow-d2d
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] New thread, visolate

[Fox Mulder]

Am 15.01.2012 21:50, schrieb gene heskett:
> On Sunday, January 15, 2012 03:38:17 PM gene heskett did opine:
> I just found 2 bugs in visolate.jar
> 
> 1. The F speeds output I had entered as 4.0ipm because my spindles max rpms 
> is 2500.  But it is multiplied by 60 in the gcode to F240 etc, so feed 
> rates will need to be entered in ips, not ipm.
> 
> 2. It can also do outline milling, possibly doing what pcb-gcode would do, 
> but once asked to, it cannot be changed back to voroni without shutting it 
> down and restarting.  Changing the checkmark doesn't change it back... ;-(  
> Neither of these are showstoppers.
> 
> Now, what do most of you folks use for converting an eagle's 'excellon' 
> drill file to gcode?  This might be the next stumbling block here.

You could report your findings on the visolate website as bugs.

And i only use pcb-gcode to generate emc gcode for milling and drilling
out of brd files. So why bother with visolate for this? :)

Ciao,
 Rainer

--
RSA(R) Conference 2012
Mar 27 - Feb 2
Save $400 by Jan. 27
Register now!
http://p.sf.net/sfu/rsa-sfdev2dev2
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] New thread, visolate

[Fox Mulder]

Am 15.01.2012 18:35, schrieb gene heskett:
> On Sunday, January 15, 2012 11:59:22 AM Fox Mulder did opine:
> 
>> Am 15.01.2012 05:25, schrieb gene heskett:
>>> Hi all;
>>>
>>> I just spent 2 hours ogling strace output while trying to get this
>>> visolate.jar to do something even if it was to tell me to go away.
>>>
>>> What I was doing was looking at the strace output for the first
>>> instance of each library etc that it wanted to open, did a mkdir -p
>>> to that location, then did an 'ls -l `locate libname' that it could
>>> not find, than added a softlink from the real file to that name, in
>>> that directory.  After 4 or 5 sessions of that, I no longer had any
>>> file not found's output by the strace. It was finding everything it
>>> wanted in that dept.
>>>
>>> This didn't effect the actual exit messages!
>>>
>>> But what I now see in the strace output is a fail of a FUTEX that
>>> could have been there all along, whatever the heck that futex is. 
>>> From that strace:
>>>
>>> [snip lots of strace success lines]
>>>
>>> futex(0xb6a95bd8, FUTEX_WAIT, 20248, NULLException in thread "main"
>>> java.lang.NoClassDefFoundError: javax/media/j3d/WakeupCriterion
>>>
>>> at visolate.Visolate.(Visolate.java:66)
>>> at visolate.Visolate.(Visolate.java:61)
>>> at visolate.Main.main(Main.java:65)
>>>
>>> Caused by: java.lang.ClassNotFoundException:
>>> javax.media.j3d.WakeupCriterion
>>>
>>> at java.net.URLClassLoader$1.run(URLClassLoader.java:202)
>>> at java.security.AccessController.doPrivileged(Native Method)
>>> at java.net.URLClassLoader.findClass(URLClassLoader.java:190)
>>> at java.lang.ClassLoader.loadClass(ClassLoader.java:306)
>>> at
>>> sun.misc.Launcher$AppClassLoader.loadClass(Launcher.java:301)
>>> at java.lang.ClassLoader.loadClass(ClassLoader.java:247) ...
>>> 3 more
>>>
>>> ) = 0
>>> exit_group(1)
>>>
>>>   = ?
>>>
>>> And I get the bash prompt back.
>>>
>>> Another strange thing is that there is an /etc/java/java.conf file
>>> that says where the .jars are supposed to be, but its being ignored,
>>> I must pass the full path to the .jar, or the error is "can't find
>>> it" in java speak.
>>>
>>> Is any of this making sense to anybody?  At this point, I'm bumfuzzled
>>> (or whatever your fav expression for confused is) and would seem to
>>> be getting worse.  Oldtimers is hell.
>>
>> I don't know why but it seems to be easier to reference the java3d libs
>> in windows than in linux. I tried it myself and i can't get visolate to
>> find the libs in linux. So i decided to refactor the jar in eclipse and
>> include the required java3d libs into the jar. So for now only the
>> native opengl libs from the java3d package are needed anymore. And this
>> is quite easy.
>> Just download my new created visolate.jar from [1] and set the ld path
>> correctly. You need to extract the two files "lib3dcode-ogl*.so" from
>> the java3d zip file from [2] and copy them where you want.
> 
> For this, I chose to follow the linkage in /etc/alternatives/jre, which 
> links to /usr/lib/jvm/jre-1.6.0-sun, and copied them to the subdirs in the 
> zip.  Then added +x perms to match the rest of those dirs contents.
> 
>> I copied them
>> to /home/user/rainer/lib and after that in the console do "export
>> LD_LIBRARY_PATH=/home/user/rainer/lib".
> 
> I used /etc/alternatives/jre as the exported path.
> 
>> Now you can run the jar file
>> with "java -jar visolate.jar".
>> But the export statement is only active in the shell where it was
>> executed.
> 
> Which is why I put that in my .bashrc, so it is (hopefully) not forgotten 
> on a reboot.  I opened a new shell but must not have put the new 
> visolate.jar in the right place,
> 
> [gene@coyote ~]$ java -jar visolate.jar
> Unable to access jarfile visolate.jar
> [gene@coyote ~]$
> 
> So I'll use the strace output and put it the first place it looks for it.
> 
> Damn, no attempt to even follow the usual java culprits path!
> 
> open("visolate.jar", O_RDONLY)  = -1 ENOENT (No such file or 
> directory)
> write(2, "Unable to access jarfile visolat"..., 37Unable to access j

Re: [Emc-users] New thread, visolate

[Fox Mulder]

Am 15.01.2012 05:25, schrieb gene heskett:
> Hi all;
> 
> I just spent 2 hours ogling strace output while trying to get this 
> visolate.jar to do something even if it was to tell me to go away.
> 
> What I was doing was looking at the strace output for the first instance of 
> each library etc that it wanted to open, did a mkdir -p to that location, 
> then did an 'ls -l `locate libname' that it could not find, than added a 
> softlink from the real file to that name, in that directory.  After 4 or 5 
> sessions of that, I no longer had any file not found's output by the 
> strace. It was finding everything it wanted in that dept.
> 
> This didn't effect the actual exit messages!
> 
> But what I now see in the strace output is a fail of a FUTEX that could 
> have been there all along, whatever the heck that futex is.  From that 
> strace:
> 
> [snip lots of strace success lines]
> 
> futex(0xb6a95bd8, FUTEX_WAIT, 20248, NULLException in thread "main" 
> java.lang.NoClassDefFoundError: javax/media/j3d/WakeupCriterion
> at visolate.Visolate.(Visolate.java:66)
> at visolate.Visolate.(Visolate.java:61)
> at visolate.Main.main(Main.java:65)
> Caused by: java.lang.ClassNotFoundException: 
> javax.media.j3d.WakeupCriterion
> at java.net.URLClassLoader$1.run(URLClassLoader.java:202)
> at java.security.AccessController.doPrivileged(Native Method)
> at java.net.URLClassLoader.findClass(URLClassLoader.java:190)
> at java.lang.ClassLoader.loadClass(ClassLoader.java:306)
> at sun.misc.Launcher$AppClassLoader.loadClass(Launcher.java:301)
> at java.lang.ClassLoader.loadClass(ClassLoader.java:247)
> ... 3 more
> ) = 0
> exit_group(1) 
>   = ?
> And I get the bash prompt back.
> 
> Another strange thing is that there is an /etc/java/java.conf file that 
> says where the .jars are supposed to be, but its being ignored, I must pass 
> the full path to the .jar, or the error is "can't find it" in java speak.
> 
> Is any of this making sense to anybody?  At this point, I'm bumfuzzled (or 
> whatever your fav expression for confused is) and would seem to be getting 
> worse.  Oldtimers is hell.

I don't know why but it seems to be easier to reference the java3d libs
in windows than in linux. I tried it myself and i can't get visolate to
find the libs in linux. So i decided to refactor the jar in eclipse and
include the required java3d libs into the jar. So for now only the
native opengl libs from the java3d package are needed anymore. And this
is quite easy.
Just download my new created visolate.jar from [1] and set the ld path
correctly. You need to extract the two files "lib3dcode-ogl*.so" from
the java3d zip file from [2] and copy them where you want. I copied them
to /home/user/rainer/lib and after that in the console do "export
LD_LIBRARY_PATH=/home/user/rainer/lib". Now you can run the jar file
with "java -jar visolate.jar".
But the export statement is only active in the shell where it was executed.

Ciao,
 Rainer

[1] http://quakeman.homelinux.net/files/visolate.jar
[2]
http://download.java.net/media/java3d/builds/release/1.5.2/j3d-1_5_2-linux-i586.zip

--
RSA(R) Conference 2012
Mar 27 - Feb 2
Save $400 by Jan. 27
Register now!
http://p.sf.net/sfu/rsa-sfdev2dev2
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] gEDA / Correcting for workpiece warpage.

[Fox Mulder]

Am 13.01.2012 23:41, schrieb gene heskett:
> On Friday, January 13, 2012 05:13:38 PM Fox Mulder did opine:
> 
>> Am 13.01.2012 22:04, schrieb gene heskett:
>>> On Friday, January 13, 2012 03:54:18 PM Fox Mulder did opine:
>>>> Am 13.01.2012 18:38, schrieb gene heskett:
>>>>> How does one run this visolate-2.1.6.jar then?  The usual "java -jar
>>>>> visolate-2.1.6.jar" without a valid input filename doesn't output
>>>>> any obvious clues to a java newbie:
>>>>> gene@coyote pcb2gcode-1.1.4]$ java -jar
>>>>> /usr/share/java/visolate-2.1.6.jar Exception in thread "main"
>>>>> java.lang.NoClassDefFoundError:
>>>>> javax/media/j3d/WakeupCriterion
>>>>>
>>>>> at visolate.Visolate.(Visolate.java:66)
>>>>> at visolate.Visolate.(Visolate.java:61)
>>>>> at visolate.Main.main(Main.java:65)
>>>>>
>>>>> Caused by: java.lang.ClassNotFoundException:
>>>>> javax.media.j3d.WakeupCriterion
>>>>>
>>>>> at java.net.URLClassLoader$1.run(URLClassLoader.java:202)
>>>>> at java.security.AccessController.doPrivileged(Native
>>>>> Method) at
>>>>> java.net.URLClassLoader.findClass(URLClassLoader.java:190)
>>>>> at java.lang.ClassLoader.loadClass(ClassLoader.java:306) at
>>>>> sun.misc.Launcher$AppClassLoader.loadClass(Launcher.java:301
>>>>> ) at java.lang.ClassLoader.loadClass(ClassLoader.java:247)
>>>>> ... 3 more
>>>>
>>>> Your are missing the Java3D package. Just download from oracle
>>>> website and install it. after that it should run.
>>>
>>> Actually it was in the repo for pclos, but wasn't installed.  Tis now.
>>> But that seems not to be a lot of help:
>>>
>>> [gene@coyote pcb2gcode-1.1.4]$ java -jar
>>> /usr/share/java/visolate-2.1.6.jar --help
>>> Exception in thread "main" java.lang.NoClassDefFoundError:
>>> javax/media/j3d/WakeupCriterion
>>>
>>> at visolate.Visolate.(Visolate.java:66)
>>> at visolate.Visolate.(Visolate.java:61)
>>> at visolate.Main.main(Main.java:65)
>>>
>>> Caused by: java.lang.ClassNotFoundException:
>>> javax.media.j3d.WakeupCriterion
>>>
>>> at java.net.URLClassLoader$1.run(URLClassLoader.java:202)
>>> at java.security.AccessController.doPrivileged(Native Method)
>>> at java.net.URLClassLoader.findClass(URLClassLoader.java:190)
>>> at java.lang.ClassLoader.loadClass(ClassLoader.java:306)
>>> at
>>> sun.misc.Launcher$AppClassLoader.loadClass(Launcher.java:301)
>>> at java.lang.ClassLoader.loadClass(ClassLoader.java:247) ...
>>> 3 more
>>>
>>> The installed java is 1.6.0-sun, but from the pclos repos.
>>
>> Seems that java somehow can't find the java3d libs. I tried it in
>> windows with the same error message. After instaling java3d it runs
>> without problem. You could try the runtime prameter "-cp
>> path/to/java3d/libs" to explicitly show java the path to the installed
>> libs.
>>
>> Caio,
>>  Rainer
>>
> Would it not be better to just move the .jar where its looking?  It seems 
> that /usr/share/java isn't exactly the root java directory.  Let me do an 
> ls -R on that tree as it exists on this pclos box:
> 
> [root@coyote java]# ls -R
> .:
> fmj/  gcj-endorsed/  java3d/  visolate-2.1.6.jar
> 
> ./fmj:
> ffmpeg-java-20071012.jar  ffmpeg-java-gpl-20071012.jar  ffmpeg-java-
> gpl.jar@  ffmpeg-java.jar@  visolate-2.1.6.jar
> 
> ./gcj-endorsed:
> 
> ./java3d:
> j3dcore.jar  j3dutils.jar  vecmath.jar
> 
> The java3d stuff is in a separate directory. Should I link or move its 
> ./java3d contents?  Or is there an ENV var that isn't set?
> 
> I put a softlink in /etc/alternatives to /usr/share/java/java3d/ as java3d, 
> the link looks good and I can ls it, but no change in the error stanza.

I don't know which is the correct location for these three files in
linux. In windows they were installed within the jre subdir under
/jre6/lib/ext.
I found a description for ubuntu at [1]. Maybe the mentioned dir on this
site helps you.

Ciao,
 Rainer

[1] http://wiki.ubuntuusers.de/java3d

--
RSA(R) Conference 2012
Mar 27 - Feb 2
Save $400 by Jan. 27
Register now!
http://p.sf.net/sfu/rsa-sfdev2dev2
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] gEDA / Correcting for workpiece warpage.

[Fox Mulder]

Am 13.01.2012 22:04, schrieb gene heskett:
> On Friday, January 13, 2012 03:54:18 PM Fox Mulder did opine:
> 
>> Am 13.01.2012 18:38, schrieb gene heskett:
>>> How does one run this visolate-2.1.6.jar then?  The usual "java -jar
>>> visolate-2.1.6.jar" without a valid input filename doesn't output any
>>> obvious clues to a java newbie:
>>> gene@coyote pcb2gcode-1.1.4]$ java -jar
>>> /usr/share/java/visolate-2.1.6.jar Exception in thread "main"
>>> java.lang.NoClassDefFoundError:
>>> javax/media/j3d/WakeupCriterion
>>>
>>> at visolate.Visolate.(Visolate.java:66)
>>> at visolate.Visolate.(Visolate.java:61)
>>> at visolate.Main.main(Main.java:65)
>>>
>>> Caused by: java.lang.ClassNotFoundException:
>>> javax.media.j3d.WakeupCriterion
>>>
>>> at java.net.URLClassLoader$1.run(URLClassLoader.java:202)
>>> at java.security.AccessController.doPrivileged(Native Method)
>>> at java.net.URLClassLoader.findClass(URLClassLoader.java:190)
>>> at java.lang.ClassLoader.loadClass(ClassLoader.java:306)
>>> at
>>> sun.misc.Launcher$AppClassLoader.loadClass(Launcher.java:301)
>>> at java.lang.ClassLoader.loadClass(ClassLoader.java:247) ...
>>> 3 more
>>
>> Your are missing the Java3D package. Just download from oracle website
>> and install it. after that it should run.
>>
> Actually it was in the repo for pclos, but wasn't installed.  Tis now. But 
> that seems not to be a lot of help:
> 
> [gene@coyote pcb2gcode-1.1.4]$ java -jar /usr/share/java/visolate-2.1.6.jar 
> --help
> Exception in thread "main" java.lang.NoClassDefFoundError: 
> javax/media/j3d/WakeupCriterion
> at visolate.Visolate.(Visolate.java:66)
> at visolate.Visolate.(Visolate.java:61)
> at visolate.Main.main(Main.java:65)
> Caused by: java.lang.ClassNotFoundException: 
> javax.media.j3d.WakeupCriterion
> at java.net.URLClassLoader$1.run(URLClassLoader.java:202)
> at java.security.AccessController.doPrivileged(Native Method)
> at java.net.URLClassLoader.findClass(URLClassLoader.java:190)
> at java.lang.ClassLoader.loadClass(ClassLoader.java:306)
> at sun.misc.Launcher$AppClassLoader.loadClass(Launcher.java:301)
> at java.lang.ClassLoader.loadClass(ClassLoader.java:247)
> ... 3 more
> 
> The installed java is 1.6.0-sun, but from the pclos repos.

Seems that java somehow can't find the java3d libs. I tried it in
windows with the same error message. After instaling java3d it runs
without problem. You could try the runtime prameter "-cp
path/to/java3d/libs" to explicitly show java the path to the installed libs.

Caio,
 Rainer

--
RSA(R) Conference 2012
Mar 27 - Feb 2
Save $400 by Jan. 27
Register now!
http://p.sf.net/sfu/rsa-sfdev2dev2
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] gEDA / Correcting for workpiece warpage.

[Fox Mulder]

Am 13.01.2012 18:38, schrieb gene heskett:
> How does one run this visolate-2.1.6.jar then?  The usual "java -jar 
> visolate-2.1.6.jar" without a valid input filename doesn't output any 
> obvious clues to a java newbie:
> gene@coyote pcb2gcode-1.1.4]$ java -jar /usr/share/java/visolate-2.1.6.jar
> Exception in thread "main" java.lang.NoClassDefFoundError: 
> javax/media/j3d/WakeupCriterion
> at visolate.Visolate.(Visolate.java:66)
> at visolate.Visolate.(Visolate.java:61)
> at visolate.Main.main(Main.java:65)
> Caused by: java.lang.ClassNotFoundException: 
> javax.media.j3d.WakeupCriterion
> at java.net.URLClassLoader$1.run(URLClassLoader.java:202)
> at java.security.AccessController.doPrivileged(Native Method)
> at java.net.URLClassLoader.findClass(URLClassLoader.java:190)
> at java.lang.ClassLoader.loadClass(ClassLoader.java:306)
> at sun.misc.Launcher$AppClassLoader.loadClass(Launcher.java:301)
> at java.lang.ClassLoader.loadClass(ClassLoader.java:247)
> ... 3 more

Your are missing the Java3D package. Just download from oracle website
and install it. after that it should run.

Ciao,
 Rainer

--
RSA(R) Conference 2012
Mar 27 - Feb 2
Save $400 by Jan. 27
Register now!
http://p.sf.net/sfu/rsa-sfdev2dev2
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] gEDA / Correcting for workpiece warpage.

[Fox Mulder]

Am 12.01.2012 19:38, schrieb Kirk Wallace:
> On Thu, 2012-01-12 at 12:55 -0500, gene heskett wrote:
>> On Thursday, January 12, 2012 12:43:08 PM andy pugh did opine:
>>
>>> On 12 January 2012 07:22, gene heskett  wrote:
>> [...] 
>>> Have you tried DesignSpark under Wine?
>>> Though gEDA might be a better bet, being native Linux. (I have never
>>> tried it)
>>  
>> I have looked at gEDA and it shows promise, but it needs to grow some std 
>> method of sizing the parts to an agreed upon std measurement method.  When 
>> the parts library is 100% contributor generated, no two parts seem to be 
>> drawn to the same scale or orientation.
> 
> If one can use it often enough to keep current, maybe once a week, I
> find gEDA much easier to use than Eagle. Eagle has an extensive library,
> but I suspect that is because it takes a PhD in Eagle to create them, so
> an interested party made sure the popular symbols were available.

To be honest i think it is quite easy to create new parts in eagle. When
i first needed it i wen through the pdf tutorial about creating new
parts. And after i made two or three parts myself it was very easy to
create new parts. :)
I agree that it isn't easy enough so that everybody can do it right
after first start. But i think most specialized programs need some time
to master it.

Ciao,
 Rainer

--
RSA(R) Conference 2012
Mar 27 - Feb 2
Save $400 by Jan. 27
Register now!
http://p.sf.net/sfu/rsa-sfdev2dev2
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] openscad -> emc

[Fox Mulder]

Am 24.12.2011 20:57, schrieb gene heskett:
> On Saturday, December 24, 2011 02:56:23 PM matthew venn did opine:
> 
>> Thanks for the tips!
>> Will try pycam. Anyone heard of skeinforge? I got recommended that too.
>> Apparently it has a milling profile as well as 3d printing. Have yet to
>> try it...
>>
>> Matt
> 
> Interesting, downloaded it.  What cad for linux can export an .stl file?
> 
> Cheers, Gene

Freecad can export STL files.

--
Write once. Port to many.
Get the SDK and tools to simplify cross-platform app development. Create 
new or port existing apps to sell to consumers worldwide. Explore the 
Intel AppUpSM program developer opportunity. appdeveloper.intel.com/join
http://p.sf.net/sfu/intel-appdev
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] 3d PDFs was Re: Engaver Tool Changing

[Fox Mulder]

Am 29.11.2011 01:12, schrieb andy pugh:
> On 28 November 2011 23:45, Fox Mulder  wrote:
> 
>> With current browsers suporting WebGL it is quite easy to embedd a X3D
>> model into a web site without any need for an extra plugin. I tried this
>> myself some time ago and it worked very good.
> 
> It sort-of works for me, I had to enable it in the "Developer" menu,
> and I did get a range of 3D views flashing up briefly, but not in any
> way that seemed to be controllable.
> 

The problem is that WebGL as part of the HTML5 standard is quite new and
the grade of implementation varies over different browsers. I think it
needs a bit more time to straighten up the standard and implementations
in the browsers. For my tests i only use latest firefox which renders
the 3D content quite well.

WebGL is only a interface to OpenGL for native rendering of 3D content
in the browser. My example with a X3D file is displayed over a
javascript from x3dom which converts the geometry to WebGL language.
There is also another javascript collection which uses COLLADA as source
format for displaying the geometry in the browser. And i think in the
future there will be more powerfull frameworks for WebGL which can read
different kinds of source 3D formats.

Ciao,
 Rainer

--
All the data continuously generated in your IT infrastructure 
contains a definitive record of customers, application performance, 
security threats, fraudulent activity, and more. Splunk takes this 
data and makes sense of it. IT sense. And common sense.
http://p.sf.net/sfu/splunk-novd2d
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] 3d PDFs was Re: Engaver Tool Changing

[Fox Mulder]

Am 29.11.2011 00:29, schrieb Kent A. Reed:
> the data sets into supporting documentation. This in addition to the 
> VRML/x3d models we were already creating and which the community didn't 
> like (we never overcame the problem of poor browser support of VRML 
> after Silicon Graphics went belly up).

With current browsers suporting WebGL it is quite easy to embedd a X3D
model into a web site without any need for an extra plugin. I tried this
myself some time ago and it worked very good. You can see the test file
at [1] which i created in blender and exported as X3D.
The documentation how to do it is from the official x3dom homepage at [2].

Ciao,
 Rainer

[1] https://quakeman.homelinux.net/files/webgl/lcd.html
[2] http://www.x3dom.org/?page_id=1101

--
All the data continuously generated in your IT infrastructure 
contains a definitive record of customers, application performance, 
security threats, fraudulent activity, and more. Splunk takes this 
data and makes sense of it. IT sense. And common sense.
http://p.sf.net/sfu/splunk-novd2d
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Generating messages regarding inputs changing their levels

[Fox Mulder]

Am 12.11.2011 07:44, schrieb Farzin Kamangar:
> Hello, dear EMC users;
>  I checked all the manuals but I did not find anything regarding
> generation of user error messages. I need to show a message when one of my
> axis drives have an alarm and
> cannot move the axis. I have connected one of the drive outputs to one of
> parallel port inputs. When the drive has faulted and its output shows its
> condition, the controller feels
> it through the parport input pin. Now I need to show a message on the
> notifications area that this drive has a fault. This way the user knows why
> the axis is not moving actually.
> Could you please let me know how I can do that. I think something should be
> done in the HAL files, so I could show my message with respect to the level
> of one of the input pins.
> Thanks a lot,
> Farzin

Maybe the if/else conditional execution [1] in combination with an
status message [2][3] could work.
But i never tried this way of conditional messages.

Ciao,
 Rainer

[1] http://www.linuxcnc.org/docview/html/gcode_main.html#r3_3
[2] http://www.linuxcnc.org/docview/html/gcode_main.html#sub:Messages
[3]
http://www.linuxcnc.org/docview/html/gcode_main.html#sub:Debugging-messages

--
RSA(R) Conference 2012
Save $700 by Nov 18
Register now
http://p.sf.net/sfu/rsa-sfdev2dev1
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] OT: 3D Surface Generation

[Fox Mulder]

Am 14.09.2011 20:26, schrieb Ben Jackson:
> On Wed, Sep 14, 2011 at 08:22:27PM +0200, Fox Mulder wrote:
>>
>> Blender has a good exporter for STL format which many CAM programs can
>> load. I use blender myself and can say that it is quite intuitive to
>> model things with it. :)
> 
> I'm not sure I've ever seen "blender" and "intuitive" in the same building,
> let alone the same sentence!
> 
> Regarding STL:  It is a collection of flat faces.  If you are modelling
> something like a violin you will want to be very careful to choose your
> export settings so that there is enough detail in the faces without
> making the file so huge it becomes a problem to work on in later steps.
> 

I think blender 2.5 is quite intuitive. The gui of the older 2.4x branch
of blender was way more difficult to grasp when you ask me. ;)
And maybe the svg export is better because it can export bezier curves
as they are.

Ciao,
 Rainer

--
BlackBerry® DevCon Americas, Oct. 18-20, San Francisco, CA
Learn about the latest advances in developing for the 
BlackBerry® mobile platform with sessions, labs & more.
See new tools and technologies. Register for BlackBerry® DevCon today!
http://p.sf.net/sfu/rim-devcon-copy1 
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] OT: 3D Surface Generation

[Fox Mulder]

Am 14.09.2011 19:10, schrieb Florian Rist:
> Hi Kirk
> 
>  > I would like to mill some 3D surfaces (violin plates) but I need a means
>  > to go from a paper plan to g-code. I have Synergy with 3D, but I would
>  > prefer to use a free and open solution.
> 
> I think the 3D model could be easily created in Blender 
> (http://www.blender.org/). I don't know about the g-code, though.

Blender has a good exporter for STL format which many CAM programs can
load. I use blender myself and can say that it is quite intuitive to
model things with it. :)

Ciao,
 Rainer

--
BlackBerry® DevCon Americas, Oct. 18-20, San Francisco, CA
Learn about the latest advances in developing for the 
BlackBerry® mobile platform with sessions, labs & more.
See new tools and technologies. Register for BlackBerry® DevCon today!
http://p.sf.net/sfu/rim-devcon-copy1 
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] HAL and INI file settings for Tool length Probing.

[Fox Mulder]

Am 08.09.2011 22:22, schrieb Don Stanley:
> Thanks Andy;
> The thing I am stumbling over is What to connect the touch switch pin to
> in the .hal file. After that I have no idea ask but I think you may have
> answered some of it.

The intenal signal is "motion.probe-input" which you should connect to
the probe switch.

a few months ago i wrote a little skript (tool-length-set.ngc) which
measures all tool lengths and writes the offset to the reference tool
into the tool table.
The reference tool is the first one in the script and has no tool number
(T0). The comments are mostly german but i think the script is not very
complex to understand. :)

Ciao,
 Rainer

(Automatisches ausmessen der verschiedenen Tool Längen und eintragen in die 
tool.tbl)

(PROBEOPEN probe.txt)

(Subroutine, welche das aktuelle Tool ausmisst)
O100 sub
O101 repeat [#<_count>]
(messe naechstes tool)
G0 Z30
T[#<_tool_nr>] M6 (insert used tool number)
G30 (fahre auf vordefinierte position, gesetzt mit g30.1)
G38.2 Z-20 F70 (messe solange bis schalter gedrueckt)
G38.4 Z20 F10 (messe solange bis schalter geoeffnet)
# = [#5063-#<_ref>]
(debug, Laengenunterschied Tool #<_tool_nr> zur Referenz = #)
G10 L1 P[#<_tool_nr>] Z[#] (tool offset speichern)
#<_tool_nr> = [#<_tool_nr> + 1] (tool nummer erhoehen)
O101 endrepeat
O100 endsub

(messe referenz tool nummer 1)
T0 M6 (kein Tool auswaehlen)
G30 (fahre auf vordefinierte position, gesetzt mit g30.1)
G49 (tool offset deaktivieren)
G91 (relative distanz verfahren)
G38.2 Z-20 F70 (messe solange bis schalter gedrueckt)
G38.4 Z20 F10 (messe solange bis schalter geoeffnet)
#<_ref>=#5063 (save reference tool length)

#<_count> = 5 (5 tools ausmessen)
#<_tool_nr> = 10 (erste toolnummer, welche sequenziell erhoeht wird)
O100 call

G0 Z5 (relative 5mm ueber dem geoeffneten schalter)
G90 (absolute koordinaten)

(PROBECLOSE)

(debug, Tool Offset ist immer noch aktiv!)
M2 (programm ende)
--
Doing More with Less: The Next Generation Virtual Desktop 
What are the key obstacles that have prevented many mid-market businesses
from deploying virtual desktops?   How do next-generation virtual desktops
provide companies an easier-to-deploy, easier-to-manage and more affordable
virtual desktop model.http://www.accelacomm.com/jaw/sfnl/114/51426474/___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] tool length probing does not work as expected

[Fox Mulder]

Am 24.07.2011 00:41, schrieb Sebastian Kuzminsky:
> On 07/23/2011 04:12 PM, Chris Radek wrote:
>> On Sat, Jul 23, 2011 at 09:51:39PM +0200, Fox Mulder wrote:
>>
>>> I hope someone can shed a bit light in this problem and give me a hint
>>> where the culprit hides. :)
>> If you are using 2.4.6 or earlier and your probe is on a parallel port
>> input, this is perhaps bug #3040806 which is fixed in the upcoming
>> 2.4.7 release (and upcoming 2.5 releases).
>>
>> http://sourceforge.net/tracker/?func=detail&aid=3040806&group_id=6744&atid=106744
>>
>> Meanwhile, you could try a v2.5_branch package from the buildbot.
> 
> If you want to preview what 2.4.7 will be like (including the 
> parport/probe bugfix), you can also use the v2.4_branch packages from 
> the buildbot.  The fix went into 2.4.6-44, so it's included in the most 
> recent version, which is 2.4.6-74.
> 
Thanks for the hint.
And yes i use the official 2.4.6 (not buildhost) version with the probe
input at the parallel port. I will try the latest buildhost version and
hopefully it works. :)

Ciao,
 Rainer

--
Storage Efficiency Calculator
This modeling tool is based on patent-pending intellectual property that
has been used successfully in hundreds of IBM storage optimization engage-
ments, worldwide.  Store less, Store more with what you own, Move data to 
the right place. Try It Now! http://www.accelacomm.com/jaw/sfnl/114/51427378/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

[Emc-users] tool length probing does not work as expected

[Fox Mulder]

Hi,

i try to make a little tool probing sequence which should work like the
homing sequence for one axis.
The z-axis should go down fast until the probe switch is touched and the
output is closed. After that it should go slowly up until the switch is
opened. For that purpose i took a look into the example file
tool-length-probe.ngc. This file works correctly except that it only
does the probing until the switch is closed (with G38.2).

So i thought i extend the sequence a bit with an additional G38.4
directly after the G38.2 to reach my goal.
Now my problem is that the G38.4 command does not work. I only get the
error message "probe is already clear when starting g38.4 or g38.5 move"
when i try to execute this command. I tried the whole sequence in mdi
mode by manually typing the commands.

working sequence:
g30 (move to probe switch reference position)
g91 (relative koordinates)
g38.2 z-20 f100 (fast probe until switch is closed)
g0 z5 (move 5mm above the switch)
#1000=#5063 (save reference tool length)
g90 (absolute koordinates)

not working sequence:
g30 (move to probe switch reference position)
g91 (relative koordinates)
g38.2 z-20 f100 (fast probe until switch is closed)
g38.4 z20 f10 (slow probe until switch is closed)
g0 z5 (move 5mm above the switch)
#1000=#5063 (save reference tool length)
g90 (absolute koordinates)

It doesn't matter if the switch still is closed or if i open it manually
before executing G38.4. The error still is the same. But instead i can
execute the G38.2 command as often as i want without any errors and it
always works as expected.

When searching for this error message i can only find the reference in
the command.c file within the emc sourcode. So i think i must do
something wrong without knowing it.
I hope someone can shed a bit light in this problem and give me a hint
where the culprit hides. :)

Ciao,
 Rainer

--
Storage Efficiency Calculator
This modeling tool is based on patent-pending intellectual property that
has been used successfully in hundreds of IBM storage optimization engage-
ments, worldwide.  Store less, Store more with what you own, Move data to 
the right place. Try It Now! http://www.accelacomm.com/jaw/sfnl/114/51427378/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] LPT "in" pinout

[Fox Mulder]

Am 28.06.2011 21:05, schrieb Viesturs Lācis:
> Hello, folks!
> 
> I am short on input pins for EMC, so I would like to set the second
> LPT port as "in", which gives more input pins.
> The 10K $ question is:
> where can I find the pinout for "in" mode?
> 
> The first LPT port by default is "out", which has 5 input pins - 10,
> 11, 12, 13 and 15. The remaining from 1 to 17 are outputs, 18-25 are
> GND.
> What about "in" mode? I had no luck finding the answer on web :(
> 
> Thanks in advance!

Hi,

the same five pins are still inputs on the second parallel port.
Additionally the data pins D0-D7 at pins 2-9 are switched to input mode.

Ciao,
Rainer

--
All of the data generated in your IT infrastructure is seriously valuable.
Why? It contains a definitive record of application performance, security 
threats, fraudulent activity, and more. Splunk takes this data and makes 
sense of it. IT sense. And common sense.
http://p.sf.net/sfu/splunk-d2d-c2
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] [OT] CAM Simulation?

[Fox Mulder]

Hi,

a very good simulation software for generated g-code and how it would be
executed are "the metacut utilities". It can simulate the stock and the
used tools to show how it will be processed.

Ciao,
 Rainer

URL: http://www.nwdesigns.com/Products/MCUNutshell.htm

Am 03.05.2011 15:55, schrieb Florian Rist:
> Hello,
> not directly EMC related, but maybe you have some advice for me:
> 
> I'm looking for a way to simulate a 5axes milling process. I'd like to 
> create an animation so show how the material is removed form the stock. 
> I don't necessarily need a ready to use program, a framework the allows 
> this kind of voxel based volumetric simulations would be fine, too. Any 
> ideas?
> 
> See you
> Flo
> 
> --
> WhatsUp Gold - Download Free Network Management Software
> The most intuitive, comprehensive, and cost-effective network 
> management toolset available today.  Delivers lowest initial 
> acquisition cost and overall TCO of any competing solution.
> http://p.sf.net/sfu/whatsupgold-sd
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users
> 

--
WhatsUp Gold - Download Free Network Management Software
The most intuitive, comprehensive, and cost-effective network 
management toolset available today.  Delivers lowest initial 
acquisition cost and overall TCO of any competing solution.
http://p.sf.net/sfu/whatsupgold-sd
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Ethernet data collector?

[Fox Mulder]

First question is what speed do you need from measurement to delivery in
the pc?
A range within Mikroseconds would be much more complex than within
milliseconds. If a few milliseconds is fast enough than an ad converter
with multiplexer and a little microcontroller could do the work and
deliver it over usb via virtual rs232 to the pc.
This is what i would do but maybe you search for a finished product
which fullfils your needs. :)

Ciao,
 Rainer

Am 11.03.2011 00:19, schrieb Igor Chudov:
> I want to collect a few analog data pieces, like 5-6 current values (AC and
> DC), some temperatures, voltages etc. is there some thing that lets me do it
> and disseminate it somehow to my Linux PC, via ethernet or USB or whatever.
> --
> Colocation vs. Managed Hosting
> A question and answer guide to determining the best fit
> for your organization - today and in the future.
> http://p.sf.net/sfu/internap-sfd2d
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users
> 

--
Colocation vs. Managed Hosting
A question and answer guide to determining the best fit
for your organization - today and in the future.
http://p.sf.net/sfu/internap-sfd2d
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] [VIDEO] PLASMA Cutting with EMC2 and USB based THC/motion interface

[Fox Mulder]

Am 03.02.2011 10:17, schrieb Steve Blackmore:
> Mmmhh -  I find the fact that it's USB based more interesting.
> 
> http://www.intra2net.com/en/developer/libftdi/index.php

And i find it interesting that they doesn't list the FT232RL as
supported device. They list the old FT232BM which needs much more
external components compared to it's predecessor FT232RL. The RL also
has more features and i/o pins to use and is little cheaper than the old
BM version. :)

Ciao,
 Rainer

--
Special Offer-- Download ArcSight Logger for FREE (a $49 USD value)!
Finally, a world-class log management solution at an even better price-free!
Download using promo code Free_Logger_4_Dev2Dev. Offer expires 
February 28th, so secure your free ArcSight Logger TODAY! 
http://p.sf.net/sfu/arcsight-sfd2d
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Talk to me about comp_file

[Fox Mulder]

Am 02.02.2011 22:13, schrieb Lester Caine:
> Steve Blackmore wrote:
>>> 110V ones are available as well ...
>> Thanks Lester:)
>>
>> They don't link a US distributor and all the countries shown are 240V
>> ones.
> I'll regret this ;)
> http://www.damencnc.com/damencnc.php?dir=cpt&time=1296680942&langId=EN&idComp=42
> Kevin will ship to the US
> 
I have bought the Kress FME 800 last year for my hobby cnc and i really
like it. :)

--
Special Offer-- Download ArcSight Logger for FREE (a $49 USD value)!
Finally, a world-class log management solution at an even better price-free!
Download using promo code Free_Logger_4_Dev2Dev. Offer expires 
February 28th, so secure your free ArcSight Logger TODAY! 
http://p.sf.net/sfu/arcsight-sfd2d
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Analog inputs, how to capture them

[Fox Mulder]

A germany distributor sells a very cheap AVR demo kit with network
interface called AVR-NET-IO [1] for only 19,95€. An already assembled
version [2] is for 27,95€ available. It uses the ATMEGA32 which has 8
analog inputs with 10bit resolution. The default firmware uses four of
them to display over the network.

For this board exist many customized firmwares in the net which uses all
8 inputs. Some of them implements a simple webserver to show the i/o
status or you could customize the firmware for your own needs.

This board is just an example of a very cheap network solution to show
analog and digital inputs over network.

Ciao,
 Rainer

[1]
http://www.pollin.de/shop/dt/MTQ5OTgxOTk-/Bausaetze_Module/Bausaetze/Bausatz_AVR_NET_IO.html
[2]
http://www.pollin.de/shop/dt/NjI5OTgxOTk-/Bausaetze_Module/Bausaetze/AVR_NET_IO_Fertigmodul.html

Am 31.12.2010 23:16, schrieb Igor Chudov:
> I would like to measure a few analog things:
> 
> 1) DC bus voltage
> 2) DC Current
> 3) Spindle motor current
> 4) etc
> 
> Is there some way to capture this and have it displayed, by EMC or
> otherwise.
> 
> I was hoping to find some analog capture module that would proadcast the
> data on ethernet.
> 
> i
> --
> Learn how Oracle Real Application Clusters (RAC) One Node allows customers
> to consolidate database storage, standardize their database environment, and, 
> should the need arise, upgrade to a full multi-node Oracle RAC database 
> without downtime or disruption
> http://p.sf.net/sfu/oracle-sfdevnl
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users
> 

--
Learn how Oracle Real Application Clusters (RAC) One Node allows customers
to consolidate database storage, standardize their database environment, and, 
should the need arise, upgrade to a full multi-node Oracle RAC database 
without downtime or disruption
http://p.sf.net/sfu/oracle-sfdevnl
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] OT: PDA's and machine control

[Fox Mulder]

Am 26.09.2010 19:04, schrieb Kirk Wallace:
> I want to read encoders with a PDA, such as for a DRO, but I would
> prefer to use something like a PC parallel port. I suspect no PDA has a
> parallel port, so does anyone know of a good PDA that has GPIO or other
> hackable ports for digital I/O? It seems the Sharp Zaurus has good Linux
> support, has anyone hacked one of these for machine control? Thanks.
> 
A PDA with full native Linux running is the openmoko freerunner. It can
run debian for example and has a native usb host interface for external
usb devices. When the case is opened there are also a few gpio pins and
an I2C/SPI Interface available. I never hacked my own one to add more
hardware but there are a few projects which did this with help of a few
gpio pins for this device. The hardware and software are completely
opensource and the schematics are available in the internet at [1] for
downloading. Maybe you can find out in the schematics if there are
enough interfaces for your purpose.

Ciao,
 Rainer

[1] http://downloads.openmoko.org/developer/schematics/GTA02/

--
Start uncovering the many advantages of virtual appliances
and start using them to simplify application deployment and
accelerate your shift to cloud computing.
http://p.sf.net/sfu/novell-sfdev2dev
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

[Emc-users] My second try :)

[Fox Mulder]

Here is my second try to mill a test layout with my selfmade cnc.

http://www.youtube.com/watch?v=UUuqJ32-q7s

After the milling finished and i removed the dust i saw that i made a
little mistake. I milled a bit too deep and therefore the tracks got
nearly milled away. :(
I calculated the milling depth and width correct but it seems that the
real width is a bit wider than in theory. So i think i have to lower the
depth a bit to get my wanted width.

At the end of the video i inserted a picture of the finished pcb.

Next thing for me is to do a few try&error mills with different depths
to find the correct one for my setup. :)

Ciao,
 Rainer

--
Start uncovering the many advantages of virtual appliances
and start using them to simplify application deployment and
accelerate your shift to cloud computing.
http://p.sf.net/sfu/novell-sfdev2dev
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] My first try to mill a simple layout

[Fox Mulder]

I made the hold down clamps especially for milling layouts. I wanted
clamps that hold the board but doesn't overlap the board so that i could
mill to the edges of the board without interfering with the clamps.
Therefore i created them on my lathe with a small sharp flange at the
top which is pressed from the side into the board. The clamps on the
down and left side are at a fixed position. For the upper and right
clamps i drilled an excentric mounting hole so i could tighten the
clamps against the edges of the board. Because of the mellow material of
these boards this works quite good. Only the material i have used is not
the best because it begins to corrode. Maybe i made the next clamps of
stainless steel. :)

I made two new photos of the clamps at [1] and [2] (high resolution with
"download 3072" button).

Ciao,
 Rainer

[1] https://quakeman.homelinux.net/bilder/details.php?image_id=7534
[2] https://quakeman.homelinux.net/bilder/details.php?image_id=7535


Am 24.07.2010 07:30, schrieb mach3:
> Nice setup.  Looks like it works nice and smooth.  
> 
> Maybe it's just my inexperience, but I haven't seen hold down clamps like
> that before.  Can you describe/show how they work?  They look pretty cool.  
> 
> Michael
> 
> -Original Message-
> From: Fox Mulder [mailto:quakem...@gmx.net] 
> Sent: Thursday, July 22, 2010 6:41 AM
> To: emc-users@lists.sourceforge.net
> Subject: [Emc-users] My first try to mill a simple layout
> 
> Hi,
> 
> i just want to show my first mill experience with my selfmade CNC and
> EMC. :)
> 
> I tried a simple layout example from eagle which i converted to g-code
> with pcb-gcode from [1]. The layout only consists of a quadratic wire
> and two letters in the middle. I filmed the milling process and uploaded
> it to youtube at [2].
> 
> I did take pictures through the various build processes of my CNC and
> uploaded them on my own picture gallery at [3]. I use a self signed
> certificate, so you have to persuade your browser to surf to my site.
> 
> For the first try i think the result is quite good but i have to do more
> testing. And i also want to build a special floating head for milling
> layouts because i think that it would produce better results for larger
> areas. But this comes when i have played a bit more. :)
> 
> Ciao,
>  Rainer
> 
> 
> [1] http://www.pcbgcode.org/
> [2] http://www.youtube.com/watch?v=7xjv8QPoLzw
> [3] https://quakeman.homelinux.net/bilder/categories.php?cat_id=169
> 
> 
> --
> This SF.net email is sponsored by Sprint
> What will you do first with EVO, the first 4G phone?
> Visit sprint.com/first -- http://p.sf.net/sfu/sprint-com-first
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users
> 
> 
> --
> This SF.net email is sponsored by Sprint
> What will you do first with EVO, the first 4G phone?
> Visit sprint.com/first -- http://p.sf.net/sfu/sprint-com-first
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users
> 

--
This SF.net email is sponsored by Sprint
What will you do first with EVO, the first 4G phone?
Visit sprint.com/first -- http://p.sf.net/sfu/sprint-com-first
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] My first try to mill a simple layout

[Fox Mulder]

Am 22.07.2010 18:26, schrieb Steve Robertson:
> 
>> Hi,
>>
>> i just want to show my first mill experience with my selfmade CNC and
>> EMC. :)
> 
> Very nice. Can you provide some details as to what kind of controllers,
> drives, screws, etc... You are using?

Sure thing.

For the electronic part i use a simple self build 3-axis L297/L298
stepper driver which is interfaced to the parallel port of the pc.

In the mechanical part i'm using stepper motors which i bought used in
the internet (no cheap printer motors). They drive the ball screw
threads over a 1:2.5 transmission ratio. The ball screw threads have a
flank lead of 4mm/turn. Only the z-axis has a transmission ratio of 1:5
because of the weight it has to lift.
I use linear ball bearing for all three axis. The whole frame is build
of aluminium and zinc plated screws. As spindle motor i use a Kress 800
FME with 3.125mm chuck.
In my picture gallery you can see most of the parts i used to build the
CNC and how i build them.

For controling i use EMC 2.4.2 which works really good so far. I only
have some strange display errors in the 3D view. But i think that the
driver for the onboard graphics card are very bad.

I'm still not finished with my CNC but it's in a state where i can first
use it for real tests.

Ciao,
 Rainer

PS: I hope i found the correct words for the technical parts. :)

--
This SF.net email is sponsored by Sprint
What will you do first with EVO, the first 4G phone?
Visit sprint.com/first -- http://p.sf.net/sfu/sprint-com-first
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

[Emc-users] My first try to mill a simple layout

[Fox Mulder]

Hi,

i just want to show my first mill experience with my selfmade CNC and
EMC. :)

I tried a simple layout example from eagle which i converted to g-code
with pcb-gcode from [1]. The layout only consists of a quadratic wire
and two letters in the middle. I filmed the milling process and uploaded
it to youtube at [2].

I did take pictures through the various build processes of my CNC and
uploaded them on my own picture gallery at [3]. I use a self signed
certificate, so you have to persuade your browser to surf to my site.

For the first try i think the result is quite good but i have to do more
testing. And i also want to build a special floating head for milling
layouts because i think that it would produce better results for larger
areas. But this comes when i have played a bit more. :)

Ciao,
 Rainer


[1] http://www.pcbgcode.org/
[2] http://www.youtube.com/watch?v=7xjv8QPoLzw
[3] https://quakeman.homelinux.net/bilder/categories.php?cat_id=169

--
This SF.net email is sponsored by Sprint
What will you do first with EVO, the first 4G phone?
Visit sprint.com/first -- http://p.sf.net/sfu/sprint-com-first
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Problem with tool path

[Fox Mulder]

Am 22.06.2010 15:33, schrieb Andy Pugh:
> 2010/6/22 Fox Mulder :
> 
>>  Maybe this ia a feature and not a bug but then i
>> need to know how to tweak or disable it.
> 
> It is, indeed, a feature.
> It might be that you need much higher accelleration limits in your INI file.
> EMC2 tries to maintain speed at the expense of path accuracy unless
> you tell it not to.
> http://www.linuxcnc.org/docview/html//gcode_main.html#sub:G61,-G61.1,-G64:
> 
> However, it looks like you are seeing a very extreme version of that,
> and I suspect that you could usefully increase your  accell terms  in
> the ini file by a few orders of magnitude.
> 
Thanks for the hint. I will give it a try at the weekend. :)

--
ThinkGeek and WIRED's GeekDad team up for the Ultimate 
GeekDad Father's Day Giveaway. ONE MASSIVE PRIZE to the 
lucky parental unit.  See the prize list and enter to win: 
http://p.sf.net/sfu/thinkgeek-promo
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

[Emc-users] Problem with tool path

[Fox Mulder]

Hi,

since a few days i finished the most parts of my selfmade cnc and i
installed emc 2.4.1 on a pc to control it. The hardware setup is fine
now and when i move the three axis it works like expected. Now i made a
simple test with an eagle layout file which i exported to gcode
(isolation milling). It is just a rectangle with two letters within it.
The layout looks like in the attached picture layout.png. The generated
gcode is in the file cnc_test.bot.etch.tap and in gcode_ok.png is a
picture of the generated toolpath for the gcode. This looks exactly the
same as in the 3d window of emc.

But now to the real problem. The inner rectangle has 90° edges so it is
just a four lines path. It is defined with these lines in the file:

G00 X-13.3652 Y13.3652
G01 Z-0.3730 F100.00
G01 X-13.3652 Y1.8747 F360.00
G01 X-1.8747 Y1.8747
G01 X-1.8747 Y13.3652
G01 X-13.3652 Y13.3652

First it goes to the upper left corner, moves the z-axis down and than
moves to the next corner and on until it is back to the starting point.
When i run it in single step mode it works as long as i let every
command finish and then do the next step. When i do more than one step
in a short time the path is changed to round edges like shown in green
in gcode_wrong.png. When i run the program without stepping through it
also has round edges. Maybe this ia a feature and not a bug but then i
need to know how to tweak or disable it.

I hope someone can help me with this. :)

Ciao,
 Rainer
(.../pcb-gcode-3.5.2.11/pcb-gcode.ulp)
(Copyright 2005 - 2009 by John Johnson)
(See readme.txt for licensing terms.)
(This file generated from the board:)
(.../cnc_test/cnc_test.brd)
(Current profile is .../pcb-gcode-3.5.2.11/profiles/emc.pp  )
(This file generated 22.06.2010 12:48:53)
G21
(Absolute Coordinates)
G90
G00 X0. Y0. 
M03
G04 P2.00
G00 Z2.5000 
G00 X-13.3652 Y13.3652 
G01 Z-0.3730 F100.00 
G01 X-13.3652 Y1.8747 F360.00 
G01 X-1.8747 Y1.8747 
G01 X-1.8747 Y13.3652 
G01 X-13.3652 Y13.3652 
G00 Z2.5000 
G00 X-9.7269 Y6.9169 
G01 Z-0.3730 F100.00 
G01 X-9.7915 Y7.0287 F360.00 
G01 X-9.8249 Y7.1534 
G01 X-9.8249 Y7.2827 
G01 X-9.7915 Y7.4074 
G01 X-9.7269 Y7.5192 
G01 X-9.6356 Y7.6105 
G01 X-9.5238 Y7.6751 
G01 X-9.3991 Y7.7085 
G01 X-9.2698 Y7.7085 
G01 X-9.1451 Y7.6751 
G01 X-9.0333 Y7.6105 
G01 X-8.7943 Y7.3714 
G01 X-8.7445 Y7.4213 
G01 X-8.7445 Y8.5459 
G01 X-8.8036 Y8.5617 
G01 X-8.9154 Y8.6263 
G01 X-9.0067 Y8.7176 
G01 X-9.0713 Y8.8294 
G01 X-9.1047 Y8.9541 
G01 X-9.1047 Y9.0834 
G01 X-9.0713 Y9.2081 
G01 X-9.0067 Y9.3199 
G01 X-8.9154 Y9.4112 
G01 X-8.8036 Y9.4758 
G01 X-8.6789 Y9.5092 
G01 X-7.8293 Y9.5092 
G01 X-7.7046 Y9.4758 
G01 X-7.5928 Y9.4112 
G01 X-7.5015 Y9.3199 
G01 X-7.4449 Y9.2217 
G01 X-7.3882 Y9.3199 
G01 X-7.2969 Y9.4112 
G01 X-7.1851 Y9.4758 
G01 X-7.0604 Y9.5092 
G01 X-5.6199 Y9.5092 
G01 X-5.4906 Y9.5092 
G01 X-5.3659 Y9.4758 
G01 X-5.2541 Y9.4112 
G01 X-5.1628 Y9.3199 
G01 X-5.0982 Y9.2081 
G01 X-5.0648 Y9.0834 
G01 X-5.0648 Y8.7233 
G01 X-5.0648 Y8.5941 
G01 X-5.0968 Y8.4747 
G01 X-5.0982 Y8.4693 
G01 X-5.1628 Y8.3575 
G01 X-6.5053 Y7.0150 
G01 X-6.5053 Y6.7933 
G01 X-6.5387 Y6.6686 
G01 X-6.6033 Y6.5568 
G01 X-6.6946 Y6.4655 
G01 X-6.8064 Y6.4009 
G01 X-6.9311 Y6.3675 
G01 X-7.0604 Y6.3675 
G01 X-7.1851 Y6.4009 
G01 X-7.2969 Y6.4655 
G01 X-7.3882 Y6.5568 
G01 X-7.4528 Y6.6686 
G01 X-7.4862 Y6.7933 
G01 X-7.4862 Y7.2827 
G01 X-7.4528 Y7.4074 
G01 X-7.3882 Y7.5192 
G01 X-7.3643 Y7.5430 
G01 X-6.3790 Y8.5283 
G01 X-7.0604 Y8.5283 
G01 X-7.1851 Y8.5617 
G01 X-7.2969 Y8.6263 
G01 X-7.3882 Y8.7176 
G01 X-7.4449 Y8.8158 
G01 X-7.5015 Y8.7176 
G01 X-7.5928 Y8.6263 
G01 X-7.7046 Y8.5617 
G01 X-7.7636 Y8.5459 
G01 X-7.7636 Y7.1534 
G01 X-7.7651 Y7.1481 
G01 X-7.7970 Y7.0286 
G01 X-7.8616 Y6.9168 
G01 X-8.3131 Y6.4654 
G01 X-8.3850 Y6.4240 
G01 X-8.4249 Y6.4009 
G01 X-8.5496 Y6.3675 
G01 X-8.6789 Y6.3675 
G01 X-9.0390 Y6.3675 
G01 X-9.1637 Y6.4009 
G01 X-9.2755 Y6.4655 
G01 X-9.3668 Y6.5568 
G01 X-9.7269 Y6.9169 
G00 Z2.5000 
G00 X-13.9103 Y0.6652 
G01 Z-0.3730 F100.00 
G01 X-14.0296 Y0.6652 F360.00 
G01 X-14.1464 Y0.6884 
G01 X-14.2564 Y0.7340 
G01 X-14.3555 Y0.8002 
G01 X-14.4397 Y0.8844 
G01 X-14.5059 Y0.9835 
G01 X-14.5515 Y1.0935 
G01 X-14.5747 Y1.2103 
G01 X-14.5747 Y13.9103 
G01 X-14.5747 Y14.0296 
G01 X-14.5515 Y14.1464 
G01 X-14.5059 Y14.2564 
G01 X-14.4397 Y14.3555 
G01 X-14.3555 Y14.4397 
G01 X-14.2564 Y14.5059 
G01 X-14.1464 Y14.5515 
G01 X-14.0296 Y14.5747 
G01 X-1.3296 Y14.5747 
G01 X-1.2103 Y14.5747 
G01 X-1.0935 Y14.5515 
G01 X-0.9835 Y14.5059 
G01 X-0.8844 Y14.4397 
G01 X-0.8002 Y14.3555 
G01 X-0.7340 Y14.2564 
G01 X-0.6884 Y14.1464 
G01 X-0.6652 Y14.0296 
G01 X-0.6652 Y1.2103 
G01 X-0.6884 Y1.0935 
G01 X-0.7340 Y0.9835 
G01 X-0.8002 Y0.8844 
G01 X-0.8844 Y0.8002 
G01 X-0.9835 Y0.7340 
G01 X-1.0935 Y0.6884 
G01 X-1.2103 Y0.6652 
G01 X-1.3296 Y0.6652 
G01 X-13.9103 Y0.6652 
G00 Z2.5000 
G00 Z5. 
M05
M02
<><><>--
ThinkGeek and WIR