Re: [Emc-users] multiple gcode streams in parallel

[Dave Caroline]

Actually this problem touches the independent axes problem that I
foresee in my sliding head lathe where an axis that will be cutting
later needs to start part way though another's coordinated move in
order to save time.

Dave Caroline

On 24/09/2015, andy pugh  wrote:
> On 24 September 2015 at 02:18, Jerry Scharf  wrote:
>> I now need to extend it to run two separate robots off the same linuxcnc
>> controller and receiver program
>
> If you need the robots to synch then I think you need to be a bit cleverer.
>
> If you imagine one robot moving in XYZ space and another moving in UVW
> space I think you need a pre-processor that "interleaves" the G-code.
> This will include breaking up an XYZ move to start a UVW move and
> vice-versa.
>
> All the axes of a G-code line reach the end point at the same time. So
> to coordinate two robots you need to combine the moves onto one line,
> otherwise you can't have both moving.
>
> --
> atp
> If you can't fix it, you don't own it.
> http://www.ifixit.com/Manifesto
>
> --
> Monitor Your Dynamic Infrastructure at Any Scale With Datadog!
> Get real-time metrics from all of your servers, apps and tools
> in one place.
> SourceForge users - Click here to start your Free Trial of Datadog now!
> http://pubads.g.doubleclick.net/gampad/clk?id=241902991=/4140
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users
>

--
Monitor Your Dynamic Infrastructure at Any Scale With Datadog!
Get real-time metrics from all of your servers, apps and tools
in one place.
SourceForge users - Click here to start your Free Trial of Datadog now!
http://pubads.g.doubleclick.net/gampad/clk?id=241902991=/4140
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] multiple gcode streams in parallel

[andy pugh]

On 24 September 2015 at 02:18, Jerry Scharf  wrote:
> I now need to extend it to run two separate robots off the same linuxcnc
> controller and receiver program

If you need the robots to synch then I think you need to be a bit cleverer.

If you imagine one robot moving in XYZ space and another moving in UVW
space I think you need a pre-processor that "interleaves" the G-code.
This will include breaking up an XYZ move to start a UVW move and
vice-versa.

All the axes of a G-code line reach the end point at the same time. So
to coordinate two robots you need to combine the moves onto one line,
otherwise you can't have both moving.

-- 
atp
If you can't fix it, you don't own it.
http://www.ifixit.com/Manifesto

--
Monitor Your Dynamic Infrastructure at Any Scale With Datadog!
Get real-time metrics from all of your servers, apps and tools
in one place.
SourceForge users - Click here to start your Free Trial of Datadog now!
http://pubads.g.doubleclick.net/gampad/clk?id=241902991=/4140
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] multiple gcode streams in parallel

[Stuart Stevenson]

Sounds to me like one 5 axis machine. You should be able to issue one line
of code. All 5 axes should move coordinated in relation to each axis home
position no matter the distance or orientation of the components.

On Thu, Sep 24, 2015 at 5:21 AM, andy pugh  wrote:

> On 24 September 2015 at 02:18, Jerry Scharf  wrote:
> > I now need to extend it to run two separate robots off the same linuxcnc
> > controller and receiver program
>
> If you need the robots to synch then I think you need to be a bit cleverer.
>
> If you imagine one robot moving in XYZ space and another moving in UVW
> space I think you need a pre-processor that "interleaves" the G-code.
> This will include breaking up an XYZ move to start a UVW move and
> vice-versa.
>
> All the axes of a G-code line reach the end point at the same time. So
> to coordinate two robots you need to combine the moves onto one line,
> otherwise you can't have both moving.
>
> --
> atp
> If you can't fix it, you don't own it.
> http://www.ifixit.com/Manifesto
>
>
> --
> Monitor Your Dynamic Infrastructure at Any Scale With Datadog!
> Get real-time metrics from all of your servers, apps and tools
> in one place.
> SourceForge users - Click here to start your Free Trial of Datadog now!
> http://pubads.g.doubleclick.net/gampad/clk?id=241902991=/4140
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users
>



-- 
Addressee is the intended audience.
If you are not the addressee then my consent is not given for you to read
this email furthermore it is my wish you would close this without saving or
reading, and cease and desist from saving or opening my private
correspondence.
Thank you for honoring my wish.
--
Monitor Your Dynamic Infrastructure at Any Scale With Datadog!
Get real-time metrics from all of your servers, apps and tools
in one place.
SourceForge users - Click here to start your Free Trial of Datadog now!
http://pubads.g.doubleclick.net/gampad/clk?id=241902991=/4140
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] multiple gcode streams in parallel

[andy pugh]

On 24 September 2015 at 12:44, Dave Caroline
 wrote:
> Actually this problem touches the independent axes problem that I
> foresee in my sliding head lathe where an axis that will be cutting
> later needs to start part way though another's coordinated move in
> order to save time.

I think this needs you to split moves. Maybe it is something that a
preprocessor could do.

If you are doing an X-move from 0 to 50 and want to start U when X is
at 25 then

G1 X50

becomes

G1 X25
G1 X50 U50

I think that that means that the X has to stop, though, or the moves
are not coordinated. (even though you don't necessarily care)

-- 
atp
If you can't fix it, you don't own it.
http://www.ifixit.com/Manifesto

--
Monitor Your Dynamic Infrastructure at Any Scale With Datadog!
Get real-time metrics from all of your servers, apps and tools
in one place.
SourceForge users - Click here to start your Free Trial of Datadog now!
http://pubads.g.doubleclick.net/gampad/clk?id=241902991=/4140
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] multiple gcode streams in parallel

[Dave Caroline]

might be better to let the two talk to each other in some way so move
starts/runs are really synchronised properly rather than rely on
buffers.
eg classic ladder or some clocking or whatever

Dave Caroline

On 24/09/2015, Jerry Scharf  wrote:
> Hi,
>
> I have hacked up a simple program to allow a client to connect up to the
> linuxcnc system and issue g code commands. It's fairly stupid, just enough
> to get what we needed done.
>
> I now need to extend it to run two separate robots off the same linuxcnc
> controller and receiver program. This brought up a question about the
> python interface.
>
> Can I issue two separate motion commands for different axes at the same
> time? One bot is a single axis on b and the other uses a,x,y,z. The motion
> for each is serialized by the nature of the network buffers, but I am
> trying to understand whether I need to serialize the commands between the
> different streams. It seems right now like if I move the machine manually
> from the GUI, the program then thinks the controller is not ready to accept
> mdi commands.
>
> I have an allocation command that is being extended to have you tell it
> which axes this client wants to move. This prevents two programs from
> fighting over a given stepper...
>
> It's really minimal code but I will share it if people are interested.
>
> jerry
>
> --
> Jerry Scharf
> FINsix IT
> 650.285.6361 w
> 650.279.7017 m
> --
> Monitor Your Dynamic Infrastructure at Any Scale With Datadog!
> Get real-time metrics from all of your servers, apps and tools
> in one place.
> SourceForge users - Click here to start your Free Trial of Datadog now!
> http://pubads.g.doubleclick.net/gampad/clk?id=241902991=/4140
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users
>

--
Monitor Your Dynamic Infrastructure at Any Scale With Datadog!
Get real-time metrics from all of your servers, apps and tools
in one place.
SourceForge users - Click here to start your Free Trial of Datadog now!
http://pubads.g.doubleclick.net/gampad/clk?id=241902991=/4140
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] multiple gcode streams in parallel

[Gregg Eshelman]

On 9/24/2015 12:57 AM, Dave Caroline wrote:
> might be better to let the two talk to each other in some way so move
> starts/runs are really synchronised properly rather than rely on
> buffers.
> eg classic ladder or some clocking or whatever
>
> Dave Caroline

That sounds good. What you Do Not Want is a race condition where you 
have two or more independent parallel processes with nothing to ensure 
they always complete in the correct order.

For an example of what can happen, google Therac 25


---
This email has been checked for viruses by Avast antivirus software.
https://www.avast.com/antivirus


--
Monitor Your Dynamic Infrastructure at Any Scale With Datadog!
Get real-time metrics from all of your servers, apps and tools
in one place.
SourceForge users - Click here to start your Free Trial of Datadog now!
http://pubads.g.doubleclick.net/gampad/clk?id=241902991=/4140
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

[Emc-users] Keeping track of machine (g53) position

[Gene Heskett]

Greetings all;

The Tormach ER20 adaptor will be more than enough extension to clear 
everything, so that problem is solved, and I am about to go carve up a 
kilobucks worth of Mahogany.  But I was unable to come up with a fly 
cutter to level the addition to the front of the jig, so that needs to 
be addressed before I actually mount a piece of Mahogany into it.

I punched a small hole in a contact piece of PCB on my jig yesterday, and 
I want to see if I can come up with a "g53 g1 f20 xn yn zn" command that 
will put the machine in a position to find, by G38.2 means the contact 
references on the jig, from which all other moves are then referenced.

I intend to leave the G54 mapping untouched from that determined when the 
machine is homed.  My code is not now doing that, so a rerun finds the 
default G54 map AFU on the 2nd run after the tool change.  Hence the 
damage to the contact, but its slightly over an inch square so I'll just 
move it 50 thou to an undamaged spot.  BTDT on the toy mill too.

And to make all the other moves from settings done in the 
G10 L2 P2 #x #y #z using G55 mapping.

This so that I can reload the routine after setting a trigger variable to 
tell it I have changed tools and that the exact same code can be 
re-used, but with a tool_diameter of perhaps .005" and a reduced to 
single pass depth cut nominally 0.125" below the top of the board.

The tool change is from a .250" end mill, to a sharp pointed 1/8" radius 
roundover shape.  So the tool_diameter becomes the diameter of the tip, 
quite sharp.  The depth of cut is then such that the outside wing of the 
roundover is right at the surface of the board.

Am I on the right track if its homed when I run it the first time,
to do this initial move to the contact points on the jig using the 
G53 G1 F20 x# y#
G53 G1 z#
to drop the tool near the contacts
Then, still adjusting only G55 with a G10 L2 P2 #x|#y|#z as the G38.2's 
run, using the results in the #53xx array to reset G10 L2 p2 (G55) and 
then doing the rest of the routine in G55 mapping.

Will this be repeatable over multi-runs without all that re-homing by 
switching back to G54 for those moves on pass/tool 2?

I think one of the mistakes in my code is the modifying the G10 L2 P1 
(G54) mapping, which for repeat runs is doing an AFU to where the 
machine goes as the original home position has been destroyed by 
diddling with the G54 mapping.  I'll march thru it come daylight and fix 
that.

Am I making sense?

I had thought a G92.1 wouild cause a restore, but that doesn't appear to 
be a solution. I doubt if I fully understand what it actually does.

=new subject=

Also, there is not a tool image for the backplot that would represent 
this tool, that I know of, unless the tool table has grown the ability 
to specify a suitable image as opposed to the tool diameter cylinder it 
uses now.  Wishfull thinking?  Future feature perhaps?

Thanks all.

Cheers, Gene Heskett
-- 
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
Genes Web page 

--
Monitor Your Dynamic Infrastructure at Any Scale With Datadog!
Get real-time metrics from all of your servers, apps and tools
in one place.
SourceForge users - Click here to start your Free Trial of Datadog now!
http://pubads.g.doubleclick.net/gampad/clk?id=241902991=/4140
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] multiple gcode streams in parallel

[TJoseph Powderly]

It doesnt sound like the task is well defined here on mail list.

If the 2 independent(?) systems do not require interpolation _between_ 
them, then M65 thru M66 could provide the 'handshaking'.
Each would have a 'busy' and a 'fin' signal.
This is extremely sequential.

Else please explain more,
and what did you see happen when
~"it didnt seem ready to accept commands" ?

TomP tjtr33

On 09/24/2015 03:00 AM, Gregg Eshelman wrote:
> On 9/24/2015 12:57 AM, Dave Caroline wrote:
>> might be better to let the two talk to each other in some way so move
>> starts/runs are really synchronised properly rather than rely on
>> buffers.
>> eg classic ladder or some clocking or whatever
>>
>> Dave Caroline
>
> That sounds good. What you Do Not Want is a race condition where you
> have two or more independent parallel processes with nothing to ensure
> they always complete in the correct order.
>
> For an example of what can happen, google Therac 25
>


--
Monitor Your Dynamic Infrastructure at Any Scale With Datadog!
Get real-time metrics from all of your servers, apps and tools
in one place.
SourceForge users - Click here to start your Free Trial of Datadog now!
http://pubads.g.doubleclick.net/gampad/clk?id=241902991=/4140
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Keeping track of machine (g53) position

[andy pugh]

On 24 September 2015 at 10:09, Gene Heskett  wrote:
> Am I making sense?

Well, you have certainly confused me.

But if you are talking about using probe results to set G54, then that
is a normal thing to do.
What I normally do is run probe_hole (in the samples) as that parks
the machine in the middle of the hole at the end, then I just "set
origin" in Touchy to zero (which is a G10 L...) command.

I keep meaning to install the pretty probe screen.

-- 
atp
If you can't fix it, you don't own it.
http://www.ifixit.com/Manifesto

--
Monitor Your Dynamic Infrastructure at Any Scale With Datadog!
Get real-time metrics from all of your servers, apps and tools
in one place.
SourceForge users - Click here to start your Free Trial of Datadog now!
http://pubads.g.doubleclick.net/gampad/clk?id=241902991=/4140
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] multiple gcode streams in parallel

[Jerry Scharf]

Hi,

These are two completely independent machines with no coordination. Many
times only 1 will be operating. I am trying to work out what happens when
both are operating at the same time. It would be simpler to have two
separate controller, but that's not where we are at right now.

jerry

On Thu, Sep 24, 2015 at 6:23 AM, TJoseph Powderly  wrote:

> It doesnt sound like the task is well defined here on mail list.
>
> If the 2 independent(?) systems do not require interpolation _between_
> them, then M65 thru M66 could provide the 'handshaking'.
> Each would have a 'busy' and a 'fin' signal.
> This is extremely sequential.
>
> Else please explain more,
> and what did you see happen when
> ~"it didnt seem ready to accept commands" ?
>
> TomP tjtr33
>
> On 09/24/2015 03:00 AM, Gregg Eshelman wrote:
> > On 9/24/2015 12:57 AM, Dave Caroline wrote:
> >> might be better to let the two talk to each other in some way so move
> >> starts/runs are really synchronised properly rather than rely on
> >> buffers.
> >> eg classic ladder or some clocking or whatever
> >>
> >> Dave Caroline
> >
> > That sounds good. What you Do Not Want is a race condition where you
> > have two or more independent parallel processes with nothing to ensure
> > they always complete in the correct order.
> >
> > For an example of what can happen, google Therac 25
> >
>
>
>
> --
> Monitor Your Dynamic Infrastructure at Any Scale With Datadog!
> Get real-time metrics from all of your servers, apps and tools
> in one place.
> SourceForge users - Click here to start your Free Trial of Datadog now!
> http://pubads.g.doubleclick.net/gampad/clk?id=241902991=/4140
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users
>



-- 
Jerry Scharf
FINsix IT
650.285.6361 w
650.279.7017 m
--
Monitor Your Dynamic Infrastructure at Any Scale With Datadog!
Get real-time metrics from all of your servers, apps and tools
in one place.
SourceForge users - Click here to start your Free Trial of Datadog now!
http://pubads.g.doubleclick.net/gampad/clk?id=241902991=/4140
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Keeping track of machine (g53) position

[Dave Cole]

On 9/24/2015 5:09 AM, Gene Heskett wrote:
> I am about to go carve up a
> kilobucks worth of Mahogany.

Have you considered using southern yellow pine or fir for a test run?

Dave

---
This email has been checked for viruses by Avast antivirus software.
https://www.avast.com/antivirus


--
Monitor Your Dynamic Infrastructure at Any Scale With Datadog!
Get real-time metrics from all of your servers, apps and tools
in one place.
SourceForge users - Click here to start your Free Trial of Datadog now!
http://pubads.g.doubleclick.net/gampad/clk?id=241902991=/4140
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] multiple gcode streams in parallel

[Dave Cole]

FWIW,

The Modbus TCP interface in Classic Ladder supports both a client and a 
server, so peer to peer networking with connections to Hal and Gcode 
interlocks would be possible with the standard software.

Dave

On 9/24/2015 7:44 AM, Dave Caroline wrote:
> Actually this problem touches the independent axes problem that I
> foresee in my sliding head lathe where an axis that will be cutting
> later needs to start part way though another's coordinated move in
> order to save time.
>
> Dave Caroline
>
> On 24/09/2015, andy pugh  wrote:
>> On 24 September 2015 at 02:18, Jerry Scharf  wrote:
>>> I now need to extend it to run two separate robots off the same linuxcnc
>>> controller and receiver program
>> If you need the robots to synch then I think you need to be a bit cleverer.
>>
>> If you imagine one robot moving in XYZ space and another moving in UVW
>> space I think you need a pre-processor that "interleaves" the G-code.
>> This will include breaking up an XYZ move to start a UVW move and
>> vice-versa.
>>
>> All the axes of a G-code line reach the end point at the same time. So
>> to coordinate two robots you need to combine the moves onto one line,
>> otherwise you can't have both moving.
>>
>> --
>> atp
>> If you can't fix it, you don't own it.
>> http://www.ifixit.com/Manifesto
>>
>> --
>> Monitor Your Dynamic Infrastructure at Any Scale With Datadog!
>> Get real-time metrics from all of your servers, apps and tools
>> in one place.
>> SourceForge users - Click here to start your Free Trial of Datadog now!
>> http://pubads.g.doubleclick.net/gampad/clk?id=241902991=/4140
>> ___
>> Emc-users mailing list
>> Emc-users@lists.sourceforge.net
>> https://lists.sourceforge.net/lists/listinfo/emc-users
>>
> --
> Monitor Your Dynamic Infrastructure at Any Scale With Datadog!
> Get real-time metrics from all of your servers, apps and tools
> in one place.
> SourceForge users - Click here to start your Free Trial of Datadog now!
> http://pubads.g.doubleclick.net/gampad/clk?id=241902991=/4140
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users

---
This email has been checked for viruses by Avast antivirus software.
https://www.avast.com/antivirus


--
Monitor Your Dynamic Infrastructure at Any Scale With Datadog!
Get real-time metrics from all of your servers, apps and tools
in one place.
SourceForge users - Click here to start your Free Trial of Datadog now!
http://pubads.g.doubleclick.net/gampad/clk?id=241902991=/4140
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] multiple gcode streams in parallel

[Ralph Stirling]

I believe the Machinekit fork of LinuxCNC is designed to allow multiple
G-code interpreters and multiple motion modules.  I'm not sure how much
has been implemented and tested along those lines yet, but you might
check in with the machinekit email list.

-- Ralph

From: Jerry Scharf [jsch...@finsix.com]
Sent: Thursday, September 24, 2015 8:05 AM
To: Enhanced Machine Controller (EMC)
Subject: Re: [Emc-users] multiple gcode streams in parallel

Hi,

These are two completely independent machines with no coordination. Many
times only 1 will be operating. I am trying to work out what happens when
both are operating at the same time. It would be simpler to have two
separate controller, but that's not where we are at right now.

jerry

On Thu, Sep 24, 2015 at 6:23 AM, TJoseph Powderly  wrote:

> It doesnt sound like the task is well defined here on mail list.
>
> If the 2 independent(?) systems do not require interpolation _between_
> them, then M65 thru M66 could provide the 'handshaking'.
> Each would have a 'busy' and a 'fin' signal.
> This is extremely sequential.
>
> Else please explain more,
> and what did you see happen when
> ~"it didnt seem ready to accept commands" ?
>
> TomP tjtr33
>
> On 09/24/2015 03:00 AM, Gregg Eshelman wrote:
> > On 9/24/2015 12:57 AM, Dave Caroline wrote:
> >> might be better to let the two talk to each other in some way so move
> >> starts/runs are really synchronised properly rather than rely on
> >> buffers.
> >> eg classic ladder or some clocking or whatever
> >>
> >> Dave Caroline
> >
> > That sounds good. What you Do Not Want is a race condition where you
> > have two or more independent parallel processes with nothing to ensure
> > they always complete in the correct order.
> >
> > For an example of what can happen, google Therac 25
> >
>
>
>
> --
> Monitor Your Dynamic Infrastructure at Any Scale With Datadog!
> Get real-time metrics from all of your servers, apps and tools
> in one place.
> SourceForge users - Click here to start your Free Trial of Datadog now!
> http://pubads.g.doubleclick.net/gampad/clk?id=241902991=/4140
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users
>



--
Jerry Scharf
FINsix IT
650.285.6361 w
650.279.7017 m
--
Monitor Your Dynamic Infrastructure at Any Scale With Datadog!
Get real-time metrics from all of your servers, apps and tools
in one place.
SourceForge users - Click here to start your Free Trial of Datadog now!
http://pubads.g.doubleclick.net/gampad/clk?id=241902991=/4140
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

--
Monitor Your Dynamic Infrastructure at Any Scale With Datadog!
Get real-time metrics from all of your servers, apps and tools
in one place.
SourceForge users - Click here to start your Free Trial of Datadog now!
http://pubads.g.doubleclick.net/gampad/clk?id=241902991=/4140
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] multiple gcode streams in parallel

[Dave Cole]

>>It would be simpler to have two
separate controller, but that's not where we are at right now.

Simple is good.

For another $200 you can get a second PC.

If this is a one-off project I'd strongly consider throwing more 
hardware at the project.

I don't know what your labor rates are, but $200 doesn't buy much these 
days.

Are you guys making a machine to assemble/process your laptop charger 
product?   That looks nice.

Dave


On 9/24/2015 11:05 AM, Jerry Scharf wrote:
> Hi,
>
> These are two completely independent machines with no coordination. Many
> times only 1 will be operating. I am trying to work out what happens when
> both are operating at the same time. It would be simpler to have two
> separate controller, but that's not where we are at right now.
>
> jerry
>
> On Thu, Sep 24, 2015 at 6:23 AM, TJoseph Powderly  wrote:
>
>> It doesnt sound like the task is well defined here on mail list.
>>
>> If the 2 independent(?) systems do not require interpolation _between_
>> them, then M65 thru M66 could provide the 'handshaking'.
>> Each would have a 'busy' and a 'fin' signal.
>> This is extremely sequential.
>>
>> Else please explain more,
>> and what did you see happen when
>> ~"it didnt seem ready to accept commands" ?
>>
>> TomP tjtr33
>>
>> On 09/24/2015 03:00 AM, Gregg Eshelman wrote:
>>> On 9/24/2015 12:57 AM, Dave Caroline wrote:
 might be better to let the two talk to each other in some way so move
 starts/runs are really synchronised properly rather than rely on
 buffers.
 eg classic ladder or some clocking or whatever

 Dave Caroline
>>> That sounds good. What you Do Not Want is a race condition where you
>>> have two or more independent parallel processes with nothing to ensure
>>> they always complete in the correct order.
>>>
>>> For an example of what can happen, google Therac 25
>>>
>>
>>
>> --
>> Monitor Your Dynamic Infrastructure at Any Scale With Datadog!
>> Get real-time metrics from all of your servers, apps and tools
>> in one place.
>> SourceForge users - Click here to start your Free Trial of Datadog now!
>> http://pubads.g.doubleclick.net/gampad/clk?id=241902991=/4140
>> ___
>> Emc-users mailing list
>> Emc-users@lists.sourceforge.net
>> https://lists.sourceforge.net/lists/listinfo/emc-users
>>
>
>

---
This email has been checked for viruses by Avast antivirus software.
https://www.avast.com/antivirus


--
Monitor Your Dynamic Infrastructure at Any Scale With Datadog!
Get real-time metrics from all of your servers, apps and tools
in one place.
SourceForge users - Click here to start your Free Trial of Datadog now!
http://pubads.g.doubleclick.net/gampad/clk?id=241902991=/4140
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] multiple gcode streams in parallel

[andy pugh]

On 24 September 2015 at 19:24, Jerry Scharf  wrote:
> There is an axis for the second robot and the hal files are all set up.
> This is about the python interface and running two separate motion commands
> at the same time.

The Python interface sends MDI, MDI commands run in strict sequence.

Depending on the application the limit3 component might be a perfectly
adequate motion controller for a single axis at a time.
(You can just set the speed and target position and it will send out a
position command to a stepgen or PID)

-- 
atp
If you can't fix it, you don't own it.
http://www.ifixit.com/Manifesto

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

[Emc-users] linuxcnc.org website layout

[Axel Zöllich]

Hi,

every time I look to the linuxcnc.org site my eyes get stuck at the missing 
space between content and left side menu. Maybe someone has the chance to do 
this little change:






Axel

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

Re: [Emc-users] multiple gcode streams in parallel

[Gene Heskett]

On Thursday 24 September 2015 04:00:46 Gregg Eshelman wrote:

> On 9/24/2015 12:57 AM, Dave Caroline wrote:
> > might be better to let the two talk to each other in some way so
> > move starts/runs are really synchronised properly rather than rely
> > on buffers.
> > eg classic ladder or some clocking or whatever
> >
> > Dave Caroline
>
> That sounds good. What you Do Not Want is a race condition where you
> have two or more independent parallel processes with nothing to ensure
> they always complete in the correct order.

Well, the epic hairball that is NFS has been retired here, in favor of 
sshfs.

I have nuked all the fstab entries related to NFS mounts.

First, I installed sshfs on all machines.

Then on each machine I have created an /sshnet/machine_name directory for 
all the other machines.  That then was "sudo chown -R gene:gene /sshnet"
If left as owned by root, then the mounting must be done using a root pw, 
which doesn't exist on our LCNC installs.

Then on each machine (logged into the others with an ssh -Y login)
do
sshfs user@somehost:/just/the/path /where/i/want/it

and to unmount:

fusermount -u /where/i/want/it

Now any file manager like mc, can cd /sshnet/machine_alias and look at 
those directories accessable by gene.

>From any machine to any machine EXCEPT from "lathe" to "shop", that gets 
a "connection reset by peer" return.  So the machine named shop is still 
being a spoilt brat, but thats a hell of a lot closer to having a 
transparent file sharing local network wide than I had at 4:00 this 
morning.

The only fly is that when the sshfs link is mounted, those directories I 
made owned by me so I wasn't asked for a root pw that does not exist, 
have been chowned back to root:root.  If they stay that way, it will be 
more tap dancing to reset that before re-initializing the share after a 
reboot.

And it didn't take but 2 folks on the TDE list to make me aware of sshfs, 
and to show the example syntax pasted above.

Now, out to fix and add the new tool code to blanket-chest3.ngc that I 
wrote about way too early this morning.

> For an example of what can happen, google Therac 25

I think I did that once before.  I guess that was an instant legend in 
its own time. ;-)

Cheers, Gene Heskett
-- 
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
Genes Web page 

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

Re: [Emc-users] Keeping track of machine (g53) position

[Gene Heskett]

On Thursday 24 September 2015 11:06:48 Dave Cole wrote:

> On 9/24/2015 5:09 AM, Gene Heskett wrote:
> > I am about to go carve up a
> > kilobucks worth of Mahogany.
>
> Have you considered using southern yellow pine or fir for a test run?
>
> Dave

clear #1 white pine, (pricy stuff too) I have about 7 feet of it left for 
test stick construction. ;-)  Used about a foot in 3.5" pieces so far. 
Somewhat cheaper to make kindling out of than Mahogany. :)

Thanks Dave.

Cheers, Gene Heskett
-- 
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
Genes Web page 

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

Re: [Emc-users] Keeping track of machine (g53) position

[Dave Cole]

On 9/24/2015 3:16 PM, Gene Heskett wrote:
> On Thursday 24 September 2015 11:06:48 Dave Cole wrote:
>
>> On 9/24/2015 5:09 AM, Gene Heskett wrote:
>>> I am about to go carve up a
>>> kilobucks worth of Mahogany.
>> Have you considered using southern yellow pine or fir for a test run?
>>
>> Dave
> clear #1 white pine, (pricy stuff too) I have about 7 feet of it left for
> test stick construction. ;-)  Used about a foot in 3.5" pieces so far.
> Somewhat cheaper to make kindling out of than Mahogany. :)
>
> Thanks Dave.
>
> Cheers, Gene Heskett

I needed some scaffolding planks for a construction/painting project 
this summer and I wanted some 2 x12 planks 8 feet long.
The local Menards had #1 grade 2 x 12 Southern Yellow Pine boards for a 
very reasonable price and I found 6 of them that were entirely free of 
knots and sapwood.
I think they were about $14.00 each.Hard to beat that.

Dave


---
This email has been checked for viruses by Avast antivirus software.
https://www.avast.com/antivirus


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

Re: [Emc-users] multiple gcode streams in parallel

[Jerry Scharf]

There is an axis for the second robot and the hal files are all set up.
This is about the python interface and running two separate motion commands
at the same time. If I can do that, then a simple threading system works,
otherwise it needs to be fancier...

jerry

On Thu, Sep 24, 2015 at 10:47 AM, Karlsson & Wang <
nicklas.karls...@karlssonwang.se> wrote:

> > ...
> > I now need to extend it to run two separate robots off the same linuxcnc
> > controller and receiver program. This brought up a question about the
> > python interface.
>
> Add an extra axis ?
>
>
> --
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users
>



-- 
Jerry Scharf
FINsix IT
650.285.6361 w
650.279.7017 m
--
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

[Emc-users] AUTO: Jan Maier is out of the office

[Jan . Maier]

I am out of the office until 01.10.2015.




Note: This is an automated response to your message  "Emc-users Digest, Vol
113, Issue 81" sent on 24.09.2015 17:12:40.

This is the only notification you will receive while this person is away.


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

Re: [Emc-users] multiple gcode streams in parallel

[Jerry Scharf]

Dave,

This is what I thought and after talking to the person, we can wait the few
days to get a separate system in for him. Everyone will be happy this way.
It's a small system so I am going the BBB/cape/machinekit way (I hate it
when the controller is larger than the machine.).

Sounds like there is little interest in the remote gcode stuff, which is
fine.

jerry

On Thu, Sep 24, 2015 at 12:20 PM, Dave Cole  wrote:

> I think everyone is trying to find a way to do what you want with the
> existing software.
>
> In order to run two different Gcode programs at the same time, you would
> need two gcode interpreters and two planners running at the same time.
>
> That is not possible with the existing software that I am aware of:
> LinuxCNC or Machinekit.
>
> If someone tells me I am wrong, I would be happy to test it.  :-)
>
> That would be very nice to have for custom machines.   As in the case of
> gantry machines that need to transfer parts to another a joining gantry
> machine where they could physically collide.
>
> The ability to plug in different interpreters and planners has been
> discussed on the Machinekit email list but I don't know how far that has
> progressed.
> Keep in mind that the ability to run different interpreters and planners
> as plug in's is much different than running multiple instances of
> interpreters and planners which is what you would like.
>
> Dave
>
>
>
> On 9/24/2015 2:24 PM, Jerry Scharf wrote:
> > There is an axis for the second robot and the hal files are all set up.
> > This is about the python interface and running two separate motion
> commands
> > at the same time. If I can do that, then a simple threading system works,
> > otherwise it needs to be fancier...
> >
> > jerry
> >
> > On Thu, Sep 24, 2015 at 10:47 AM, Karlsson & Wang <
> > nicklas.karls...@karlssonwang.se> wrote:
> >
> >>> ...
> >>> I now need to extend it to run two separate robots off the same
> linuxcnc
> >>> controller and receiver program. This brought up a question about the
> >>> python interface.
> >> Add an extra axis ?
> >>
> >>
> >>
> --
> >> ___
> >> Emc-users mailing list
> >> Emc-users@lists.sourceforge.net
> >> https://lists.sourceforge.net/lists/listinfo/emc-users
> >>
> >
> >
>
> ---
> This email has been checked for viruses by Avast antivirus software.
> https://www.avast.com/antivirus
>
>
>
> --
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users
>



-- 
Jerry Scharf
FINsix IT
650.285.6361 w
650.279.7017 m
--
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] multiple gcode streams in parallel

[Jerry Scharf]

thanks for the input so far.

I know a second controller would be a better solution, but I need something
this week... I will look at some hardware for the next couple weeks. A
couple hours of programming now would save us time, which is our critical
resource.

I know about machinekit, but had trouble getting it running on the system
in question.

I was really hoping that someone would just answer the simple question of
whether having two different command streams could have motion commands
operating simultaneously on different axes of the same linuxcnc controller.

They don't let me near the production assembly systems. :) I just build
things for engineering and test. Yes, a small 65W laptop adapter is a
really cool thing.

jerry


On Thu, Sep 24, 2015 at 8:32 AM, Dave Cole  wrote:

> >>It would be simpler to have two
> separate controller, but that's not where we are at right now.
>
> Simple is good.
>
> For another $200 you can get a second PC.
>
> If this is a one-off project I'd strongly consider throwing more
> hardware at the project.
>
> I don't know what your labor rates are, but $200 doesn't buy much these
> days.
>
> Are you guys making a machine to assemble/process your laptop charger
> product?   That looks nice.
>
> Dave
>
>
> On 9/24/2015 11:05 AM, Jerry Scharf wrote:
> > Hi,
> >
> > These are two completely independent machines with no coordination. Many
> > times only 1 will be operating. I am trying to work out what happens when
> > both are operating at the same time. It would be simpler to have two
> > separate controller, but that's not where we are at right now.
> >
> > jerry
> >
> > On Thu, Sep 24, 2015 at 6:23 AM, TJoseph Powderly 
> wrote:
> >
> >> It doesnt sound like the task is well defined here on mail list.
> >>
> >> If the 2 independent(?) systems do not require interpolation _between_
> >> them, then M65 thru M66 could provide the 'handshaking'.
> >> Each would have a 'busy' and a 'fin' signal.
> >> This is extremely sequential.
> >>
> >> Else please explain more,
> >> and what did you see happen when
> >> ~"it didnt seem ready to accept commands" ?
> >>
> >> TomP tjtr33
> >>
> >> On 09/24/2015 03:00 AM, Gregg Eshelman wrote:
> >>> On 9/24/2015 12:57 AM, Dave Caroline wrote:
>  might be better to let the two talk to each other in some way so move
>  starts/runs are really synchronised properly rather than rely on
>  buffers.
>  eg classic ladder or some clocking or whatever
> 
>  Dave Caroline
> >>> That sounds good. What you Do Not Want is a race condition where you
> >>> have two or more independent parallel processes with nothing to ensure
> >>> they always complete in the correct order.
> >>>
> >>> For an example of what can happen, google Therac 25
> >>>
> >>
> >>
> >>
> --
> >> Monitor Your Dynamic Infrastructure at Any Scale With Datadog!
> >> Get real-time metrics from all of your servers, apps and tools
> >> in one place.
> >> SourceForge users - Click here to start your Free Trial of Datadog now!
> >> http://pubads.g.doubleclick.net/gampad/clk?id=241902991=/4140
> >> ___
> >> Emc-users mailing list
> >> Emc-users@lists.sourceforge.net
> >> https://lists.sourceforge.net/lists/listinfo/emc-users
> >>
> >
> >
>
> ---
> This email has been checked for viruses by Avast antivirus software.
> https://www.avast.com/antivirus
>
>
>
> --
> Monitor Your Dynamic Infrastructure at Any Scale With Datadog!
> Get real-time metrics from all of your servers, apps and tools
> in one place.
> SourceForge users - Click here to start your Free Trial of Datadog now!
> http://pubads.g.doubleclick.net/gampad/clk?id=241902991=/4140
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users
>



-- 
Jerry Scharf
FINsix IT
650.285.6361 w
650.279.7017 m
--
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] multiple gcode streams in parallel

[Dave Cole]

I suspect there is actually a lot of interest in being able to run 
multiple paths at the same time, but I also think it is likely a lot of 
development work to make it happen.

Dave

On 9/24/2015 4:05 PM, Jerry Scharf wrote:
> Dave,
>
> This is what I thought and after talking to the person, we can wait the few
> days to get a separate system in for him. Everyone will be happy this way.
> It's a small system so I am going the BBB/cape/machinekit way (I hate it
> when the controller is larger than the machine.).
>
> Sounds like there is little interest in the remote gcode stuff, which is
> fine.
>
> jerry
>
> On Thu, Sep 24, 2015 at 12:20 PM, Dave Cole  wrote:
>
>> I think everyone is trying to find a way to do what you want with the
>> existing software.
>>
>> In order to run two different Gcode programs at the same time, you would
>> need two gcode interpreters and two planners running at the same time.
>>
>> That is not possible with the existing software that I am aware of:
>> LinuxCNC or Machinekit.
>>
>> If someone tells me I am wrong, I would be happy to test it.  :-)
>>
>> That would be very nice to have for custom machines.   As in the case of
>> gantry machines that need to transfer parts to another a joining gantry
>> machine where they could physically collide.
>>
>> The ability to plug in different interpreters and planners has been
>> discussed on the Machinekit email list but I don't know how far that has
>> progressed.
>> Keep in mind that the ability to run different interpreters and planners
>> as plug in's is much different than running multiple instances of
>> interpreters and planners which is what you would like.
>>
>> Dave
>>
>>
>>
>> On 9/24/2015 2:24 PM, Jerry Scharf wrote:
>>> There is an axis for the second robot and the hal files are all set up.
>>> This is about the python interface and running two separate motion
>> commands
>>> at the same time. If I can do that, then a simple threading system works,
>>> otherwise it needs to be fancier...
>>>
>>> jerry
>>>
>>> On Thu, Sep 24, 2015 at 10:47 AM, Karlsson & Wang <
>>> nicklas.karls...@karlssonwang.se> wrote:
>>>
> ...
> I now need to extend it to run two separate robots off the same
>> linuxcnc
> controller and receiver program. This brought up a question about the
> python interface.
 Add an extra axis ?



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

>>>
>> ---
>> This email has been checked for viruses by Avast antivirus software.
>> https://www.avast.com/antivirus
>>
>>
>>
>> --
>> ___
>> Emc-users mailing list
>> Emc-users@lists.sourceforge.net
>> https://lists.sourceforge.net/lists/listinfo/emc-users
>>
>
>

---
This email has been checked for viruses by Avast antivirus software.
https://www.avast.com/antivirus


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

Re: [Emc-users] multiple gcode streams in parallel

[robert - Innovative-RC]

On 24/09/2015 12:44, Dave Caroline wrote:
> Actually this problem touches the independent axes problem that I
> foresee in my sliding head lathe where an axis that will be cutting
> later needs to start part way though another's coordinated move in
> order to save time.
>
> Dave Caroline
>
yep i have the same problem and not just with sliding heads, with a dual 
turret lathe where dual path control is much needed also.

one machine had two controls linked but it seems abit of a old school 
way todo things now days realy. where my sliding head has dual path 
control. i know newer controls can have much more than 2 program paths 
running.

  we just issue !L as a wait code when we want one program to wait for 
the other to get to a point, other wise they run at there own pase, we 
use !Ln also works and easyer to keep track off in a program, is the 
only way when more than 2 programs are running at once. (i know fanuc 
use M codes todo this,  above is from a Mitsubishi)

everything is in one program file so no loosing one half of the program, 
and 1 file to send to the machine.

here is a quick sample progam for reff if any one is intrested , from  
lathe with subspindle , power tooling,
http://pastebin.com/rcbu8mVa

here is a 3 path control machine in action
https://www.youtube.com/watch?v=C5YovD99rNM


-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.


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

Re: [Emc-users] Keeping track of machine (g53) position

[Gene Heskett]

On Thursday 24 September 2015 15:31:23 Dave Cole wrote:

> On 9/24/2015 3:16 PM, Gene Heskett wrote:
> > On Thursday 24 September 2015 11:06:48 Dave Cole wrote:
> >> On 9/24/2015 5:09 AM, Gene Heskett wrote:
> >>> I am about to go carve up a
> >>> kilobucks worth of Mahogany.
> >>
> >> Have you considered using southern yellow pine or fir for a test
> >> run?
> >>
> >> Dave
> >
> > clear #1 white pine, (pricy stuff too) I have about 7 feet of it
> > left for test stick construction. ;-)  Used about a foot in 3.5"
> > pieces so far. Somewhat cheaper to make kindling out of than
> > Mahogany. :)
> >
> > Thanks Dave.
> >
> > Cheers, Gene Heskett
>
> I needed some scaffolding planks for a construction/painting project
> this summer and I wanted some 2 x12 planks 8 feet long.
> The local Menards had #1 grade 2 x 12 Southern Yellow Pine boards for
> a very reasonable price and I found 6 of them that were entirely free
> of knots and sapwood.
> I think they were about $14.00 each.Hard to beat that.
>
> Dave

Yep, very reasonable. Lowes is a few miles closer than Home Depot, in a 
different direction.  So they usually get first crack at supplying 
kindling making material.

Cheers, Gene Heskett
-- 
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
Genes Web page 

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

Re: [Emc-users] Keeping track of machine (g53) position

[Gene Heskett]

On Thursday 24 September 2015 15:32:02 andy pugh wrote:

> On 24 September 2015 at 20:12, Gene Heskett  wrote:
> > So any diddling of maps will be done with a G10 L2 P2 axis's which
> > will adjust G55, leaving G54 in the as homed condition. IOW do the
> > probing while G54 is in effect, but modify G55 with the results of
> > the probing.
>
> It might help to note that G30.1 stores the current absolute position
> into #5181 et. seq.
> http://www.linuxcnc.org/docs/html/gcode/gcode.html#sec:G30-G30_1

I'll have to look at that.  What I do know from todays hacking so far, is 
that starting at homed, followed by an immediate G53 G1 X0 y0 z0, g10 L2 
P2 x0 y0 z0 which is what the display says after being homed (which is 
45mm's from the top of the post, and z home is parked at the instant the 
switch opens.

But when my code runs, doing that same G53 G0 x0 y0 z0, the display says 
z is at -1.9 something inches when it stops!

I can touch it off to zero it, but it comes right back (to 4 decimal 
places right of the dot) on the next run.

It is also yelping about program exceeds z axis limits, making me do a 
run anyway, single step by single step until I can see that it has 
screwed up or is about to.

What it is, is frustrating.  Nothing in a z move in this program takes it 
within 1/2" of the z home switch, or below the nominally 10" of down 
travel (270mm's) allowed in the ini file.

FWIW, there is an A axis defined in the .ini so it will be there when and 
if I cobble up a separate driver for my nearly worthless but motorized 
4" table.  Is there an interaction between the A axis even when there 
are no A axis moves in the gcode?  ISTR someone once claimed there 
was...

This is a dog and pony show that I've hit way to many times, and its an 
old enough squawk it seems to me like it should at least report the 
erronious line of code so we might stand a snowballs chance in hell of 
finding it in a 500 line program.  What debug= will enable that tracing?
Preferably tracing and reporting only the Z affecting code bits in this 
case.

Make me ask no one in particular if its 5 o-clock yet, someplace.  Since 
its past 17:00 here, I'll take a diabetic near (2.3%) beer and the 
manual for its debug info.

Thanks Andy.

Cheers, Gene Heskett
-- 
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
Genes Web page 

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

Re: [Emc-users] Keeping track of machine (g53) position

[Gene Heskett]

On Thursday 24 September 2015 17:27:29 Gene Heskett wrote:

> On Thursday 24 September 2015 15:32:02 andy pugh wrote:
> > On 24 September 2015 at 20:12, Gene Heskett  
wrote:
> > > So any diddling of maps will be done with a G10 L2 P2 axis's which
> > > will adjust G55, leaving G54 in the as homed condition. IOW do the
> > > probing while G54 is in effect, but modify G55 with the results of
> > > the probing.
> >
> > It might help to note that G30.1 stores the current absolute
> > position into #5181 et. seq.
> > http://www.linuxcnc.org/docs/html/gcode/gcode.html#sec:G30-G30_1
>
> I'll have to look at that.  What I do know from todays hacking so far,
> is that starting at homed, followed by an immediate G53 G1 X0 y0 z0,
> g10 L2 P2 x0 y0 z0 which is what the display says after being homed
> (which is 45mm's from the top of the post, and z home is parked at the
> instant the switch opens.
[...]
> Make me ask no one in particular if its 5 o-clock yet, someplace. 
> Since its past 17:00 here, I'll take a diabetic near (2.3%) beer and
> the manual for its debug info.

Which it turns out, has been excised from the new users pdf, pointing us 
at a header file in the source tree, which I of course do not have.  
Really now?  9 out of 10 of us are running the binaries from the 
buildbot, so remarks about genius's come to mind.

So please, put that data in the ini file DEBUG = 0x format back 
into the users docs so we can at least submit meaninbgfull terminal 
snapshots when stuff goes aglay.

> Thanks Andy.
>
> Cheers, Gene Heskett

After posting this tirade, I went back out and redid the z axis homing so 
it thinks z=0.000 is about 2mm's from the top of its available travel.  
This means I have to pre-position it above the switch or it runs the 
wrong direction looking for it.

Is there some way to make that initial search direction automatic?

And somewhere in all the restarts to play with that, the program exceeds 
z limits errors went away.  So now I am fine tuning that program I wrote 
for the toy mill to run with the slightly different offsets that moving 
the jig from the small machine to the bigger one needs to do its jig 
locations on.

Marching along, adding more global vars so there is only one place to 
change when something needs changed, the ER20 let go of the mill and 
dropped it into the jig, digging a slot about an inch deep in perfectly 
solid white ash.  And it was tightened about 1/4 turn from broke when 
the spindle was started.

This hex nut on the adapter has an eccentric internal ridge that may not 
be properly engaging the groove where the two angles meet on the outside 
of a collet, and the only reason I can see to explain its existance at 
all is to engage that groove on the collet to forcibly disengage the 
collet from the taper when the nut is loosened.

If that is the reasoning, I don't think this nut is working as it was 
designed to.  Once today I installed a mill and snugged it up pretty 
good, only to have the mill display a 1/8" run out at the end of the 
mill because this eccentric internal ridge apparently didn't drop 
properly into the groove in the collet, but was wedging it crossways.

Should I ask for a better nut, or will this one break in given enough 
time?

You fellows are the ER-8/16/20/32 experts. I never saw one of them in 
person until about 16:00 yesterday.

Anyway I am back to making progress, if I can keep a tool solidly 
mounted.

Thanks everybody.  Shuteye time in WV.

Cheers, Gene Heskett
-- 
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
Genes Web page 

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

Re: [Emc-users] multiple gcode streams in parallel

[Gregg Eshelman]

On 9/24/2015 1:20 PM, Dave Cole wrote:
> I think everyone is trying to find a way to do what you want with the
> existing software.
>
> In order to run two different Gcode programs at the same time, you would
> need two gcode interpreters and two planners running at the same time.
>
> That is not possible with the existing software that I am aware of:
> LinuxCNC or Machinekit.
>
> If someone tells me I am wrong, I would be happy to test it.  :-)
>
> That would be very nice to have for custom machines.   As in the case of
> gantry machines that need to transfer parts to another a joining gantry
> machine where they could physically collide.

What would be useful in such a setup is an ability to pass information 
directly from one LCNC host to another to coordinate functions such as 
transferring a pallet with work on it to another machine for additional 
operations.

One example would be two machines doing an operation that takes 30 
minutes, both feeding to a 3rd machine doing an operation that takes 
less than 30 minutes. With all three communicating, the 3rd machine 
could report when it's free and whichever feeder is farthest along on 
its job would be prioritized to transfer its pallet next.


---
This email has been checked for viruses by Avast antivirus software.
https://www.avast.com/antivirus


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

Re: [Emc-users] multiple gcode streams in parallel

[Karlsson & Wang]

> ...
> I now need to extend it to run two separate robots off the same linuxcnc
> controller and receiver program. This brought up a question about the
> python interface.

Add an extra axis ?

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

Re: [Emc-users] Keeping track of machine (g53) position

[Gene Heskett]

On Thursday 24 September 2015 08:27:17 andy pugh wrote:

> On 24 September 2015 at 10:09, Gene Heskett  wrote:
> > Am I making sense?
>
> Well, you have certainly confused me.
>
> But if you are talking about using probe results to set G54, then that
> is a normal thing to do.

But I want to preserve g54's inital mapping so I have a recovery for the 
next tool ability, so when I am done with this edit session, g54 will 
not have been touched, and G55 will be used as the remapped co-ordinate 
system.  That way, on a rerun after changing the tool and setting the 
new tool true with an editor, I still have  g54 to use to find the g38.2 
targets without wrecking the targets.

So any diddling of maps will be done with a G10 L2 P2 axis's which will 
adjust G55, leaving G54 in the as homed condition. IOW do the probing 
while G54 is in effect, but modify G55 with the results of the probing.

Is that clearer?

> What I normally do is run probe_hole (in the samples) as that parks
> the machine in the middle of the hole at the end, then I just "set
> origin" in Touchy to zero (which is a G10 L...) command.

This is hand coded as there isn't a hole, but 2  sides and the screw in 
the bottom holding it, half of a brass cube for the 1/4" end mill work, 
and nothing but a pcb for a z contact when the tool has been changed to 
the pointed roundover.  The only 2 other things that are changed is the 
tool_diameter and the z_depth, after the tool length has been found by a 
G38.2.

> I keep meaning to install the pretty probe screen.

I looked at that too, and may, if I don't forget about it.  But too, I 
want to bring in the camera I knocked off the toy and put it on this 
machine. Theres lots of space inside the bottom of the head casting on 
the GO704, might even be enough to hide the C920 if I can make it work 
on this machine but have not yet managed that trick.  But that is a 
couple priority ranks below making this gcode Just Work(TM). ;-)

Thanks Andy.

Cheers, Gene Heskett
-- 
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
Genes Web page 

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

Re: [Emc-users] multiple gcode streams in parallel

[Dave Cole]

I think everyone is trying to find a way to do what you want with the 
existing software.

In order to run two different Gcode programs at the same time, you would 
need two gcode interpreters and two planners running at the same time.

That is not possible with the existing software that I am aware of: 
LinuxCNC or Machinekit.

If someone tells me I am wrong, I would be happy to test it.  :-)

That would be very nice to have for custom machines.   As in the case of 
gantry machines that need to transfer parts to another a joining gantry 
machine where they could physically collide.

The ability to plug in different interpreters and planners has been 
discussed on the Machinekit email list but I don't know how far that has 
progressed.
Keep in mind that the ability to run different interpreters and planners 
as plug in's is much different than running multiple instances of 
interpreters and planners which is what you would like.

Dave



On 9/24/2015 2:24 PM, Jerry Scharf wrote:
> There is an axis for the second robot and the hal files are all set up.
> This is about the python interface and running two separate motion commands
> at the same time. If I can do that, then a simple threading system works,
> otherwise it needs to be fancier...
>
> jerry
>
> On Thu, Sep 24, 2015 at 10:47 AM, Karlsson & Wang <
> nicklas.karls...@karlssonwang.se> wrote:
>
>>> ...
>>> I now need to extend it to run two separate robots off the same linuxcnc
>>> controller and receiver program. This brought up a question about the
>>> python interface.
>> Add an extra axis ?
>>
>>
>> --
>> ___
>> Emc-users mailing list
>> Emc-users@lists.sourceforge.net
>> https://lists.sourceforge.net/lists/listinfo/emc-users
>>
>
>

---
This email has been checked for viruses by Avast antivirus software.
https://www.avast.com/antivirus


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

Re: [Emc-users] Keeping track of machine (g53) position

[andy pugh]

On 24 September 2015 at 20:12, Gene Heskett  wrote:
> So any diddling of maps will be done with a G10 L2 P2 axis's which will
> adjust G55, leaving G54 in the as homed condition. IOW do the probing
> while G54 is in effect, but modify G55 with the results of the probing.

It might help to note that G30.1 stores the current absolute position
into #5181 et. seq.
http://www.linuxcnc.org/docs/html/gcode/gcode.html#sec:G30-G30_1

-- 
atp
If you can't fix it, you don't own it.
http://www.ifixit.com/Manifesto

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