Hi Tom

re 'retreating from a circle to the apex of a cone':

the way it is now: the retract move goes to a absolute position or an offset of 
the current position where retract was hit (actually  using the current feed as 
I havent figured out how to properly pass or set this information).

how that position/offset is set is up to discussion, currently I just use HAL 
pins; the Fanuc controls seem to 'hand down' this value from G-code

wrt circle: it is possible to look at the type of the move in progress. Those 
would be from motion_types.h:

#define EMC_MOTION_TYPE_TRAVERSE 1
#define EMC_MOTION_TYPE_FEED 2
#define EMC_MOTION_TYPE_ARC 3
#define EMC_MOTION_TYPE_TOOLCHANGE 4
#define EMC_MOTION_TYPE_PROBING 5
#define EMC_MOTION_TYPE_INDEXROTARY 6

so it is fairly basic information, you dont have any feedback about the actual 
Gcode source context at this point.

You could, depending on the motion type, act differently on the motion type. I 
dont know how generally useful this would be, so I'm sticking to the Fanuc 
stuff for now on the assumption these operations were thought out and solve 
some real problem; I wouldnt really know where to start based on past linuxcnc 
discussions and wiki pages.

Note motion is an RT component in kernel, so the escape hatches to 
configurability are narrow. It would be interesting to think through how to 
make these things configurable through an optional user-programmable RT 
component, but I'm unsure whether HAL is the proper interface to do this. Or 
push this back to userland somehow; thoughts welcome.

There are many other issues to look at, eg spindle state, at-speed, mist/flood 
handling. Those can be snapshotted too, but the actions on retract and recover 
will vary with needs.

- Michael


Am 24.04.2012 um 23:58 schrieb Thomas Powderly:

> Michael Haberler,
> 
> re: the interrupt,retract, recover, continue
> Hello,
> 
> extremely interesting
> 
> can i try the code?
> or
> could you try retreating from a circle to the apex of a cone?
> ( that is equivalent to the 'safe position' in sink edm orbits )
> ( or,  apex of a pyramid while walking round the base of a square orbit )
> 
> being able to 'run away' to a safe place is.... way kewl
> like up to a safe plane but to one side of a path... (too many ideas
> occurring now )
> 
> if you can get a WEDM Fanuc operator manual
> you'll find buttons
> Cycle Interrupt,
> BreakPointReturn ( return to point in path where wire (tool) broke,
> just a signal that automatic automaticly causes a set of hdwr changes
> ( power flush wire drive...)))
> StartPointReturn   ( dont try to continue, restart the whole contour )
> CycleContinue     maybe other buttons
> 
> I'm not pushing WEDM, but it's a classic somethngWentWrong / fix-em-up
> problem handled by Fanuc
> 
> again, greatly appreciate your work
> thx
> tomp
> 
> ------------------------------------------------------------------------------
> 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-developers mailing list
> Emc-developers@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-developers


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

Reply via email to