--- In DynoMotion@yahoogroups.com, Tom Kerekes <tk@...> wrote:
>
> Hi Dave,
>
> Not sure what you mean by adding messages.  But you should be able to do everything using a subroutine without changing anything in the Interpreter.
>
> Then later if you modify the Canned Cycle to include M Codes you can use existing M Codes (Since they will be hidden from the Operator it shouldn't matter which ones they are).
>
> If an M Code calls a C Program you can easily add any required delay in the C Program.
>
> Regards
> TK
>
>
>
>
> ________________________________
> From: daves3891 <fahque99@...>
> To: DynoMotion@yahoogroups.com
> Sent: Tuesday, March 26, 2013 11:09 AM
> Subject: [DynoMotion] Re: Motion Clamps (Again)
>
>
>  
> That sounds exactly like what I need.
>
> I did change a few other things to the interpreter code, I also made MAX_MCODE_ACTIONS_M1 21 instead of 11. This has allowed me to get a valid call back from the GUI.
>
> But if I am modifying the interpreter then I guess I don't need to worry about the GUI callback as much.
>
> I might also need to add a delay after the M100 / M101 (clamp unclamp)
>
> Would I basically just be able to add the correct messages to the rs274ngc (eg, set a M code)?
>
> Thanks
> Dave
>
> --- In DynoMotion@yahoogroups.com, Tom Kerekes <tk@> wrote:
> >
> > Hi Dave,
> >
> > I think there is a few more things to do to add additional M Codes.   That change just tells the GCode Parser that it is ok for the MCode to be in the GCode, but additional code would be required to actually make it do something.  As well as GUI changes to allow the GCode to be configured by the User.
> >
> > I'm thinking that the best approach since you must conform to existing GCode and are able to modify the Interpreter would be to modify the Canned Drill Cycle to add the additional Clamp functionality.  I think this would be relatively straightforward with several MCode Calls added.
> >
> > I suggest using a Subroutine first to prove the concept.  Once it is working you can modify the Interpreter's Drill Cycle to do the same thing.
> >
> > The Subroutine would consist of:
> >
> > #1 Move XY using G0
> > #2 M100 - call c program to wait until in position and clamp (abort on timeout)
> > #3 Drill Z using G1
> > #4 Retract Z
> > #5 M101 - unclamp
> >
> >
> > Does this sound reasonable?
> > Regards
> > TK
> >
> >
> >
> > ________________________________
> > From: daves3891 <fahque99@>
> > To: DynoMotion@yahoogroups.com
> > Sent: Tuesday, March 26, 2013 7:20 AM
> > Subject: [DynoMotion] Re: Motion Clamps (Again)
> >
> >
> >  
> > Hi Tom,
> >
> > I modified the interpreter because it was giving me a unsupported M code error when I was using M15, I just modified the _ems[] in the rs274ngc file to use group 10 instead of -1. I think this was the correct way to do it but I am not 100% sure.
> >
> > I do not need a dummy canned cycle, I need the standard drill cycle but have it activate the clamps before the Z starts moving and deactivate when it is retracted (only is M15 has been set in the beginning of the G code)
> >
> > The motion clamps will only be used with canned drilling cycles
> >
> > We have this feature on one machine that is a custom DOS based software with a galil board and a second newer machine that has a Fanuc controller and does the motion clamps with the built in ladder logic. I would like to get this working on our third machine with the kflop and if it is successful and reliable use the same system to replace the galil system.
> > I need the system to be accurate as some of the work pieces we drill are $20k+ for just the material.
> >
> > I looked at adding something in to the interpreter code itself, but I really do not fully understand how I could get a message from the rs274ngc to the kflop. I am confused about the flow from the interpreter messages to the board.
> >
> > That is why I was thinking the C program monitoring the z position might work well, but I am still open to suggestions on the best way to accomplish this.
> >
> > Thanks
> > Dave
> >
> > --- In DynoMotion@yahoogroups.com, Tom Kerekes <tk@> wrote:
> > >
> > > Hi Dave,
> > >
> > > I don't really understand what you did with the M15 and M16 or why you needed to modify the Interpreter source code to make an M Code just set a bit.
> > >
> > > An M Code will naturally wait for any previous motion to stop before executing.  And the Interpreter can be set to wait for an M Code to finish before going on.  So there isn't really any need to start/stop the Interpreter.
> > >
> > > From the earlier thread I thought you wanted to use the canned drill cycle with dummy Z motion?
> > >
> > > Regards
> > > TK
> > >
> > >
> > >
> > >
> > >
> > >
> > > ________________________________
> > > From: daves3891 <fahque99@>
> > > To: DynoMotion@yahoogroups.com
> > > Sent: Monday, March 25, 2013 1:02 PM
> > > Subject: [DynoMotion] Motion Clamps (Again)
> > >
> > >
> > >  
> > > I had discussed how to add motion clamps on CNCzone and I had a few more questions and I decided to move the topic here.
> > >
> > > I have modified the interpreter to let me use M15 and M16 with the call backs so that part is done.
> > >
> > > Now I want to use a C program to do the following for the motion clamp cycle.
> > >
> > > Here is a basic program flow
> > > - Check for the motion clamp bit (set by M15)
> > > - Monitor for the X and Y axis to both stop
> > > - Pause or stop the interpreter/buffer
> > > - Activate the Motion clamp output
> > > - Wait for 1 second
> > > - Start the interpreter
> > > - Wait for the Z axis to return to the start position
> > > - Pause or stop the interpreter/buffer
> > > - De activate the Motion clamp output
> > > - Wait for 1 second
> > > - Start the interpreter
> > >
> > > Does this sound feasible with a C program?
> > >
> > > What is the bast way to monitor if the axis has stopped moving and pause the interpreter?
> > >
> > > Any suggestions would be appreciated.
> > >
> > > Thanks
> > > Dave
> > >
> >
>