Re: [CAD_CAM_EDM_DRO] MasterCAM Tutorials, postprocessor, and NC file generation questions

On Feb 8, 2008 5:49 AM, Stephen Wille Padnos <spadnos@...> wrote:>

> Michael Fagan wrote:
>
> >[snip]
>
> >
> >> >Also, since my Taig has manual, non-preset tools (it's got an ER-16
> >> >spindle), I always run my ops as separate programs, one program for
> each
> >> >tool.
> >>
> >> You can use all your tools in the one program if your set up right. I
> >> start with tool one Zero 'd to stock surface or a reference point and
> >> when Mach3 stops for a tool change, simply change tool, jog back to
> >> surface or ref point, press zero Z then press cycle start, and off you
> >> go again.
> >>
> >>
> >I run EMC2, so I'll have to see how to do that. I'm not sure how it's
> >configured to react when it reaches a tool change cycle, I'll have to
> >check that out. If it doesn't work with a tool change, it might work
> >with a M01 optional stop, I'll try it out and see.
> >
> >
> There's a sample config that uses a program called
> "hal_manualtoolchange". This is a program that pops up a dialog box
> when you need to switch tools. The core motion controller just outputs
> a request for the new tool and waits for a signal that it has been
> changed - you connect those signals to whatever you want (ladder logic,
> physical I/Os, the manualtoolchange program - whatever). You can see
> how it's used in the sim/axis config.
>
> You can't do it the way Steve mentioned with EMC2 - you can't perfom
> manual motions while a G-code program is running, and a toolchange is
> part of the running program.

That's what I thought, since all the buttons are grayed out when a
program is running or paused.

>
> With a sensor on the machine (a simple contact switch) and a little
> extra stuff in the G-code program, you can have EMC2 measure the length
> of the new tool once it has been changed. There's an example in the
> file "tool-length-probe.ngc" in the nc_files directory.

I'll check that out. Do you actually use a tool length sensor of some sort?

>
>
> >So far all of the parts I have run have been plates, which I secure to
> >my tooling plate. A couple of dowel pins and 2-4 #10 SHCS hold the
> >plates to the table. I run all of them through one tool (usually 6-10
> >pieces for each part), then load the new program, new tool, and run
> >them all through again. I've been checking pretty carefully and I
> >haven't had any registration problems between the different runs. The
> >first op is usually the most precise, bearing holes, for example, and
> >then the rest of the ops are for various bolts, standoffs, and such.
> >
> >
> If you don't move the workpiece on the table, and it's well clamped
> down, there should be zero registration error. Any errors you get
> between tools (I'm assuming X/Y here) is a mechanical problem with the
> machine.

I haven't had any problems yet. Most of the parts have been in the
assembly stage over the last couple of days, and if things like
bearing press fits and gear center to center distances (the most
critical parts) weren't working I'd be hearing about it.

>
>
> >> Steve Blackmore
> >>
> >>
> >Thanks for the good advice.
> >Michael
> >
> >
> - Steve
> (the other Steve)
>