I use G92 to set the XY and Z on my plasma cutter like so...

Touch off by probing down in a subroutine then set the Z by the amount 
of hysteresis my switch has.
G92 Z 0.070

In the main cutting program I set the X Y to 0
G92 X0 Y0
lots of cut moves... then clear the G92 values stored.
G92.1

You could take a look at the parameters 5211 - 5219 after doing a G92 to 
see what is changed. I would but for some strange reason all my 2.4.7 
sim configs are broken now....

John

On 2/5/2012 9:54 PM, gene heskett wrote:
> Greetings Guys;
>
> 2 questions actually:
>
> 1. What is the schedule saying about the renaming of this list to
> linuxcnc-users@etc?
>
> 2. I apparently am having trouble understanding the G92 family docs, which
> state:
> ------------------------
> G92 makes the current point have the coordinates you want (without motion),
> where the axis words contain the axis numbers you want. All axis words are
> optional, except that at least one must be used. If an axis word is not
> used for a given axis, the coordinate on that axis of the current point is
> not changed.
>
> When G92 is executed, the origins of all coordinate systems move. They move
> such that the value of the current controlled point, in the currently
> active coordinate system, becomes the specified value. All coordinate
> system’s origins are offset this same distance.
>
> For example, suppose the current point is at X=4 and there is currently no
> G92 offset active. Then G92 x7 is programmed. This moves all origins -3 in
> X, which causes the current point to become X=7. This -3 is saved in
> parameter 5211.
>
> Being in incremental distance mode has no effect on the action of G92.
>
> G92 offsets may be already be in effect when the G92 is called. If this is
> the case, the offset is replaced with a new offset that makes the current
> point become the specified value.
>
> It is an error if:
>
>      all axis words are omitted.
>
> LinuxCNC stores the G92 offsets and reuses them on the next run of a
> program. To prevent this, one can program a G92.1 (to erase them), or
> program a G92.2 (to remove them - they are still stored).
> ----------------
> Ok
> Then:
> ------------------------
>      G92.1 - reset axis offsets to zero and set parameters 5211 - 5219 to
> zero.
>
>      G92.2 - reset axis offsets to zero.
>
>      G93.3 Restore Axis Offsets
>
>      G93.3 - set the axis offset to the values saved in parameters 5211 to
> 5219
>
> You can set axis offsets in one program and use the same offsets in another
> program. Program G92 in the first program. This will set parameters 5211 to
> 5219. Do not use G92.1 in the remainder of the first program. The parameter
> values will be saved when the first program exits and restored when the
> second one starts up. Use G92.3 near the beginning of the second program.
> That will restore the offsets saved in the first program.
> ------------------------
>
> That seems to intimate that using a G92 axis number not only stores the
> desired position in the axis named, but also stores the rest of the axis
> positions in 5211-5219.
>
> What I am trying to do is add some position offsets on a per axis basis,
> but if a G92 actually updates all axis's to the current location except the
> one named in the G92 command, then the subsequent execution of a g92.1 it
> seems is restoring bogus numbers, as is evidenced when a new home operation
> is done, the z reading of 4.0681 doesn't revert to 0.0000.
>
> So, assuming the machine is sitting about .040" above my little brass tube
> I set in the pallet, nominally at X-0.3 Y+0.2, where I run holefinder.ngc
> in order to locate the precise location that brass contact within half a
> thou both ways.  This routine uses G38.2 a total of five times, and sets
> #100 to the real X location, and #101 to the real Y location.  I then use
> those #vars to move the machine to a position that is X-0.100, y0.000 from
> what should be the boards left front corner, and home x and y there.
>
> Then I raise it far enough to change out the drill chuck holding the
> conical tipped metal probe out for a collet and a 1/8" shank engraving bit,
> which raises the effective tool tip by about 4" since the collet is that
> much shorter than the drill chuck when its mounted,about 0.1" and run it
> down slowly with the down key until the bit contacts the pcb triggering the
> stop during a jog function.  At that point, I home the Z.  Its a bit
> abitrary as I do that same G38.2 once in the o<tedautoz>  call and use G92 Z
> to install a very small offset, a thou or less, in Z so I can fine tune the
> depth of the engraving done in the top.etch file.
>
> Now, in the remainder of the files, 4 more to complete the board, and at
> the present I am using the holefinder to locate a hole drilled in the
> board, which when the board is turned end for end, is used to install the
> offset value into the x so that the etch and bot.drill will result in the
> drill bit tips exactly meeting in the center of each hole.  I know I will
> have to break this holefinder into 2 files since I don't want to muck with
> the Y location when running it the second time.
>
> Obviously I have to use G92 to install the Z offset that compensates for
> the drill length, that is unavoidable from my reading.  I have no lengths
> in my tool table since I'm using a drill chuck, making those lengths
> arbitrary and requiring I G38.2 to find this out after every tool change.
>
> So what is the sequence of G9X's I must use in the remaining 5 files it
> takes to complete a board?, and which will not result in the machine moving
> its Z zero up in the air, both in axis display, and visually, about 4" for
> every tool change after the 2nd one.  It seems to me that I need to restore
> its reference such that if I ran it to X0.0Y0.0Z0.4, it should be sitting
> exactly over the corner of the board and not very high above it since that
> was located by a probe of similar length as one of the smaller drill bits
> originally.  That would seem to need a G92.3 at some point in each tool
> probe.
>
> At that point I seem to be lost and can't find the tree I want in all this
> forest.
>
> Thanks for any clues, or cluebats as the case may be, that will help solve
> my apparent loss of Z references problem.  Ideally, I would like to put all
> this stuff in the callable autoz function as that simplifies the editing of
> the pcb-gcode outputs to just adding the o<(ted/bed)autoz>  call after each
> M6 T# in the rest of them, and adding a spindle speed in the top of each
> file since pcb-gcode, while asking how long it takes for the spindle to
> spin up, doesn't ask for, or issue an S command.  Duh...
>
> I'll be reading that doc some more, but it all hinges on: Does a G92 Z#
> diddle the other 8 vars in the 5211-5219 group, or just the Z offset?
> It seems that what I want to do is restore the whole snapshot exactly is it
> was when last HOME'd, so that axis displays where it is in reference to
> this original HOME'd position.  Perhaps there is such a command and I'm not
> recognizing it?
>
> Thanks again guys.
>
> Cheers, Gene

------------------------------------------------------------------------------
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
[email protected]
https://lists.sourceforge.net/lists/listinfo/emc-users

Reply via email to