x axes continuos

hi to all
i have a particular machine (two axes)
the x axis move a long film for about 3000 mt only forward
the y axis ( 0-200mm ) print on film a pattern that repeats every xxx mm along the x axis
the start of y is the same of the y end but not x .
what would be the best solution for managing the axes so that the cam can pass only the first pattern?

best regards
Alessandro

Hi

If you are feeding the output directly from CAM, you are probably going one stage too far in automation and removing the flexibility you need.

If you had the print as a gcode subroutine, you could just incrementally move the X axis forward to the next desired start point and call the sub.
Would just need to touch off at the start point at the beginning and then start the outer loop of the program for however many iterations.

There are other variations you could do loading files instead.

regards

hi
I would think of something like this:

#10
g21
g90
g91
o103 repeat [5]
o101 if [#10 GT 0]
(AXIS,hide)
o101 endif

(cam program start)
g1 x50 y0 f1000
g1 y50
g1 x50
g1 y-50
g92 x0
#10=[#10+1]
(cam program end

(AXIS,show)
o103 end repeat
m2

repeat -->>>> 10000000
it is right but :
in the display, the cone tool moves and exits the video.
I would like the cone tool to remain in the middle of the video and would shift the path
what do you think?
Ale

in the display, the cone tool moves and exits the video.
I would like the cone tool to remain in the middle of the video and would shift the path
what do you think?

It is not the depiction of the cone tool you need to worry about, it is that when the subroutine starts
it will reverse the incremental X move and print on top of the last print

You need to put X at 0 on the current print, do the incremental X move and then reset X0 to the current position
www.linuxcnc.org/docs/devel/html/gcode/gcode.html#sec:G10-L20

regards

PS
G92 X0 should work, but as you are never going back to previous positions so don't need to restore etc, G10 is simpler

I have a feeling that i have seen G92 produce some strange results in the plot before also, whereas G10 should bring the
XY intersection indicator and tool cone back to the new touch off point

hi ArcEye

I tried G10 L20 P1 X0.0 before endrepeat and seems to be right:

g54
g21
o103 repeat [5]
g91
g1 x50 y0 f1000
g1 y50
g1 x50
g1 y-50
g10 l20 p1 x0.0
o103 endrepeat
m2

in the video instead, the cone tool continues to move in the positive x direction and does not return to the start point of the G54.
Something is missing?
Regards
Ale

In the video instead, the cone tool continues to move in the positive x direction and does not return to the start point of the G54.

I see what you mean, the plot shows the new origins but does not alter it within view.

What sort of X axis do you have that is notionally 3000 metres long?

A conveyor belt, or are you just winding a spool on the film and the bed underneath remains still, or are you combining either of those with a moving bed?

regards

It is a tape on which is printed a pattern.
The tape is moving along the x axis as if it was a conveyor belt.
The y-axis (200 mm stroke) pursues him according to the cad cam design .
This is repeated for the entire length of the tape which varies from 100 m to 3000 m.

Some idea for displaying a pattern only during repeat?

thank's
Ale

Either way I always have the same visualization

Unfortunately, I cannot see any way to get the display of the print in the same place each time.

Due to the assumptions made by the interpreter, a move of 100mm say, will always result in the next print being displayed at a point 100mm from the first one.
The concept that the X axis has essentially not moved, rather the workpiece has moved across the X plane, is not one that is catered for.

If it was displayed, it would display in exactly the same place that there was already a trace from the last print, so the usefulness might be lessened anyway.

The only software solution would be to automatically centre the origin point within the screen whenever it is reset.
OpenGL programming is not my forte and such an implementation would be wrong for every other conventional machine, where the origin point is displayed within the context of a finite work envelope.

If your film was on a powered reel, moving across an essentially static table, I could envisage setups that moved it without using the X axis and thus would produce the plot you desire.
( advancing it outside the knowledge of the plot interpreter through an M code activated motor movement say)

The main thing is the code will produce the repetition you require hopefully, you may just have to live with the plot.

regards