Hi Graham, it works very well. You move to any position inside of the object. Then call the gcode file. It will start up the m1299 macro which allows you to calibrate the center startpoint, size, and resolution of the array. You can set length of the leg and the degree of separation and the feedrate.
Then it will ask you to name the saved points file. At that point it turns control back over to the Gcode program.
At that point it makes a calibration run and then proceeds to the sub program. It then proceeds as programed from the calibration macro to perform the array. At completion it closes the points file.
It is not 100% able to get to all the possible nucks and crannies in one cycle but it will gather the bulk of the points and then allow you to use OTHER routines to gather the rest of the isolated areas. 1/2 arcs, 1/4 arcs, square inside and outside arrays, along with the circle arrays, etc
I am working on making a CMM page for MACH that will allow a host of functions to gather data points. Now that we have tip comp for the g31 that is a reality. Now IF we could get jog probing we would have it whipped.(;-)
PLease let me know when you have a copy of your book, my young ones need a good primer to fall back on my manual coding is so rusty the flakes fall off of me every time I sit down and write some code. I am trying to get them to learn it the manual way first then progress on to CAM. That way the CAM produced code will make sense.