We are in the process of changing all of our CMM programs to run in absolute speeds (MOVESPEED and TOUCHSPEED) to help reduce the measurement variation from CMM to CMM (4 different CMMs using 3 different controllers). I had Hexagon right me an application to do this in all of our programs but need to change the settings in each probe file (previous programmer literally created a different probe file for each program) and I really do not want to go into each probe file and make the change from % to absolute. Is there a an easier way to do this as there are 300 plus probe files?
I'm glad you're taking this on. Huge amount of waste is a common thing when it comes to probes in my experience thus far. Like
andrewb91 mentioned standardize this and you'll save a bunch of time a year. I wasn't on board with this idea at first but about a year later it clicked and took me no time to implement it as it was a new employer and new setup (about a year or more).
We keep a spreadsheet of probes and their tips, under each probe name column are program files that utilize them. Tips are in the order of calibration to avoid crashing while auto calibrating them all. Total calibration takes 1.25hr vs who knows exactly how much but we usually got told something from production after 2hrs of down time to calibrate to get the gears turning again.
I used a script program from here, I forget who posted it, it lists all probe files used in each program file in a directory. After that list is compiled, start to eliminate duplicate probes or find out which one probe can do the job of 2 for example a 1x20mm can probably do the same as a 1x21mm can. I cut my list down to a few from a couple dozen. Once you go thru and do that find out which tips are actually being used versus the ones you're calibrating, this is where I saved the most time in our auto calibration program, there were way to many unnecessary tip angles being qualified that no one program called/needed.
As far as probe settings... I manage them all from a subroutine every part routine calls. That way any time I need to adjust something, that is standard, I make one change in one place instead of every program. Touch speed, movespeed, max accel speed, prehit/retract, check distance, etc... oh and sounds like you're getting away from percent to absolute good move.
Tips are in the order of calibration to avoid crashing while auto calibrating them all.
This will be fixed with 2019. It gives you the option to run collision detection on calibration. I have a probe that is about 12" long and has to calibrate 25 angles that I have to select in order. With the collision detection, it adds moves to avoid collisions with the column and I no longer have to select the calibration order.
Correct. All of the old programs were written with machine speed set to percent. Repeatability of measurements between CMMs is as issue as some of our tolerances are +/-.0002". Two CMMs max out at 500mms and the other two max out at 300mms and while not a huge difference repeatability study shows up to .0003" variation when running them in percent. Absolute measurement study cut this in half to .00015"
