I program offline. I have an issue that when I open my programs on the cmm's, my clearplanes and movepoints change, sometimes by many inches! I've actually had it occur on existing programs we've run a hundred times. Has anyone else experienced this? It's a pretty regular occurrence for us.
I program offline, but I have also programmed online in the past.
I only see this issue if I change something associated to an alignment, or add a local alignment, to a program that has run data.
I had this happen yesterday when troubleshooting a program online. I added a local alignment and said "yes" to the prompt, and everything seemed to be OK. (I usually say yes to the prompt offline to ensure my features stay where I originally intended.) The local alignment was a 1inch difference from what it was before, and I even did my due diligence and recalled my current alignment after the feature first, before creating the local alignment. I continued the program and all was well but later on in the program, it errored out for over-travel and all moves and clear planes were 25in off and the measured values of all features after that haven't been checked yet were off the same. Luckily, it errored out and didn't plow trough the part. I have a feeling that editing online is backwards from editing offline as far as when to say "yes" or "no" to that prompt, but I haven't tested that.
What I prefer to do is have a fresh, data free copy of the program on the network that no one else has access to and the CMMs run their program locally. If I make any adjustments to a program, I do it to the fresh copy and release it to the machines. This ensures no glitches happen and it's frustration free. It also ensures any weird glitches that happen over time can be fixed with no work other than transferring a file.
Thats what works for me but im sure there are 10 other solutions out there.
"Answering Yes/No incorrectly after changing an alignment would change everything, not just move and clearance points, so we can rule that out."
"I have noticed that this is more likely to occur if a program is stopped for any reason during prove-out."
Answering yes or no would do that, yes. This is occurring before I execute anything at all. No alignment updates, which would also do it. I'm literally opening the program, and the movepoints/clearplanes are changed. There's nothing going on in between there. In the other instances you've mentioned, I've seen that occur, and the reasons are obvious, as you've mentioned. That's why I'm here. There is no obvious explanation for what's occurring and I wanted to see if others have seen it. Apparently there are similar occurrences. We're having Hexagon app support come to our shop for two days to witness some of the strange issues we deal with.
Twice within the last three weeks the cmm lost the ability to load and remove probes from the probe rack, while switching probes during calibration! I've reloaded the master probe, reset the tips, re-found the sphere and recalibrated the rack without any issue whatsoever, but it will not load the master probe back into the rack. We just have issue after issue like this. Another common one is skipping a clearplane that causes a crash. The clearplane is clearly there, but it's as if pcdmis never read it. These issues occur on programs that have run 100's of times and on new programs alike. We had a probe run into the granite at 0/0 during calibration! The tech said it took 8 or 10 hits, moved away from the sphere and drove right into the table! How does that happen? This is not a homegrown calibration program either. It's the standard pcdmis calibration.
I'm not knew to this. I've been programming for about 18 years now. I'm certain I'm not making absent minded errors that are causing these issue. I have a template that I begin all my programs with and it's in a very specific order, so nothing is being missed, and all the parameters are preset in that template. I'm confident everything is correct. Monday I'll see if I can pull some of it from the program to post here.
"Answering Yes/No incorrectly after changing an alignment would change everything, not just move and clearance points, so we can rule that out."
"I have noticed that this is more likely to occur if a program is stopped for any reason during prove-out."
Answering yes or no would do that, yes. This is occurring before I execute anything at all. No alignment updates, which would also do it. I'm literally opening the program, and the movepoints/clearplanes are changed. There's nothing going on in between there. In the other instances you've mentioned, I've seen that occur, and the reasons are obvious, as you've mentioned. That's why I'm here. There is no obvious explanation for what's occurring and I wanted to see if others have seen it. Apparently there are similar occurrences. We're having Hexagon app support come to our shop for two days to witness some of the strange issues we deal with.
Twice within the last three weeks the cmm lost the ability to load and remove probes from the probe rack, while switching probes during calibration! I've reloaded the master probe, reset the tips, re-found the sphere and recalibrated the rack without any issue whatsoever, but it will not load the master probe back into the rack. We just have issue after issue like this. Another common one is skipping a clearplane that causes a crash. The clearplane is clearly there, but it's as if pcdmis never read it. These issues occur on programs that have run 100's of times and on new programs alike. We had a probe run into the granite at 0/0 during calibration! The tech said it took 8 or 10 hits, moved away from the sphere and drove right into the table! How does that happen? This is not a homegrown calibration program either. It's the standard pcdmis calibration.
I'm not knew to this. I've been programming for about 18 years now. I'm certain I'm not making absent minded errors that are causing these issue. I have a template that I begin all my programs with and it's in a very specific order, so nothing is being missed, and all the parameters are preset in that template. I'm confident everything is correct. Monday I'll see if I can pull some of it from the program to post here.
