turning an alignment

[[Mi...]]

ok, I have a few parts that I programmed with the longest length in the x axis. Runs fine on the old Eclipse but it's too long for the Contura and I'd like to turn it 90°. I turned it by recreating the base alignment and when I opened the rotate stylus dialog box the simulation of the probe was somewhat confused. I immediately closed without saving. Is there a way to rotate a programmed part correctly? Was I on the right track?

[[No...]]

Feel....urge to.....must...resist...!!
Nooo......

viewtopic.php?f=19&t=2620

[[Mi...]]

thanks?... er...um.. I think??

Yeah, pretty much what I did and when I was checking my probe orientations I noticed the simulation of the probe and body were in disarray. A sure sign to me that something was wrong so I reacted properly and paniced. Maybe I just had to adjust my clearance planes before checking my probe directions.

[[Ne...]]

Michael,

Calypso does not require your base alignment orientation to match the machine orientation. I do not know how the simulation
software works because I do not use it. All you had to do to run the part on your Contura was to physically move the part and
update the stylus system. For example if you had a stylus at 0,90 you might have to change it to 90,90. For another example any stylus that were at 0,0 could the stay the same. I have parts in airplane coordinates and my base alignment isn't even square to the machine axis nor my x,y,z origin even located on the CMM.

I hope this helps,

Neil

[[Mi...]]

Neil, thanks, that makes sense. All I would need to do is turn the part 90° and change all my RDS orientations by the same 90° and I'd be good to go. The only thing I would need to do is add a note to turn the part 90° in say, the CCW direction from as shown. I know the operators and they would be upset and there would be great wailing and gnashing of teeth.... I would rather change the program if that is as easily done. I wonder if I would need to change the space axis on every operation. If that is the issue then I would rather just turn the part and deal with the wailing!

[[Ky...]]

Just make sure if you have a star probe that the probes and clearance planes are correct after rotation. That is the only thing I have found that Calypso doesn't do well when you rotate a part in a proven program.

[[Mi...]]

Kyle, no star probe. I would like to make an L probe for my RDS but that's another thing. When you talk about rotating a part do you mean physically or in the program?

[[Ky...]]

I mean, if you program it one way then rotate it in your program Calypso will change almost everything except the star probe and clearance planes for star probe.

[[No...]]

Please sign in to view this quote.

Don't worry about the little joke 😉 This topic is becoming quite repetitive lately.

Please sign in to view this quote.

Until a few years ago Calypso required you to orientate the part on the simulated machine table along the same axis directions as the machine coordinate system. Otherwise weird things would happen, like tilted machine parts and probes.
Those times are over, but I still wouldn't bet on everything being properly displayed and (collision) checked in cases like this. The simulation still has its nasty quirks.

[[Am...]]

Are all the probes used in the program in the 1 position?

[[Mi...]]

Kyle, yes, it worked fine and everything looked good until I looked at the probe simulation.

Amel, yes they are all in position 1.

Norbert, I kinda saw it coming, 😃 . A lot of my programming is done in simulation since the programs are usually large. It's easier to do them at my desk and then move to a machine to prove them. So when the simulation went wonky I pulled the plug right away. I lost a couple programs when that sort of thing happened in the past and I couldn't afford it this time. I think it worked correctly, I think it might have been fine and it was just the simulation looking funny. Maybe I'll save the program as a junk program and try again, that way if it gets corrupted in some way I won't feel like jumping out a window or something.

[[No...]]

Some screenshots I kept from the old days. Not modified! (only the part greyed out)

[[Mi...]]

yeah that's basically it. It's just in the simulation then and doesn't effect the program? I was wondering if after turning a part in the program do I need to change all my Space Axis callouts in each feature or does it do that automatically?

[[Mi...]]

ok, did a little experimenting. I created a plan with a basic rectangle part that had a few holes in the top and sides, aligned it and measured it all quickly. Turned the RDs and did the sides, some hole to hole locations, TP's, flatness, a couple XY locations for the holes... stuff like that. Ran the program, everything good.

I then changed the direction of the planar rotation and turned the part, kept the clearance planes as is for now. Then changed the clearance planes the way I wanted (because I just don't trust calypso to have my best interests at heart... ). Changed the probe directions and ran the simulation. Yup, the probe simulation went wonky. All my hole locations were backwards, in the feature they would read like X.375 Y9.750 but in the characteristic they still read like X-9.750 Y.375. I went into the feature and turned them off and back on and the characteristic corrected itself. I then changed my hole to hole characteristics from x to y since they were in that direction now and noted that everything else measured good. I also noted that the probe was no longer broken apart in the sim.

It seems like it didn't like the X and Y's switched but as soon as they were corrected by turning them off and then on again all was right with the world. So turning a part with the alignment works, just a bit of maintenance afterwards. I didn't have to change any Space Axis, that was done automatically I guess. It looks like if the probe is wonky, there is something in the program that might need to be double checked and it is usually shown with a red mark in the characteristic.

------------

Now I created the same plan but I didn't create any characteristics. Turned the part same as before and the probe came apart. This time I had to make all the changes as well as having to change the space axis this time. I got a feeling this is going to take some time to get right.

[[No...]]

Please sign in to view this quote.

Well, it depends...

But I would never trust that what the sim is showing me is a close reproduction of reality. Because the simulation is not exactly a simulation in the usual sense. If it was, it would have to measure "real" points on the model and react accordingly. Meaning that if you turned the model by 5 degrees, so the probe touches it early, it would have to take exactly the resulting point and use it for further calculations. But this is not the case - and not the intention, by the way.
All it does is to move the machine according to the programmed features and paths and check for collisions on the way. But the resulting points are purely artificial and have nothing to do with the CAD model and its orientation. They're only generated from the nominal feature geometry. That's why you can never fully trust that what you see on the screen depicts reality.

[[Mi...]]

yes I understand that. But then again, I never saw my probe come apart and move around like this either, 🤣