Jump to content

Model replaced and reference lost


---
 Share

Recommended Posts

My 3D model was either corrupted or damaged, but it wouldn't render properly, and even though the program ran fine, I couldn't add more features so I decided to replace the model with the same thing by inserting a new one. Right after that, I noticed that when I measure a vertical cylinder on the left side with the right probe (#3) it works fine, but when going around to scan the right half with the left probe (#5) it approaches the part from 4-5 inches away and cannot "find" the part and alarms out. The same thing happened with other features using the similar method of measuring and I cannot figure out the reason. The base alignment seems to be right because it does other measurements, only the ones that go around the part with different probes are the problem ones.
I am almost certain it has to do with clearance planes, except cannot find information on it.
Clearance planes are the cause of many of my issues and not able to clarify most of them. I am a CNC programmer, dealt with coordinate systems for decades but ZEISS's clearance planes defining rapid moves, references seem to be a little difficult to adjust to.
Can anyone throw me a lifeline?

Most appreciate any help
Link to comment
Share on other sites

When you put the new CAD model in, did you do the CAD model comparison? This may help to restore the features to your model and allow to add new features and re-establish the clearance zone.
Link to comment
Share on other sites

Or, when you put the new model in, did you first make sure it had the same origin as the program?

The way I do this is load the model up in a brand new program, then create a base alignment using features selected from the model. Using the same features and whatever rotation/offsets are used in the real program. Once that is done I save the model.

Then you can load that model into the original program and everything is in the right spot and orientation.
Link to comment
Share on other sites

Please sign in to view this quote.

Where would I save the model to, separate from the program, which folder? Any special place, or do I designate any good old folder and import back from there?
Link to comment
Share on other sites

Please sign in to view this quote.

It doesn't really matter. You can overwrite your original CAD file or save it as another name or location. The important thing is to remember the new name and location so you can bring it into your measurement plan again after the transformation.
Link to comment
Share on other sites

Please sign in to view this quote.

Yes you can save the model wherever you like for future use. Each Calypso program with a model creates a copy of that model in its own directory. So once you load the new model in and save the program it has overwritten it within the program directory.

Open any program that has a CAD model with windows explorer and you'll see files like model.fct, model.jpg, model.sab - those are the files that will get overwritten when you load a new model into a program.
Link to comment
Share on other sites

I appreciate all replies and they are helpful with the model issue, but with regard to the probe movement I am still in the hole.
For some reason, even with a different program, or any like program as I tried more than one, that ran fine before, I seem to have a problem when probing anything on the right side of the part, and that corresponds with the original problem I was having. I show this on the attached picture that has a CP -X, and a 0.200 retract distance coming in at 280.00 degree to sweep around the feature for 160.00 degrees and still approaches the part from over an inch away, naturally never touching the surface.
Any ideas for the possible cause?

probing too far out.JPEG

Link to comment
Share on other sites

Please sign in to view this quote.

sounds like bad manual alignment.

but first lets fix your issue with the cad model. its possible multiple things might of screwed things up.

Hopefully you have a copy of the program that has the corrupted cad model without any of the changes you made after loading the new model.
Load that up.
Load the new model.

Now look at your program's base alignment. What features does it use? there should be three main ones that define the x y and z axis.
open up the first feature from the program then click on that feature on the cad model.
Look at the axis that its being defined. The program will show 0. The cad model should also show 0 for that same axis. If it does not, then you need to adjust the position of the cad model only.
Do the same thing for the other features of the base alignment.

In order to move the location of the cad model only, you go to CAD > Modify CAD Entities > Hierarchy Tab > Click on the Cad model only that should end in .stp or whatever file extension your cad model uses. > That will highlight your cad model > Transformation.
There you can enter how much each axis is off on the cad model to the program.

Lets say Plane1 defines your X axis on the program for your base alignment. So it should display 0 on the nominal section when you open up that Plane1. Then when you click on that same feature on the CAD model, it shows that the nominal for X axis is 2.334.

So then you go to CAD > Modify Cad Entities and then put -2.334 on the X axis of the Transformation Tab. then hit apply.
That will move the cad model feature to where it is on the program.

Once its fixed up, run the base alignment using manual alignment, and it should match up to before.
That should fix the issue of your probe going a different direction

just remember to use 4 points for planes and circles and 8 for cylinders.
Link to comment
Share on other sites

Please sign in to view this quote.

Can you show a screenshot of your base alignment? If that is the part you're checking then having a feature on the planar rotation might throw the probes off if it's a symmetrical part all around.
Link to comment
Share on other sites

Please sign in to view this quote.

I checked for the location of the model and it shows 0 on all 3 axis, and I did another manual alignment and it still does the same thing: scans 1 half of the cylinder on the left just fine, then as it moves to position to scan the 2nd half on the right but moves away from the part too much, over about an 1.5" or so, on the Y axis, (-Y ) then alarms out because it cannot find the part.

To be accurate: I scan the cylinder toward the top end with the "down" probe #1 with a 360-degree sweep, then for the lower circle of the cylinder I use the right probe #3 on the left half and move to the right side to use the left probe #5 to scan the right half, and that's where the issue is, on the right side only.

I have used this method many times to scan round features in two steps and never had a problem doing so. In the CNC programming world I tell the machine to move to a point by a G code either rapid or feed, and it is cut-n-dry, I see and can predict every move, clearly, but here with the CMM, I have no clue (well, some clearance data) what makes it move to a particular place besides the direct probing move, cannot single step it to see where I am in the program, what the next move may be, and that adds to the "mystery". Clearance planes are still clouded in a fog of mystery to me, and I must have been lucky for years creating hundreds of programs successfully and I thought I had an understanding of it, but apparently, I don't. It is clear to me that I am doing something wrong, naturally, but cannot find what it is, and Calypso stubbornly isn't giving me a clue as to what the remedy might be. It sometimes wants to change the clearance planes automatically after I make a change, and I agree to it, but the result remains the same, it moves out on -Y too much to probe the part.
Link to comment
Share on other sites

Please sign in to view this quote.


And I am still looking for a good source to learn and clarify the concepts and the relationships between clearance groups and clearance planes and anything and everything related to them. I suspect most of my troubles originate there.
Link to comment
Share on other sites

Stupid Question.

If there any chance you screwed up calibrating that left probe? Look at the Stylus system management window at the coordinate values. Your left and right should both have about 0 in Y, but Positive and Negative X values. Z should be basically the same between both.
3535_1e7c1180718468a8ca6be5e25170b385.png
It knows where the part is so it's nothing like that. But if that left probe was manually calibrated incorrectly. Remember after you manually requalify a probe, any calibration program will just keep doing whatever you did manually without error. So if that Left probe was accidentally calibrated as something else, it won't get fixed until you calibrate it manually, correctly.
Link to comment
Share on other sites

It looks like that may be the solution!

Just remember you will have to do it manually with the "Qualify Stylus", that's so far off that it's going to miss the ball just as much as it was missing the part.
Link to comment
Share on other sites

Please sign in to view this quote.

That seemed to have been the problem. I re-qualified the probes and now they scan properly, as intended.
I have no idea what caused the left probe to be off , but thanks for the advice, it fixed it.
Link to comment
Share on other sites

Please sign in to view this quote.

Glad to hear it!

For it to be off that much, I suspect it was calibrated after the Calibration Sphere had been moved, but not redefined with the MasterProbe.
Link to comment
Share on other sites

Please sign in to view this quote.

Well, I am the only one using the CMM, so either I qualified a probe without a qualified sphere, OR, it happened in an unknown/unexplained way - it is a bad news either way 🙂 but I am glad I learned something new today.
Link to comment
Share on other sites

 Share

×
×
  • Create New...