All Activity

And I did that in an area that was flatter. The problem is that I need actual points and not a plane in some areas. I could create a plane that way, but I would still need to extract the point from that plane to get the point with its IJK. Maybe if I tell it I am making a plane, select one point and then explicitly change the feature type to a point ... when I get a free seat I will try this.

Sounds like you might have some good insights. I originally tried an rps alignment. I came to the conclusion that one of the key regions I am taking data in is a problem. The customer-defined alignment points are mostly on the edges of thin walls, maybe 0.08" thick so the manual alignment has to be pretty tight and the part has to be made to tight tolerances for this to work. But there are some key points that are out on a curved surface that have a pretty big affect on the alignment and can cause the alignment to rotate if I don't absolutely nail the initial placement of those points. (They are a floating datum target.) So I tried to create a start alignment that is more practical. It is telling me that my basic alignment and start alignment are something like an inch off from each other. I am not certain how it figures that out so I want to ignore that. When it goes into CNC mode to take the base alignment, the first four points that it takes are all at one end of the part. They are almost enough to fully lock down the alignment, but not quite. They are enough to know if the alignment is grossly out. All of those points measure just fine. They are right where they are supposed to be. But when the CMM goes to the far end it is just off somehow. it might be off by an inch. It is hard to say as the machine errors out. I used "define points" to create most of this stuff, but people are suggesting that I use space points so I will go that route next. But the more I think about it, I don't think that is enough to account for the issue.

Ok, so you can create new plane feature and now you can click on model which will make this "not defined yet" plane position and size adjust by your clicked points. It's not modifying already created planes, but for new it's doing it. Same for cylinders, spheres and so on.

Follow up: I got this picture from ZEISS Portal. They make a distinction between the actual normal and the nominal normal. They then draw two points. It looks like they label the plane point as using the actual normal. I would assume that they get that from the force direction applied to the probe. In the text explanation they don't distinguish which of those two normals they are using. They also don't give the second point a name. So which normal vector do they use? Further, if I am using a space point, are they really going to project it out into space like that? Will the software really just screw with the measurement and give you no reasonable answer for what trying to measure the point?

@Benjamin Bellville, hang in there. Your problem is real, but it's solvable. My first CMM gig was measuring large machinery that sounds similar to your current challenge. It's entirely possible to measure parts that are not orthogonally perpendicular to the CMM's coordinate system, including parts without planes and other prismatics. What Peter and Martin suggested regarding space points vs CAD points should solve your point placement issue. There are several different types of points in Calypso. The CAD point option tends to snap points to the CAD entities as though they're magnetic. Space points should work better here, even though the names sound counterintuitive. Regarding the alignment challenge, there are two options I would consider: 1. Non-standard alignment options such as best-fitting / RPS (reference point system). You can use a cluster of points to define your workpiece and constrain degrees of freedom. This works well on parts without prismatic features like planes, cylinders, etc.. 2. Standard alignment with a defined rotational angle from the "Special" menu. If you can create a standard alignment but just need to define that it will be clocked at a known angle, you can use the special button in the alignment window to rotate the alignment by this angle.

I really don't know the difference between a point and a surface point. As for extracting planes from models, I cannot. Most of the surfaces are not recognized as a plane by the software. I can click on multiple points on the CAD model and iterate through

The active probe is also considered when clicking points on a model. (If your probe is angled, and constrained to the coordinate system of the CMM, your points obviously will have to be too!)

You could also do a 3d best fit alignment and loop the alignment

Thank you. I will try this as soon as I get a free seat.

Sorry, I cannot share pictures or really much more information.

I'm not entirely sure either. As you can see, the graphics I provided is below the nominal and it shows slightly to the right of the midline on the scale.

Never looked that hard at the scale, usually looking at deviation. Pulled up a report and noticed the first dimension graphic appears centered around the means even though it is above the means dimensionally. Maybe scaling because of the large tolerance on the dimension? Again, never paid much attention to the graphic, I believe it is intended as a graphical aid, but you piqued my curiosity.

Are you clicking points on a CAD model with the "Define points" option? This defaults to the Touch Point evaluation method (snaps vector to nearest coordinate axis) and rounds your nominals to the "Round-off measure" setting under Extras->Settings->Measurement... Instead try the "Define a Space Point" option. This will maintain the nominals from the CAD model and set the evaluation method to Space Point, which also uses the correct vector from the CAD.

@Richard Shomaker I don't have FFS, but I assume you can use formula in recall feature points. I would use PCM to create a list with desired elements and recall that list using formula. You have to change a loop index in every element, but you can come up with some "automatic index extract" from name of element or something and you just copy paste that element x115 😄 nameOfElement = "Point" noOfElements = 4 noOfSections = 3 loop = 1 recallElementsList = "list(" for i = 1 to noOfElements for j = 1 to noOfSections recallElementsList = recallElementsList + qm() + nameOfElement + "_" + i + "_" + j + "(" + loop + ")" + qm() + "," next j next i recallElementsList = mid(recallElementsList,1,(len(recallElementsList) - 1))) recallElementsList = recallElementsList + ")" //display(recallElementsList) recallElementsList = getParameterNamed(recallElementsList)

Are you clicking only point or surface point? You can also extract plane from model and do points on it - so vectors will be fine

Are you able to share an image of the CAD or part?

I suspect dirt on contact area inside RDS, but i don't own RDS so i can not tell how it looks inside. This problem is only in this position or is it occuring in other configurations?

I did, but there was no improvement.

Did you try restarting machine?

We are having an issue with the probe head on our ZEISS CONTURA CMM, as shown in the video. Does anyone know how to resolve this issue? We urgently need to take measurements. Thank you!! 1790819277603-1342595881688479684-g2617570072669844278_VY7PIJil.mp4

I apologize , it´s confusing, i was questioning if someone used the projected tolerance zone.

Hi Marek, sorry for the late response. Yes, the deformation and clamping force are calculated using the deformation model. A more sophisticated quadratic deformation model used in Virtual Clamping will deliver more realistic results for example. If you use De-Warp I recommend watching the eLearning. It also covers the deformation model. Nanno

Hello Akos, what is circled in the right screenshot is the simultaneous requirement, which was added with Inspect 2026. A positional check with spherical tolerance zone is possible if the datum system supports it. Nanno

Good morning everyone, FYI: development is currently working on this to be more user friendly, i dont know about the planned timeline but there will be a change about this matter with the upcoming CALYPSO Versions. Best Regards Marcel

Sounds like calypso is just super buggy with this and badly programmed. Thank you all for your advice. I ended up starting a clean file and rewriting the program from scratch. I was not that hard of a program and only took about 6 hours.