All Activity

I made a post about this a while ago. I'm seeing more and more designers use dynamic profile. This will become a big deal sooner rather than later.

Is there really still no dynamic profile modifier?

Hello Nanno!, Thank you very much for your help with my question. The part was scanned with a Hexagon Romer Absolute Arm 85, and the scan software was PolyWorks. I only use ZEISS INSPECT 2026 for the inspection. Since the scans do not come from a ZEISS system, I understand that combining measurement series is not an option for me. I will try to merge the two scans in PolyWorks first and then import the single mesh into ZEISS INSPECT. If that does not work, I will test the trial of the Professional version. Best regards, Anu

It used to be if your stylus failed qualification the results element would return the value of the previous sigma information. Is it still that way?

Hi Anu, if you have two separate meshes like STL, it is not possible to combine them with the free version of Inspect. The professional version supports multipart projects. Here you could align the meshes using the markers and other features, then combine them after stitching (ZEISS Quality Tech Guide) or some other mesh processing. You might try the trial version of professional first for this. Do you scan the parts with Inspect? Then you could combine the measurement series in one project, which will also result in a smoother result, because the measurement series are polygonized together. (ZEISS Quality Tech Guide) The measurements are also transformed accurately by reference points with this method. Nanno

Hi Robert, i could reproduce the error by switching to different inspections with an open simulation, without even connecting to another CMM via the settings, i will create a BUG for development! Thanks for the help! Best Regards Marcel

Hi Sven, does this exception appear when you open the Basesystem? This issue is reported in 8.0 and is solved for the upcoming 8.0.20 Servicepack, it might be that the same problem is within 8.4 and should be fixed with the next Patch Version. Best Regards Marcel

Hello, I am using ZEISS INSPECT 2026 (free version) and need help merging two scans of one part into a single mesh. The part was too long to scan in one go, so I scanned it in two sections and have two STL files. The scans overlap in the middle of the part. In the overlap I placed four magnetic letter markers (just physical markings), one on each side, and they are visible in both STL files. Goal: one complete, accurate mesh of the whole part.

Hi Goce, How to best fit those points to the CAD after importing? Is there a way? What is the proper work process? the normal Local Best-Fit alignment works, if you select points of your point cloud like you would on a mesh, for example right click -> select all points. The point cloud might need to be aligned by an inital alignment like 3-2-1 first to get a realistic search distance in the best-fit. Also how can I re-align the CAD to different position in space? There is a catetogy "single element transformation" under operations -> alignment, or you chose another coordinate sytem during CAD import. The single element transformations are not parametric like an alignment from mesh to cad and should be used at first in the project because it will later break dependencies of elements. Example "Set matrix": ZEISS Quality Tech Guide Is there a way to polygonise this type of point cloud? There is a function for this: Polygonize Point Cloud If the point cloud is polygonized the workflow is like with a mesh (STL, polygonized from ATOS, ...) If you work with Airfoil inspections, there is an App with CMM workspace available, which sets up CMM data for inspection with Airfoil Inspection: CMM data preparation Nanno

Hi, One option to get this immediately would be to use linear sizes for the inspection of cylinders or other features of size. Either with normal labels or with the 3D labels. It has also the benefit that your results are in conformance with the international ISO-GPS or the american ASME standards. Then it simply does not matter anymore what is set at the fitting element as the software computes everything automatically for you in accordance to the selected standards and the needed association (fitting) method. The following methods are available in linear sizes (named either according to ISO 14405-1, if selected, or based on ASME Y14.5)

A little bit of information before my question, I am using a spectrum CMM with a passive Vast XXT with rotating head, my reference sphere is at 180 for angle 0 for tilt and it positioned as close to center as I can get. I am also running calypso 2019. I am attempting to set up a program to automatically qualify my probes using the probing system qualifications characteristic, I have to manually put on and take off my master probe to get reference sphere location due to rack space limitations. Then I run the program, it will do my first probe a 3mmx50mm no problem but then it will attempt to do all my other probes, and it will hit a limit switch usually on my X axis sometimes Z. I'm stuck on what I need to change or if there is another way of doing this.

It does happen that it's needed. We have a customer from chip making industry that wants 5 micron runouts on entry radius - thru hole - exit radius. Its needed for guiding a beam thru the device.

wish zeiss would build something in natively, I mean what year is it ? sheesh

I found only this via python

I'd probably also suggest to use Curve results points as well.

Is there an easier way? Unfortunately not at this moment.

We've got the FixAssist. It's amazing how often it can't help.

Hey Marco, Glad you got it working! you should be able to click the number in the top RH corner and either use the mousewheel or change the value to whatever you want. If you want to customize it beyond that, I Included a "Prompt_for_ai.md" - just download / extract the codebase from github, give Codex/Claude code access to that folder, copy the prompt and follow it up by "Id like to change the color legend to ........." give codex/claude some example pictures of what you're looking for. let us know how it goes!

Is there a way to show the fitting method used for a fitting cylinder? I tried creating a custom label but I couldn't find a keyword that exposed the fit method. Let me know if you have a solution for this.

I am not sure what you are getting at. Are you using a result element to calculate something or visually acknowledging the Sigma and Radius values? FYI Many users use a CMM program that reports the Qualifications of probes and setting tolerances on sigma values. I do this, and I chart the sigmas over time to track the condition of each stylus tip using Excel, simple Bell Curve for normal distribution.

As usual, solid advice. I thought the part resembled a Swagelok VCR Gland perhaps something like a "SS-4-VCR-P" (Or A‑Lok, Tylok, etc), I have literally made about half a million of these in my machining career, I do not recall a runout requirement before, but I do know that radius is a critical sealing surface (If it is a VCR Gland).

Are you sure that the templates that PiWeb is using are installed in the correct location and that any paths that PiWeb is writing information to are accessible? Maybe it's trying to write the PDF file to a specific location, and it can't because it's write-protected and/or the location doesn't exist?

Measure "Circle on a Torus" in at least 6 locations (60 degree separation). Since you are using an XXT the initial point pickup will need a reverse over-scan for optimal accuracy (a second circle path that starts where the first ended and go back over the scan to pick up lost point data from stylus normalizing). Depending on size (hard to scale from image), it may be beneficial to constrain the size for optimal fitting. Do this in 6 places minimum with even distribution/spacing. Create a Point > Recall > Select 1 circle. (creates center point) Repeat for each circle scan path you created. Create a circle > Recall points > Select points you just created as center points for torus scans. Apply the runout control to the circle construction you've just created. There are ways you could do this with patterns and indexing as well but for ease of direction, I provided the generic manual steps. Alternatively you can scan around the torus D1 in 4-5 places (full diameter scans at different points on torus) and then create intersection planes to create a circle from feature points in as many places as you prefer without adding cycle time. Each approach has its benefits. One question comes to mind though: Are you sure the torus requires a runout check? This is not standard in most scenarios on a turned component as the end radii are usually clearancing and not sealing surfaces.

I skipped 2023, and went with 2024, as it's supposed to be a true 64 bit software, for more ram access. I had intended on going with 2025, but now not so sure. The simulation I did on 2027 went smoothly, so I have some hope. I did mess around some with 2025, for some reason I had to recreate all my machines from scratch, and put a number of some kind, at the end of the machine name, or I would get some weird error.

???