All Activity

I'm not sure you're overall part size, but its all gonna come down to data quality and resolution, I'm not as familiar with the tscan hawk, but i'd recommend an ATOS Q 12m ~100mv. You dont want garbage in or you're gonna get garbage out. construct a nominal surface--> MP Fitting element and reduce the normal deviation significantly. use that actual selection in a surface comparison as opposed to a gdnt profile, tweak the opening angle and normal deviation. have you looked into a zeiss cmm with an active scanning sensor/probe? those things can traverse over some pretty large deviations...

If you would like to use a singular point, you could use a an "offset point". The offset would be the distance between the elements. Once you have all of the elements on the same plane, then you can select all four points and use a fitting plane.

How are you extracting the features out of the model? With define points?

Version of Calypso? File format of CAD? I have never face this - only if model was not correct, but models from SolidWorks ( in SAT and IGES ) are correct once i extract key features for base alignment and setup. Other extracted features are then ok.

When you use this option, the settings are applied to any characteristic that you create. In essence, the end result is the same. The only differences are 1) the features used for your base alignment do not get the benefit of having filters and outliers applied to them and 2) when looking the features, you might see differences in actual values vs. the results on the report. Unfortunately, when using this option, the Outlier Elimination settings for pre-filters don't change between circular and linear features. Maybe someone will post that it doesn't matter. In my 2nd Basic class in 2008, the instructor suggested just checking all 12 checkboxes and followed with that it was better use the default settings even though they may not be the correct settings. In other words, the wrong settings are better than no settings.

Capitalize the 'W' and it should work -- .printStringWithoutExponentF Could also try .asNumeralString and .decimalString

This happens with all CADs. there's no issue in alignment and references.

Lets say I set up my default filter/outlier settings, ex: Then I extract a feature. I notice that in the Evaluation settings, Filter/Outlier isn't toggled on by default, even with the default settings set up. My question: Do I still need to toggle them on to get to them to work? Or are they automatically on in the background because I set up the defaults, and I only need to toggle them on if I want different settings from the default?

Has anyone ever used the title name's extension? I found it in the PCM cookbook. It doesn't seem to accomplish what it says it does, or anything for that matter. But I'm also a novice, and not sure if I'm applying it right. every attempt I've made, the "d" is still there. Not a big deal at all, mostly just curious if I'm too stupid to apply it, or if it just doesn't work.

What Keith and Richard said. The likely difference between the reported flatness and the Min./Max. in the grid you referenced in PiWeb relates to the alignment that the Min./Max. are generated from. Flatness uses minimum feature as the standard evaluation method, which isolates the unrelated mating envelope (i.e. doesn't relate to a datum reference frame). This is a key advantage of digital systems such as a CMM. The feature can be isolated to itself to evaluate form characteristics such as flatness.

The Max is 0.0038, and the Min is 0.0000 = 0.0038 Flatness. It's location is just where it is respective to its alignment - flatness is irrelevant to location/orientation.

Thanks everyone this helped a lot. After reviewing my setup, the issue came down to a combination of weak geometry and measurement strategy: Datum C was built from very skinny slot planes, making the symmetry plane unstable That instability was amplified when projected back to the origin My measurement strategy (grid scan) wasn’t sufficient to properly capture rotation on that geometry At first I thought I had just programmed something wrong or was missing a setting, but it’s clear now the root cause is the geometry and how it’s being measured not just programming. This also explains why the center plane was inconsistent (especially in Y), while the center of the radii gave stable, repeatable results. On the next part in the family, the same datum is rotated ~43° from the base alignment, which likely makes this even more sensitive to the same issue. Appreciate the input everyone.

I would presume that the corner box values are simply "where are the min/max points located relative to the alignment", which is seperate from the Measured value that is meing evaluated to the Minimum Feature plane.

I ended up figuring it out! It was linked to the spatial feature that was in that alignment. The plane that was used for the x origin and spatial alignment was recalled from a 2d curve. I thought this was a bug, so I tried it in the latest version of 2025 with the same outcome. With that said I believe this is a bug, and I have opened a ticket with Zeiss. How I got it fixed was I had to re-add a plane and recall feature points to recreate the plane. Just recalling the points in the same plane did not work. I don't know how this got messed up or why it reacts like this, but that is how I worked around the issue.

Hi - I was questioned about these plots about why the min/max in the corner chart doesn't add up to the overall Flatness, and I wasn't sure. Should it? I never really paid attention until now. These were (6) diameter planes compiled into (1) plane. Note: I have a few of these scans and none come close regarding the expected ABS MIN+MAX= flatness. Thanks!

Hi Jens, any update on this ticket? Are you working on a solution for enhanced visualisation of the scan data during scanning?

1st reason can be not exact model. 2nd reason - did you first alignment from extracted features?

"Hi all, I'm consistently running into an issue in Calypso where extracted CAD nominals don't perfectly match the 2D drawing coordinates. For example, a circle dimensioned at exactly X=50.000 on the print extracts as X=49.9561 in Calypso. The circle's diameter extracts perfectly, the CAD geometry is correct, and I am using the proper extraction tools and alignment.

@Alaa Nwesry Not that I am aware. This task is much easier when handled outside the software with an automated script. Do you have any experience scripting?

How that part looks like? Curve in one plane or it goes in all three axis? I would bet that secondary alignment is somewhat rotated and moved. If you want to have less red points, then increase search distance on a curve. At least you will be able to see what is scanned.

I've had a few parts with this scenario. I used an RPS alignment. I measured the two pads at the same height as planes/recalled them into one plane/recalled the one plane as a point. Next, I measured the other two pads as planes and recalled them into points. I then plugged them into the RPS alignment along with my datum B and C features.

yes, it is possible to set limits on a global or per stylus system basis - https://portal.zeiss.com/knowledge-base?id=3366633

Bump. Still running into this issue 😞 Has anyone seen this before?

you can fill out this form for training sign up assistance! - https://forms.office.com/pages/responsepage.aspx?id=RCIEKFG71kyANHd2-jcD6CVrq-2wTgpPoBq-ltyu5ypUMTgwSDhNSlA4TDQ0TEJQRjRCREVCVDZOWS4u

Does Observer require a dashboard, third-party software? I pull the ProgramDurationRecord.xml file into excel and chart it, it's a PITA.