All Activity

@Maciej B, Autorun is your best option per the workflow you mentioned, including the automation cell + barcode scanner. It's not impossible to run two parts / two CAD models in the same measurement plan and parse it to two reports, but you would be using a complicated workaround to do what Autorun already accomplishes.

Well, you can not run one program and get two protocols. But what is with a robot? If it can place parts, then run code, which will run a program, then why not let robot place two parts, then scan and run part1, then scan and run part2? Both would have another pallet program with defined strategy, but same program.

Maybe I didn’t explain myself well. The machine and robot with a separate panel don’t matter here. I just wanted to explain why I need this program. The robot works with the machine, and for now, everything is fine 🙂 Until now, we only used the program for one part. Now we want to measure multiple parts in one setup. You mentioned a miniplan, which could be a good idea. My question is: can I make two seperate reports? One for the main base part (first part) and one for the miniplan (second part). I know the characteristic will have different name, but that’s okay.

I am no expert in codes for Calypso, but there are some here. What i know they are using bar code and search field to find correct program to run. If you have PCM licence, then this can be doable. For program - normally you can not have two protocols in one program using same characteristics. You can have setup autorun for having twice the same program, only in run parameters you can use miniplans, or using different strategy. Miniplans will use doubled characteristics ( you have to copy them, rename them and use different features for different probe / safety plane ) Using different strategy is using same characteristics and features, you have to only edit all feature's strategy for different probe / safety plane. Using robots and code reader is outside of my knowledge and wasn't mentioned in initial post. I would advise you to modify title to bring someone with knowledge.

@Martin Jánský thanks for answer. This program will be run by CNC operators from production. The machine works together with a robot that mounts a previously prepared vise onto the machine table. The operator does not even need to approach the measuring machine. From the robot interface, they scan a QR code with the drawing number (the measurement plan has the same name), and the program runs automatically. So in this case, the auto-run function probably does not make sense, right? Can I have a standard measurement program with two CAD models in it and still get two separate measurement reports?

Hi, we have a small article with @Jochen Heller's post as a picture here on these values: https://techguide.zeiss.com/en/zeiss-inspect-2026/article/introduction_to_cp_pp_cpk_ppk_values.html

In autorun you can have two programs in same panel. Then with "CTRL" on keyboard you can select more programs to run - you fill parameters ( header ) and it will perform two measurements on one go. But this is not different from usual work. Unsless you would like to have both parts in one program ( same part, but different probes ) that can be done too.

Hi Jochen! Thank you for your response. So does Cp have something like a filter that removes the outside measurements? Or does it accept only a certain percentage out of tolerance? because in the example I posted I have a correct Cp, but I have something like 5 measurement out of tolerance. I don't understand this thing.. thank you very much

Thanks for the document, Keith. Much appreciated. 🙂 At some point in the past someone had posted settings specifically for the evaluation of curves, both open and closed. Do you have any insight on that?

Hi Guys, I would like to prepare multiple measurement plans for the two small parts below. I have never measured multiple parts in a single setup before. Could you give me some guidance on where to start? It is important for me to get two separate reports after measurement and the programs cannot be identical, because, one will require an approach from -X and the other from +X

Filters & Outliers.pdf

Hi Marco, this behaviour is (can be) "normal", due to what cp and cpk evaluates. What are Cp and Cpk? Cp, or Process Capability Index: Measures the potential capability of a process under the assumption of perfect centering. Cpk, or Process Capability Ratio: Reflects on the actual process mean, presenting a more realistic image of the process capability.

Hi Jeff, Can you expound on this? I'm not sure I understand how this creates individual features instead of nested features when using pattern. That would be great, as this has always been somewhat of a pain to do. Currently I use alignments, copy, and keep position to create individual features of a pattern.

Hi Jordan, with an automated system like this, you can use the Kiosk mode: ZEISS Quality Tech Guide It's very customizable with python, but the basic setup can be done by dialogues in the software. I would still recommend to reach out to your local support for the first setup. There is a setting available, which will create a trend project of the used project template in the background. Measuring always takes place in an empty project template. Saving individual projects or just the .stl for example is also possible. Nanno

The bigger question for me is why and when to filter. I never filter stuff unless I have a reason to after looking at the results and/or form of the feature. I like raw data in all my features and will only filter on the characteristic side if I do filter. Automatically filtering everything has bitten me more than once, especially on tight tolerance bores and curves.

Optical comparators work great for this. Zeiss also makes something called FixAssist that aligns things for you.

I think a circumstance you might need to recalibrate after a shutdown is if you use the machine origin/coordinates for something in your inspection process. When the machine homes I think it resets the machine origin. It might be an inconsequential amount, but I suppose it could have an effect on accuracy. I don't know this for a fact. Just basing this on other machines that I've used.

Hi Nanno, I’m taking advantage of this thread to ask for a clarification: which standard or regulation is used as a reference for the calculation of Cp and Cpk? Because in the case shown in the image, even though I have some values outside the tolerance limits, the Cp is greater than 1.33, while the Cpk is not acceptable.

Hi Jayveous, you need to create/edit a measuring setupZEISS Quality Tech Guide and possibly move the part to the new fixture ZEISS Quality Tech Guide. Exporting it from your program before the import into Zeiss Inspect in the same coordinate system as before should also work. It's part of the Scanbox training also. If there are any doubts please contact your local support, a wrong measuring setup can lead to hardware crashes. Nanno

When Calypso changes to a different feature I assume there is some internal flag we can catch at the time the feature name is activated for measuring. We can see the next feature to be measured at the top of the Calypso screen just before the measurement starts.

Get one of these, they are inexpensive and can aid in setting any simple or compound angle you need.

A machinist protractor or an optical comparator also work well.

Here's what you do to find the exact angle. Start touching the shaft of the probe against the reference sphere. Make sure to get points high and low on the shaft and go all the way around the reference sphere. Your goal is to create a cylinder feature inside of Calypso. Your A1 A2 angles on the feature are the angle the probe is angled. You then enter these values into the appropriate spots in Calypso when you qualify it.

I think only option is using python script. That's why i am working on our own filling table, which will suit more our needs of evaluation selection.

Lots of questions here...thank you in advance! When building a stylus with movable parts, what is the trick to getting the movable parts at the correct angles? And how critical is it? I am building the stylus pictured. It can be rotated 360° (blue arrow) and the knuckle can bend back and forth as well (green arrow). Let's say that when the RDS is at 0°/0° I want the stylus to face forwards towards me and be lifted up 45° towards me. How do I ensure that it is exactly centered / forward and exactly 45° up? What if it is not exact? How far away from that can I be before it creates an issue? What angle does the reference sphere have to be at in order to qualify it without a collision? So far our reference sphere has always just been straight up. If I have to change the angle of the reference sphere, how do I ensure it is at the correct angle? When I go to qualify it, if it does the thing where it guesses what angle the probe is at, but it's not what I intended, do I change the values to match what I intended, or do I adjust the stylus and try again until it matches? I believe once it is set up, I could set it up as different probes with different A angles, correct? But the B angle would always be 0° and the probe would always be 45°. Is that correct? What if in the future I wanted to use this same set up but have the knuckle point the probe at 30° rather than 45°. I imagine I would need to create a whole new stylus system, correct? But wouldn't that require a separate adapter plate so that the CMM would recognize whether it was the 30° setup or the 45° setup? Would it make more sense to leave this one the same and buy an identical setup and make a whole new configuration?