All Activity

I think ellipse is the least usable element. What you want to achieve? Sketch maybe? 🙂

Yea, after more thought, that is the conclusion I have come up with as well.

PiWeb 2025, I can't figure out why it does not show my probe radius in the PlotProtocol page.

Can we do this, intersect an Ellipse to a Cylinder?

@DWC This is what I'm leaning towards. Especially since it seems to happen on all teeth scans, which are by definition, larger than the 4 tooth scans. My next step is to carefully track exactly when GearPro finishes generating the report, haven't had time to test it today. I should be able to check the report status window to narrow down exactly where Calypso is in the report phase. I'm using the function below in Calypso: I've looked thru the GhostScript documentation, but without knowing exactly how Zeiss implements it I can't see exactly what the triggers are or when it times out. I was hoping for an easy button where someone else had seen the exact same thing.

@RPSmetrologyTech It took me awhile to be able to articulate to bosses and peers what the base alignment was exactly, and how it doesn't have anything to do with DRFs. For any new guys reading this - the base alignment defines how the part is oriented and positioned in space for stylus navigation by establishing a coordinate system. The end. 🙂

2 years ago during my first year as a programmer, I was under the impression that the DRF should be the BA if possible. Just recently I switched my method to relying on secondary alignments rather than all the characteristics under the base alignment.

Thank you, the typo of 50 instead of 5 was indeed the issue. Thankfully it is only on that one feature.

Hey Everyone, Just need to double-check my work on the true position of a hole pattern to a single datum. Do these need to be measured as cylinders or is circle on a cylinder allowed? Here is my characteristic with redacted info. Here is a mockup drawing. Appreciate the help!

Ding ding ding. You outlier eliminated all of the points. lol.

I could work with you to show you some proof of concepts of an idea. Shoot me a DM.

@Chad Watton I have absolutely no experience with Ghostscript, GearPro, or even gears. I did a quick a quick search on here about how people are using Ghostscipt, and it seems some people are using it with PCM. This part stood out to me - In most scripts (the good ones at least), there is error handling to include timeout logic for a function, and if the function cannot be completed within a certain amount of time, it will silently fail and continue processing the script. This prevents a script from hanging. Perhaps there is a timeout setting being tripped where GearPro PDF is not fully written before stitching starts? Perhaps adding a delay in the script or with PCM will fix the timing issue. This is interesting, and may point to temp files not being cleaned up correctly (or maybe Ghostscript runs out of memory or hits temp file limits?) When the stitching fails, is the GearPro PDF already generated? Or does it generate seconds AFTER Ghostscript completes the stitching?

Yes, I do have PCM but i know very little on how to use it. I have done some formulas within the calypso (i know that is very basic PCM) but I know there is a pcm license here.

I would think Calypso would default to empty Primary, Secondary and Tertiary windows and not auto-populate with Base Alignment. How many users think "I created my BA using the DRF, I must be all set."

@Jordan McCann This seems....high. Change it to 5 and see if it calculates.

Above are the Screenshots you requested

Perhaps the Intersection is translating along the Cylinder axis (until it intersects the Plane); whereas, the projection is just using the origin (or centerpoint) of the Cylinder and (looking for a different word besides "projecting" 😂)...projecting it directly to the Plane

I mean, there are certainly ways around that. Do you have access to PCM?

I believe the order of the projection is correct, but the results for position are quite different. If I choose the cylinder for the position then choose projected tolerance zone, and choose the plane for Feature 1, and set offset and reference length to 0, I get very similar results, less than 10 microns, compared to using the intersection of cylinder to plane, so I believe the intersection is the correct way to do it. I just was curious to why the cylinder projection is so much different.

This is definitely the initial plan. But there will also be the need to a full program test as we have found a couple instances where the programs are recalling alignments and creating new ones and the new ones weren't fully following the part. So unfortunately, I think a full test will need to happen too. Between the way the parts are fixtured and alignments built on alignments built on alignments there is quite a bit to check. Both program repeatability as well as fixturing. Which is one of the reasons i am trying to push for fixtures as the way of fixturing.... some of them are quite impressive.

Consider narrowing the scope (at least initially) to critical characteristics & characteristics that have produced "bad results" (until after requalification). Hopefully, this can reduce your runtime enough to run a short repeatability study. It doesn't necessarily need to be a full Gage R&R.

i believe it was around 5. That is one of the things I need to do. From the sounds of it, the programs have been proven out to not crash but never really fully verified for repeatability. There has been much talk about doing this. No one has ever had the time. And with the programs being so long, it takes forever to do also. Most of the programs run between 30 mins and 1.5 hours. So to do the repeatability it will be a couple day thing on some of these parts. But it needs to be done.

I can't, Unless there is a way to allow the operators to select individual characteristics to re-run in Autorun, that isn't going to be an option. I would love to be able to utilize autorun but the programs are too long to rerun a full program to recheck one or two characteristics. They have one program that is an hour and a half long.

...how many is "several" is there any available CMM time to remeasure the same part, for a repeatability study?