PCDMIS 2020 R2 True Position

[Vikingroovin]

I have an old program that was proven out and worked fine up until I installed the 2020 R2 software now the True position is way off even if I run known good parts. anyone else have this issue? did they change something in the software?

[MetsoMetrology]

Which version did you upgrade from? I don't recall having any TP issues going from 2019 to 2020.

[MetsoMetrology]

One thing I remembered is that in 2020 you can no longer edit theoretical values in the TP output window, so you have to make sure all your theos are correct in the edit window.

[JacobCheverie]

There is a new Geometric Tolerance command that replaces XactMeasure. Please post the relevant portion of your code (features, alignment(s), datums, dimensioning, etc...) and we may be able to find the discrepancy.

[DAN_M]

There have been BIG changes to the software. The latest revision of Y14.5M has redefined/changed the default calculation math for some important stuff & the software has evolved along with it to suit these industry requirements.

Hexagon has posted some really good videos with their senior level people explaining the changes, really helped me out a lot. Highly suggest you start looking into this stuff.

The above being said...you're gonna have to post that code or we can't help you my friend. There are way too many potential things that could be messing you up.

[neil.challinor]

This is where to find the link to the webinar recording https://www.pcdmisforum.com/forum/pc...419#post481419
There is also a wealth of information in the help files here : https://docs.hexagonmi.com/pcdmis/20...rol_Frames.htm

[jjewell]

Dan,

You say big changes to the saftware, but OP is talking solely about TP...its calculated on the most basic math and its never going to change...its simple trig really....he needs to post code so we can see where the problem lies...I wish people would learn that they cannot come here and say, " What is wrong or going on"....thats like saying, "why doesn't my car start"...could be dozens of things, but without being able to eliminate some variables, we are just throwing darts in the dark...( Which I have done when I was younger and ended up with one sticking out of my forehead after !! LOL )

[DAN_M]

jjewell

Hi Jim, big changes have occurred to the software my friend!

For years, the industry has been struggling to come up a "UNIVERSAL" best fit math type when having a software "calculate" what a feature looks like. This is in effort to improve data correlation from supplier to supplier/across the board.

The release of Y14.5M-2019 settles this discussion and instructs the industry to STOP using LESTSQU as the DEFAULT best fit math type and use something called "Constrained L2" as the DEFAULT. When you use "XACT" and create a feature control frame, PC DMIS will now default to this "Constrained L2" math.

Constrained L2 (very similar to ISO's Chebyshev) eloquently eliminates rock in the software's calculations and best duplicates what you'd get if you had the datum resting on a surface plate. Even with really waffly/deformed "chip" looking parts (that are supposed to be flat) this calculation minimizes distortion in a highly repeatable way and will yield results that correlate to a much higher degree with other methods.

I have done some testing myself. Measured a simple T.P. using 2019 then 2020R2. Even though you're totally correct and the formula for T.P. remains the same, the PLANE for the alignment gets calculated differently AND the circles that are being measured look slightly different to the software so the results came out slightly different (we're talking 0.0005 difference on T.P. and 0.0005 difference on dia size...but for me thats a good amount of my tolerance).

Hexagon has published some webinars about this stuff, suggest you give it a look! Hope this info was helpful.

[Schlag]

OCD flaring up ....... no such thing as TRUE POSITION ........ just POSITION..... '

[DAN_M]

if the CMM says its good, its a true position. if it rejects it, its a FALSE position =(

[Schlag]

Cant find that in any of my standards and I don't remember that question in my Y14.5 certification test but I do agree 100% so I guess i would have answered it wrong....