erroneous readings in 3D measuring

In the problematic files are all the measurements out by the same factor, or is it only some individual measurements that aren't working?

New to this forum so looks like if I ACCEPT your answer, then I cannot reply in this same thread so I will write this message and then submit to see what happens.

It might be the same factor for perpindiculars. I think hole diameters are measuring OK on the problematic file(s). I will verify these subjects on Monday 2010-02-22 at the workplace.

Thanks for your post USVAR

We often see the entire file scaling wrong (due to a mistake in the import options dialog, or a file not having a scale factor in the header - common issue when using U3D as an intermediate step). If the same object measures OK in some aspects but not in others then it's a bug - not unexpectedly so, but if it is, there won't be a fix until the engineers track it down and roll something into a dot release.

I don't suppose you have (or can mock up) a file with the problem that you can post a link to?

UVSAR
I very much appreciate your input but providing a link would be a no-no since it is proprietary and because of the fact that I have not been directed or charged with what I am doing. I am a bit of a maverick delving into solving problems that are not actually part of my sandbox.

Most likely tommorrow (Monday here) I will have some time to compare several of these pdf files to get a better grip on what is consitent or otherwise between files.

I do not recall seeing a scale factor in the header. Would that be in the header or top of the 3D pdf window? There is a thingy in the lower right corner of the pdf 3D window that says 1mm or something similar even after I go into options and change to inch measurements . Also in that same little lower right hand box it shows Radius even when I change the option to Diameter. And yet when I measure a hole, it gives me either radius or dia correclty depending upon which I have designated in the options.

What is U3D? U for Unigraphics?

U3D is "universal 3D format" - one of the intermediate file types people often use to export data into Acrobat when it won't read the native files, and one of the ways Acrobat stores mesh data internally.

Most 3D/CAD file formats have a variable within their header that says what one internal unit equals in real-world units. Acrobat displays this on the bottom line of the 3D measurement popup (which is why it doesn't change when you switch units) and will use it to set the overall scale unless you manually override it on the "Import options" dialog when you create the 3D annotation. Once imported, you can't edit the value without re-importing the model, but you can turn off auto-scaling in Acrobat "General Preferences / 3D Measurement" by unticking the first checkbox.

We often see the scale factor going astray when people bounce through an intermediate file format - for example if you export from 3DS Max into an OBJ file, then import that into Acrobat.

Having said that, a bad or missing scale factor will only ever result in the entire measurement system being out. It won't cause some types of measurement to be correct and others not, as Acrobat doesn't support the concept of non-cube space (different scales for X,Y and Z); so if that's what you're seeing, it's a bug.

UVSAR,
Thank you. This is very helpfull.
I did hve time today to measure 5 different parts and their pdf 3D models and compare. I learned several things.

Two the models based on your last comment have bug(s).

measured on part/measured on pdf 3D model Scaling factor P/N
1.993/0.0784 = 25.42 220-1044
0.157/0.0064 = 24.51
0.490/0.0169 = 25.78 also had readings of 0.019 vs 0.0169
0.993/0.0786 = 25.36 220-1043
0.157/0.0057 = 27.54
This was using faces. However except for the 0.490 dimension, these are not planer faces. But using linear edges on another part without bugs and with non-planer surfaces I get good measurements whereas on these to buggy guys linear edges give me other erroneous values.

Also on these two models, when in modes for "3D snap to radial edges" and "3D radial dimension", It allows snap to two different points (where there are none) and gives an erroneous dia of circle based on a chord between the two points.

On the other two parts (220-1008 & 270-1605 for my reference) that do not have bugs, where measuring between two no-planer surfaces and using snap to linear edges, I had to zoom way in to get a good reading. Apparently when zoomed in tight, it takes a tangent to the curve and treats it like a short linear line?In Preferences for number of significant digits it is pretty crazy but workable.

If I tell it either 4 or 5 significant figures, it gives me 3 sig # or 2 decimal places.
If I tell it 6 significant figures, it gives me 7 sig # or 6 decimal places. (This on model 210-4111)

On model 220-1008
It gives me 4 decimal places if I tell it 4,5, 0r 6 sig #'s.

Therefore I just tell it 6 sig #'s for everthing because we do need 4 decimal places on some holes.

You wrote: "Once imported, you can't edit the value without re-importing the model, but you can turn off auto-scaling in Acrobat "General Preferences / 3D Measurement" by unticking the first checkbox.

What does this do to turn off auto-scaling? Can you please elucidate?