Jump to content

Why this Element geometry alignment doesn't work?


---
 Share

Recommended Posts

Hi,

I want to know if a am wrong with something.. why this alignment does not work? With the first cylinder i can block 2 translation (X and Y) and 2 rotation (around X and Y), with the 2nd cylinder i block the rotation around the Z axis and with the plane I block the Z translation, all the 6 freedom degrees. Why it doesn't work? Thank you very much.

Line-Line-Plane.png

Link to comment
Share on other sites

Hi,

 

my guess would be that either in nominal or actual your cylinder are not perfect parallel. To compute the chosen type of datum system (or alignment) this is a needed requirement.

Unfortunately, the software cannot established a parallel secondary datum feature from your cylinder line (and this is all information that can be taken into account due to the fact that you have chose "Geometry" as computation method) due to fact that the rotation point to make the secondary cylinder parallel to the first one matters (i.e. you will get different results, if you rotate the secondary one around the base or the top or the middle point).

Now you have the following possibilities

  • You solve the problem for you own, i.e. you have define a method to established a parallel cylinder from the cylinder data.
  • Or you switch to ISO computation mode, that will solve this problem using ISO 1101 conform fitting methods. 

Hope this helps

Christoph Schult

  

Link to comment
Share on other sites

Ok, I tried, but the result is not what I want. I tried another alignment  just to verify how it works, but I don't understand. With the alignment in the first image, the result is the second image and it is not the result I hope like the 3rd image (with a simple local best fit by selecting like the 4th image)3_Planes.thumb.png.bca30d16e72cdc32f0f04bbff321993e.png

Geometry_3D_Coomparison.png

3D_Comparison_Local_Best_FIT.png

Local_Best_FIT_Selection.png

Link to comment
Share on other sites

So, I can never use geometrical alignment because the result is NEVER what I want and I prefer use the RPS or local best fit which I find paradoxically more precise.

Link to comment
Share on other sites

Please sign in to view this quote.

Maybe, for your application this is the wrong alignment. If you want to have a good common adjustment above all areas then local best-fit is indeed the better alignment method. Instead if you try to simulate an assembly then an alignment by geometric elements gives you the tools to evaluate such function requirements.

Please sign in to view this quote.

It's quite simple. This alignment is only a shortcut for the following construction (of course a little bit simplified but the idea should be visible).

  1. Create a nominal datum system using your nominal elements. Due to the fact that you have specify 3 elements and some special modifier are stripped from the dialog this should result in a datum system with no remaining degree of freedom. From this one a nominal coordinate system can easily determinate.
  2. Create a actual datum system using your actual elements and determinate the actual coordinate system.
  3. Compute a transformation starting from the actual coordinate system to nominal one
  4. Apply this transformation to your actual mesh

To clarify, this uses the logic define in ISO 5459 (the GPS/GD&T-datum system norm) and expect that all elements a parallel or perpendicular to each other. If this is not the case a internal fitting is used to find some elements that fulfill this requirement.

This allows you to see the mesh in the alignment that is defined from your datum system on your technical drawing. Due to the fact that datum systems should be a representation of a certain function of your part is often useful to have the same alignment for reporting or additional inspection.

Please sign in to view this quote.

Depending on your task maybe alignment by geometric elements is the wrong alignment method. But I hope the computation and application for this alignment are now more understandable.

Best regards

Christoph Schult

Link to comment
Share on other sites

I would agree with you if the two results (second and third image) were similar to each other, but the difference is really big, about 0.10 mm on plane 3. I think GOM on alignments should improve because plastic is not mathematics and some degree of "mediation" should be there. This is my opinion.

Link to comment
Share on other sites

  • 2 months later...

Sorry, I can answer here much easy in German:

Ebene 1 auf die große Planfläche, ( Z)

einen Zylinder auf den großen Innen - Durchmesser in Bezug auf die Ebene 1 (Y) ( so wird eine Schräglage des Zylinders verhindert ) 

Ebene 2 auf die senkrechte ( grüne ) Fläche im unteren Teil im 3 Bild ( X )

Alle Körper als "Reffenzierte Konstruktion" über I-Inspect wählen

Dann lässt sich das als Ausrichtung über Geometrieelemente definieren - danach kann man es ja noch in Y Richtung drehen, um die beiden Bohrungen über das Bildschirmraster auszumitteln.

 

 

Link to comment
Share on other sites

 Share

×
×
  • Create New...