Creation of datum system from common datum. Offset parallel planes

[[Yt...]]

Hi 

Two options are available when creating a datum system using common datum planes, one of which is offset. 

Position of the resulting plane: middle or attached to the first plane.

Both options gave me fundamentally different results.

Is it possible to explain when and how the above options work?

Do all the planes need to be at the same height before they are set as a datum?

If the resulting plane is determined by the first plane in the list, what is the meaning and role of the other planes defined in the drawing?

I need help understanding the issue.

Thanks a lot

[[Ch...]]

To our experience if you have a common datum (consisting from multiple parallel planes) at your technical drawing the following cases can appear.

1. All planes are coplanar => it doesn't matter what option you are choosing, due to the fact that the computed offset is 0.0
2. Planes are parallel but not coplanar; all TEDs are based on one of these planes
   => this plane is the first one you are putting into the datum field and you are choosing on first element
3. Planes are parallel but not coplanar; all TEDS are based from a middle plane
   => you are choosing middle plane

Planes between 2 and 3 differs only in location, not in normal. Setting one plane as zero allows us to provide you the same TEDs as on your technical drawing and not having an offset between the TEDs computed from the software and the one from your drawing.

In short, the software computes the plane according to the ISO 5459 and afterwards it moves it the the specified location alongside the computed normals to give you better TEDs at position tolerances.

 

[[Yt...]]

Hi Christoph

Thanks for your response. 

Just more clarification for me.

My question came from doubt regarding position and profile tolerances, when the results vary between the two different options of the result plane. 
The color map also shows a substantial change in the alignment of the scanned file with the CAD and then of course, also regrading to the correctness of the CNC process. so it seems that the calculation of the plane location changes, and not just the calculation of its normal. What about the drawing request for common datum out of the whole 5 pads? 

I'm taking about a part with 4 coplanar pads and offset one which is 15.39mm higher. They are all should function together. 
How can the alignment variation be explained as a function of the method of the resulting plane? In addition, I need to have the same measurement method for myself and for our supplier.
Can you please explain to me more about the desired process and whether these two calculation methods come from a standard guidlines?
Is it required or is there such a guideline for not coplanar common datum, to avoid this issue and first perform an offset for the upper plane relative to all 4 others and create a uniform plane in advance?

 

[[Ch...]]

Yes, it is not unexpected that the results vary from these two methods. Reason for the difference is that in actual your plane have a distance that differs from the nominal distance. Therefore your designer has to be answered the question from which point all the TEDs have to be computed.

From somewhere in the middle between the planes (but this means that the distance between this middle plane and your 5 planes differes between nominal and actual). Or do you designer choose one plane as zero point (then of course your other plane have a distance different then 15.39 mm).

Both interpretation are valid according to the corresponding standards due to the fact that is not defined how the datum situation plane is established from the assiociated datum. Therefore we are allowing both interpretation inside our software. 

In short we are established the associated datum as it described the standard. For further details please have an look in the corresponding standard. Afterwards we are established the datum situation plane, either at one plane or in the middle of all planes.

But to be honest I am afraid the problem is too complex for the forum. Therefore if any question left please contact your local partner or our local support team so that you can discuss the remaining question on your given project. 

 

Best regards

Christoph Schult