Showing results for 
Search instead for 
Did you mean: 
Showing results for 
Search instead for 
Did you mean: 

Issues with different accuracies


Issues with different accuracies


Can I have different accuracy within an assembly? What can be the issues of same?

Let me elaborate this.

Suppose I have an assembly (MAIN.ASM) having 4 parts (A,B,C and D.prt) and one sub assembly (SUBASSEMBLY.asm).Sub assembly has two parts (E and F.prt).

All parts, sub assembly are having relative accuracy of 0.0012 and all has been assembled into main assembly having relative accuracy of 0.0012.

Now in a scenario, I need to add a feature into E.prt which is asking for absolute accuracy of 0.0001. I am changing accuracy and updating part.

With these update, will there be any effect on MAIN.ASM like regeneration failure or constraint failure or any other failure?

What would be effect on interference or clearance with this update?

Hope I have clarified my doubt. Feel free to ask in case of any confusion.




I assemble parts with different accuracies all the time and have no major issues.


please see following link: Your model accuracy? Other users described some problems around model accuracy.

What would be effect on interference or clearance with this update?

- Not able to answer your question .

The only problem l have had is TOP-DOWN design.

- if l share/propagate geometry from main skeleton to subskeletons or parts, than it can fail due different model accuracy.

If possible, keep one model accuracy in whole project.

More expirienced users give you better tips...


Milan Bonka


You have to be careful about setting up something like this.  It requires planning.  Typically, you also have the discipline to minimize your dependencies to required interfaces only and a type of skeleton (core Creo) that the assembly extension provides.

The biggest problem I see with this is when you decide to start using cross-component-constrained parts in a new project.

The way I look at it, if you go this route, treat your assembly as a single inseparable part.  Otherwise, be prepared for reference searches that seem to have nothing to do with your current effort.

I am not able to understand why I need to use assembly as inseparable part? It's already having accuracy of 0.0012 which all other models are having... I have changed accuracy of part e... If part e is acceptable by main assembly, then why it's not acceptable by other models?

Did you mean this restriction or something else by saying inseparable part?

This was more a response to top-down design than accuracy.

My consideration is that  parts that have a lot of external references are not usable in other projects until those references are transferred to the new project.

During most of my career, even in the engineer's mind before CAD, was this top-down development techniques (pretty much what drafting board designs did) that just tantalized the entire engineering community.  In practice, after the evolution of CAD all these year, top-down design is a very disciplined niche within the design community.  When you work with production, change management, document control, etc., these techniques become too fragile to manage for most small to midsize organizations.  Invariably, you will have follow-on product from your initial product.  This is typically where stuff hits the fan and engineering managers get racked over the coals.  I consider the top-down development for the smaller organizations as a skunk works modeling effort that will, at some point, be thrown over the wall to get cleaned up.

This certainly doesn't say that one should understand the method.  It is the dream of many of us that use to manage design with Mylar and colored pencils.  You will soon figure out why discipline is required.

Any specific reason for failure in top down design? Can failure be avoided by taking extra effort?

Specific reaon from my praxis:

- when some dimmension are MUCH MORE LARGER than others dimmensions

- construction of more complicated duct (transitions square-circle or square-octagonal)

- duct is 3000mm long, but the wall is only 5mm

- using TOP-DOWN design


Problems during geometry propagation:

- square corner (R5) "was messed = failure" --- subskeleton geometry and final part coudn´t be cunstructed

- the flat wall looked like "crumpled paper"

Source of problems:

- my vigilance ( and experiments with accuracy )

- main skeleton accuracy A

- subskeleton accuracy B

- final part accuracy C ( and it was sheetmetal part )


- set same model accuracy in whole project

Hope it can helps...


Milan Bonka

In my experience, there will not be any issues in the assembly unless there is an intersecting feature like an assembly cut or any merge features....that intersects with the part having different accuracy.

Agreed, Manjunath RV.

A horrible problem we had with assemblies is if the accuracy of some of the constituents is at the default (0.0012 relative in inch units), a global interference check showed everything being okay, when there was, in the "real world" an interference between two parts. I don't recall the combination of accuracy settings in the parts and assemblies, but luckily it showed up when the machinist was programming the parts and noticed something was not adding up. Just another consideration.

That's why I still do drawings   You'd be amazed at how much cost reduction I catch while creating detail drawings.


Attention: Creo 7.0 Customers
Please consider upgrading
End of Life announcement here.