CATia 5 Model Tree Structure and Modeling

[pedja14]

I am relatively new to Catia 5 and the thing that I find somehow confusing is "suggested" model organization using bodies and than assemble them using Boolean operations.
So I would like to ask experienced CATIA users what is the benefit of making model tree that complex creating a new body for almost any modeling entity when you can do it much more simpler in one body.
In that tutorial "suggested" technique I even found a bizarre thing such as putting a single hole in a separate body and than assemble it after.
If holes and pockets automatically remove the material what is the meaning of using the Boolean operations?

Thanks

 

[Configurator]

That is absolutely the wrong way to use CATIA V5.

Where did you find this "suggested" techinque?

V5 is FEATURE BASED. A hole will perform better than a removed cylinder. A pocket will perform better than a removed prism.

If I were you, I wouldn't put much stock in whatever document you are reading. It was probably written by a V4 user. If it came from Dassault, I'll be extremely surprised.

 

[pedja14]

I haven’t obviously made myself clear enough. Suggested technique does use the hole and not cylinder. However it puts the hole in a new body in which btw the hole is only feature. In the and it assembles the "hole" body to the main one.
My question is: Why they suggest to separate the features in the separate bodies when you can use only one.
This way you get complex model tree structure even for the simple part? What is the benefit if any?

This suggested techniques:

http://www.amazon.com/exec/obidos/t...2-1416714-5822403?v=glance&s=books&n=507846is

the source for

 

[itsmyjob]

Hi,

I hope we can meet 1 of the 2 authors in Phoenix and ask them about that.

Also I am surprised they would suggest 1 feature per body. This is not the best way to use CATIA V5.

A lot of people ask me about how I would build a solid, my answer is still the same : Make it easy to understand and to modify. Then think about optimizing update...

A Tree to long (more than 100 features in 1 body) might not be easy to 'read'. Even if I don't need to use boolean operations to build my solid I will use some to make it easier to understand.

...also, I almost forgot, the geometry got to be exact. A 10" pad is not a 10" 1/16 ... Even if the tree looks good your geometry is not.

After the COE topgun, the 5 best users should explain how they built their solid... and why...

indocti discant et ament meminisse periti
Eric N.​

 

[Configurator]

^^ Agreed. Assembling a few bodies here and there just to make navigating the tree easier is a reasonable thing to do. I can't think of any reason why anyone would impose a one feature per body limitation. I can understand monodetail sketches, but putting a hole into a seperate body and assembling it to the solid is a waste of time (and probably resources), plus it would make the tree harder to read.


If you read the reviews of the book, one of the two authors isn't even a V5 guy.

 

[pedja14]

Thank you for responses
It wasn’t suggested 1 feature per body, however in one particular example it was suggested to put a simple hole in a separate body.
Most of the other bodies had a few features.
But one or a few does it really make difference?
I don’t think so.

 

[catiajim]

There is some power to putting together bodies, not only in Model Organization, but in terms of creating PowerCopies and UserFeatures, as well as Knowledgeware Applications. I might give someone an example of a single hole in a separate body. But only to show the difference between a "negative" and a "positive" body, and what the Assemble operation does as opposed to the Add or Subtract operation. I would not tell them that that is any way to work in normal circumstances.