Part Body Modelling

[warburton1]

Hello all,

I have a question or rather I am questionning the general rule where it is considered good practice to model in a seperate Body to that of the Primary Part Body.

The reasons given are;-

1) Easier to edit

2) Easier to correct if something becomes corrupted

3) All elements can be either assembled or cut and pasted into Part Body at the end.

I remain unconvinced to all of these and would welcome any comments that could either reinforce the principal or disprove it.
My own opinion is that for;-

1) Ease of editing is based upon your modelling process rather than where it is carried out. Robust modelling / horizontal modelling / skeleton modelling do not really require the use of a seperate part body to be effective. This is exemplified by alternative feature based modellers that do not use multiple part bodies to create robust models with. Unless of course Catia is admitting to a fatal flaw relative to item 2 - corruption in models.

2) Non-corrupt data can be cut and pasted out of a solid at any time and does not need a seperate body initially, this could be created as and when required.

3) The assembly of the final part body becomes an additional process in the assembly of the solid. I prefer the "Keep it simple" concept over this approach. As for cutting and pasting it all into the partbody, well, again could this not be the cause of some future corruption / error.


As I say any comments would be appreciated.

Thanks Gary W.

 

[jackk]

Gary... in my opinion

1) no easier then working with the first part body or any part body

2) same as 1)

3) I've never experienced any corruption/errors when using the primary body for boolean assembles or with assembly design. (I'm not sure I follow your comment)

Are you modeling more than one part in a CATPart?

Where did you get these 'general rules'? Maybe I'm missing something here, but I think these rules are garbage!

...Jack

 

[Ashvanth]

Hi warburton

As jackk is saying there r no general rules

Whatever u have said above is from one individuals perception.

Its absolutely their freedom to work all commands in one single Body or multiple bodies.

The reason for going to multiple bodies r as follows.
This is what i feel.

1.Imagine a big part say for instance where tree runs to some 10-15 scrolls of mouse on the catia screen.
I wud say a person doing all commands with this part under one single body is the having so much of patience.

2.I really agree on ur point below

"Ease of editing is based upon your modelling process rather than where it is carried out. Robust modelling / horizontal modelling / skeleton modelling do not really require the use of a seperate part body to be effective. "

3.But what if some feature has to created with reference from any other command say trim or split which is at the beginning of the tree and after which u have done some 50 splits after that( just for example) or as many trims as i have said then its going to be a hectic job to get the trace of the Split or whatever command it is. just think of it.( the above said is WRT wireframes and surfaces).

Hope u can really get what it is by now.

Thats the reason people go into many bodies of course rename it with proper associative names governing the commands and perform operations in it and do all the needed (i.e the results) in the main body.
that is for easy understanding purpose.

bye
Ashvanth

 

[warburton1]

Thanks for your feedback Jack and Ashvanth,

Jack I actually have the same opinion as you, I think these rules are garbage. Probably stemming from what someone has heard somebody else talking about around the water cooler!!.

As for the corruption that is their argument not mine. I think that this comes back to my point about the half listening into conversations around the water cooler.

Asvath, I understand what you mean about complex large tree simplification, using multiple bodies, however, I think the problem here is that the training fraternity have taken it that every model should be constructed in the same manner.

I work under contract for a company that is currently retraining its staff to v5. I am a certified catia professional and have thus far escaped the training schedule, however, those coming off the training course are all being taught to do all their modelling inside an alternate body to the part body. Also one of the users have read about this approach in a book by Kleismet and Karam.

Still if anyone could explain to me with reasons that can be conclusive, why it is necessary then I am willing to listen. Thus far I cannot justify the use of what I consider to be a waste of time and effort for little or no gain.


Gary W.

 

[rsingh75]

Hi,

There is no hard and fast rule as how a model has to be built,

1)Ease of editing :-

I totally accept that it depends on the modeling process but it's would prefer to do the model into chunks and do, (which means do it in many open bodies, If required). I will do this type of modeling because

a) the amount of time to update the model is less, since the model is built in chunks for example modeling of engine is done in cores like the front core, top core, side core etc.. We will not be able to find the difference is small models but when doing a big model, this time really counts.

b) More than one person can work ( This is applicable for big models like engine...)

c) Every person has a unique style of modeling, If any corrections have to be done to the model then the best person to do will be the one who has modeled and if he is not there and the model is given to a different person then the amount of time taken to understand the model is high.

2) Easier to correct

This is a relative term with first point, If the model is built in chunks and if any correction has to be done then the change will be done in that chunk and update the whole model ( this is something like creating models in a model).

3) All elements can be either assembled or cut and pasted into Part Body at the end.
This is the difficult task if modeled in chunks, We face problems in assembling the two chunks but again the i feel the problem is worth it's effort.

I think my views might be wrong also, will appreciate if onyone can tell more about it.

Regards

Ravindra

 

[Azrael]

The only feedback I can get U on to start with a separate body is if you don´t know how to build the model it can be a goog idea due to the boolean operations are limited with the start body. Say that you under your modeleing see that you need to use your body as a subract body you can´t do it but on the other hand you can use the replace feature so I really don´t see the point.