Assembly constraints

[Sachin]

Hi while assembling a component, in automatic constraining, when we select two flat surfaces of the child and parent components, the constraint type get changed to "Normal" automatically. Can we control this auto-switch to "Coincident" by any means?

This thread is inactive and closed by the PTC Community Management Team. If you would like to provide a reply and re-open this thread, please notify the moderator and reference the thread. You may also use "Start a topic" button to ask a new question. Please be sure to include what version of the PTC product you are using so another community member knowledgeable about your version may be able to assist.

[Dale_Rosema]

Here are a couple of links to product ideas for this.

http://communities.ptc.com/ideas/1132

http://communities.ptc.com/ideas/1128

Thanks, Dale

[Sachin]

Are there any plans set by PTC to work upon these ideas?

[StephenW]

They are already implemented. Use the following config.pro options:

For eliminating the normal constraint from happening by default:

comp_angle_offset_eps 91

comp_normal_offset_eps -1

Add this option to get it to default to coincident instead of distance:

auto_constr_always_use_offset never

[StephenW]

Try using these for eliminating the normal constraint from happening by default:

comp_angle_offset_eps 91

comp_normal_offset_eps -1

Also, you might add this option to get it to default to coincident instead of distance:

auto_constr_always_use_offset never

[DRFaust]

By setting the config options

comp_angle_offset_eps 91

comp_normal_offset_eps -1

auto_constr_always_use_offset never

Assembly works perfect as would expect it to.

Thank you very much

[Sachin]

Steven that is really a good practice..!

[JonathanHodgson]

I'd argue that depends entirely on what you're modelling. For parts that are a 'matched set', and will only ever be assembled with the others of that one set, it works well. When we model a gearbox casing assembly (so for example a 'bellhousing', maincase and end cover) we model them on a common coordinate system (although not the car coordinate system...)

However, when we model a bearing (to take the other extreme) it would clearly be ridiculous to model it on the gearbox coordinate system, because not only might it be used in more than one position within this gearbox, it may also be used in any number of other gearboxes - some of which may not even have the same shaft orientation (transverse vs longitudinal).

IMO, as soon as there's even a possibility that a component may be used in more than one position, functional constraints (i.e. insert bore on shaft; mate face to abutment) are the only way to go. They also represent design intent, whereas fixing a component at an absolute point in space doesn't...

(I did recently watch a colleague modelling a gearbox with many - if not all - parts on absolute coordinate systems. It looked a royal PITA when he needed to update the position of (what should have been) an assembly, and he was also trying to represent a sliding component - used in two places - in both 'engaged' and 'disengaged' positions!)