Anchoring parts Vs. Three constraints.

[Phil]

Anchoring parts Vs. Three constraints.

When I create an assembly in Alibre, I generally use 3 constraints per part to prevent them moving, the first one is constrained to the assembly datums and the others are built up from there. This is the method I was taught on a Pro/Engineer course about 7 years ago.

My colleagues are self-taught 3D users and have only used Alibre - they tend to use constraints to get the part in the right place then they "anchor" the part - this works OK in Alibre but things seem to fall apart when the assembly is exported as a STEP into Solidworks, Pro/E or UG.

What is the purpose of the "anchor part" command in Alibre? Is it not just as easy to apply another constraint (or two)?

 

[MilesH]

I can't see how the fact that things "fall apart" on imported STEP files can have anything to do with how they were originally constrained?

There are known issues with importing STEP Assemblies from Alibre into Solidworks, I guess that's what you're experiencing. There seems to be disagreement about the STEP implementation...

 

[Phil]

When importing to Pro/Engineer, the assembly that has the anchors seems to ratain the one or two constraints that have been applied then ignore the anchor.

Consider a part that has two constraints so that it is in the correct place (but still has freedom of movement in one direction) which, has been anchored - within Alibre you cannot move the part at all.

Now, the imported assembly in Pro/Engineer seems to have retained the two constraints and the freedom of movement in one direction - however, the part will now be mis-placed in that free direction.

In the second instance, where the parts have been constrained in three directions within Alibre, the imported assembly is just as you would expect it.

 

[MilesH]

Curious.....

AFAIK, no data on constraints is stored in the STEP file. The positioning is set on Export and if things move on Import to another program that's due to discrepancies in the STEP implementation.

What happens if the part is simply Anchored, without any redundant constraints?

 

[bhershberger]

What is the purpose of the "anchor part" command in Alibre? Is it not just as easy to apply another constraint (or two)?

I used to constrain the first part to the assembly datums like in Proe, and then build up the other parts from there like you said. However, at an Alibre training session I went to, they suggested anchoring the first part to the origin (which is easy to do because you can insert the part right at the origin) and then build up from there. That seems to work pretty well.

I would be afraid to anchor successive parts because I usually have to change something, and I’d hate to have to redo part locations...

Cheers

 

[wfpelletier]

Re: Anchoring parts Vs. Three constraints.

When I create an assembly in Alibre, I generally use 3 constraints per part to prevent them moving, the first one is constrained to the assembly datums and the others are built up from there. This is the method I was taught on a Pro/Engineer course about 7 years ago.

Like some of the other posters here, when I place my first part in an assembly in Alibre, I tend to place the part at the origin and use the 'anchor part' function to fix the position of the part. However, I have also used the '3 constraint' method as well. (I do not use the 'anchor part' function for subsequent parts, however.) Even though this seems to work just fine, I have heard some say that it is considered a better practice in 3d parametric software to use constraints to position your first part. (I think I saw a posting on the SolidWorks board suggesting this. However, in SolidWorks, I almost always use the SolidWorks equivalent of 'anchor part' to locate my first part.)

I think either method is acceptable. Unless I do not want my first part to be placed on the assembly origin (which is pretty rare), I usually will anchor the first part in an assembly model. However, I have a suspicion in the back of my mind that it might be better to use constraints to locate the first part in an assembly. (Since I am usually too lazy to do this, I prefer the 'anchor part' method!)

wfpelletier

 

[leeave96]

Normally I try to use constraints to fix a part in space or with respect to another part, however, there is an exception and it is an aggrivating one. When mating hole center distances are close, but not exact, like out past 6 places (caused by making one part in context with another). When I can't get the mates to work-out, but for all practical purposes it doesn't matter, then I will fix the part - just so I can move on.

Bill