What are those arrows in an work axis constraint?

[oransen]

Silly question I'm sure, but I'm a programmer not a mechanical engineer. That is my excuse anyway.

You see these two arrows on this axis-axis constraint?

Does this mean that axes have a direction? If so how can I find out what it is?

I mean when I create an axis, no arrow appears, so how do I find out the direction? If there is one...

TIA unwashed and somewhat slightly dazed...

[imajar]

Yes, axes have a direction.  I personally don't know of a way to view that orientation outside of viewing those arrows while making a constraint - maybe somebody else does.

In the API, you can drill down and get the direction of the axis as a vector.  (workaxis.line.direction)

[jtylerbc]

It is a direction.  However, until recently (last couple of releases), direction on an axis constraint didn't really exist.  The axis constraint aligned the two axes, but controlling the direction of the alignment was done using additional constraints.

I suspect that since it is relatively new, making the direction more easily identifiable just hasn't happened.  In normal (non-programming) Inventor usage, the only time it matters is while creating an axis-axis constraint - which also happens to be the one time you can see the direction arrows graphically.

[johnsonshiue · Accepted solution]

Hi Folks,

Indeed, each axis has a direction. However, it should be mostly irrelevant in modeling except assembly constraint. When constraining a component, an Axial Mate can have two solutions (180 deg apart). Though mathematically, either solution is valid, it is not acceptable from mechanical designer's perspective. The component can flip to an undesirable direction leading to other failures.

A while back, there was a change to Angular Constraint to allow an aligned vector (Explicit Reference Vector) so that the direction can persist. In 2018, we made a change to Axial Mate constraint for similar reasons due to negative comments on Mate direction stability.

When creating an axis, the user cannot tell the direction. Nor is it relevant.

Many thanks!

[imajar]

Actually, if I can disagree on one point - I think being able to see and control axis orientation is relevant.  There have been times that I wished an axis or line was oriented the opposite direction to agree with my sign conventions.  I've never made a fuss over it because it is easy to work around. 

On a similar note, there have been other times that I wished there were better inputs/outputs for getting/applying mathematical transformations.

[SharkDesign · Accepted solution]

You can also see tiny arrows in the constraint dialogue if that helps you understand the constraint you are placing. 

[oransen]

@johnsonshiue, thanks for the detailed reply. Does this image summarize what you mean...?

So if you have a plane to plane constraint already in place you should choose "undirected" axis constraint?

I suppose it makes sense because who can tell which direction a hole or circular cut axis should go...?

[SharkDesign]

Personally I always use undirected, if you constrain it properly you should never need to tell it which direction. 

From what I understand, the direction was added because sometimes in large assemblies the direction of the axial mate can flip to the wrong side and cause errors and this was the solution to stop it happening. 

[oransen]

jameswillo, yes, I suppose undirected axes are more natural, and face mate and flush are more natural for organizing the "direction" of the objects.

[SharkDesign]

If you don't use a plane mate with an axial, then it can essentially move
in a direction infinitely is the way I look at it.

[oransen]

Yes.