Unable to Create Driving Dimension

[pmcnicoll]

Please see the attached f3d file. I have also included screenshots to illustrate the issue. I cannot dimension the angle which the left arcs end at, as Fusion tells me that there is a conflicting dimension. The conflicting dimension that it lists is the very same dimension. One dimension should not conflict itself. I have dimensioned the other side of the profile in this way, and previous experience would suggest that it is a sound method.

[whittakerdw]

One thing you could try is drawing a construction line perpendicular to the blue line and constrain that with a perpendicular constraint. I'm sure there are other ways but that is the one I found to be most convenient.

[lichtzeichenanlage]

Here you go: 

[pmcnicoll]

Thank you. I actually did attempt this solution; I couldn't get it to solve
on my machine.

[lichtzeichenanlage]

Shouldn't be a machine thing. Perhaps (but not likely) an OS thing. I'm using a Windows 7 based machine. 

However. I've recreated the solution and here is the export. 

[pmcnicoll]

Oh I agree but that's simply what I experienced.

I'm running Windows 10; it seems unlikely but anything is possible.

[whittakerdw]

That's very odd. The perpendicular line should have constrained the blue line. As long as it were 90 degrees to the blue line as shown in the picture then it should have worked. But regardless I hope the other option was a little more helpful to you.

[pmcnicoll]

Thank you, this appears valid.

However, we shouldn't have to bend over backwards and play guessing games with Fusion to make a dimension work. Both of our dimensions are parametrically identical, so the fact that one works and one doesn't is indicative of a real issue. It's a waste of time to have to figure out just how Fusion wants something dimensioned - I should be in the driver's seat, not the passenger's.

For this reason, I have left the issue unsolved. 

[whittakerdw]

A way for the driven dimensions to be located easier would be something to submit to the IdeaStation

[chrisplyler]

@jeff.strater 

[jeff_strater]

@chrisplyler , I think there is one missing constraint here.  The end of the pair of arcs is not quite aligned with the center of those arcs.  If I add a coincident constraint, then I can add the angle dimension as a driving dimension.

[chrisplyler]

---

@jeff_strater wrote:

@chrisplyler , I think there is one missing constraint here.  The end of the pair of arcs is not quite aligned with the center of those arcs.  If I add a coincident constraint, then I can add the angle dimension as a driving dimension.

---

Very interesting.  How is it geometrically possible - if one of the arcs is an Offset from the other - that their endpoints are not INHERENTLY coincidentally aligned with their center point?

And even if one wasn't an Offset from the other, and so they could have different included angles, I should still be able to set a driving angular dimension between the two straight lines, right? Like the picture below. So why can't I set one in the other setup?

[pmcnicoll]

To answer your first question, the paths of the arcs have already been defined, but their length is still variable. Much like if you start a line at the origin, and you define its angle from the horizontal, the line will appear black. The line's second point (not origin) will be undefined, as the length is not defined. But only the point is undefined; the path is clear. If you think about it, the line isn't going anywhere; it can only change length. The same is true here. As for your second question, I share your sentiment. That was the reason for my post. 

[pmcnicoll]

Thanks for the reply. I understand your comment about the coincidence to the center point; this would definitely be a relation that I should have included.

However, we still have an issue. I should be able to define the angle of my line with or without a centerpoint coincidence. Right now, I can't do that without making a driven dimension. Purely by concept, the fact that a blue line would automatically create a driven dimension is wacky. 

[chrisplyler]

---

@pmcnicoll wrote:

To answer your first question, the paths of the arcs have already been defined, but their length is still variable. Much like if you start a line at the origin, and you define its angle from the horizontal, the line will appear black. The line's second point (not origin) will be undefined, as the length is not defined. But only the point is undefined; the path is clear. If you think about it, the line isn't going anywhere; it can only change length. The same is true here. As for your second question, I share your sentiment. That was the reason for my post. 

---

But if one is an Offset from the other, their endpoints are coincidentally aligned with their center. That's the result of Offsetting an arc in Fusion. Go ahead and drag one endpoint of one of the arcs...the corresponding endpoint of the other arc will follow. They will always be coincidentally aligned with their center.

So I don't understand why you adding that constraint made any difference. #1 it ought to be coincident to the center point of the arcs inherently, and #2 we ought be able to drive it's angle even if it weren't.

I suspect a bug.

[chrisplyler]

---

@pmcnicoll wrote:

Thanks for the reply. I understand your comment about the coincidence to the center point; this would definitely be a relation that I should have included.

However, we still have an issue. I should be able to define the angle of my line with or without a centerpoint coincidence. Right now, I can't do that without making a driven dimension. Purely by concept, the fact that a blue line would automatically create a driven dimension is wacky. 

---

Yup. I think it's a bug. Constraining a line coincident to some point should not be a requirement of driving the angle of that line. And in this case, in your sketch, it's inherently coincident anyway.

[pmcnicoll]

I attached an example of two concentric arcs whose ends are joined by a NON-centered point. 

I think I may have missed something you said earlier. These arcs aren't offsets; they are simply concentric. Perhaps an offset would suit me better next time in a situation like this. 

[chrisplyler]

Right. In your original video, however, the inner arc IS an offset from the outer arc.

Somewhere in the secret code that is Fusion 360, the line that connects the two endpoints isn't being recognized as coincident to the arcs' center point, although it obviously must be. When Jeff added the coincident restraint, it SHOULD return an over-constrained error.