Inventor 2019 Axis-to-Axis Mate Constraint Default To Forced "Opposed" Solution

[jletcher]

I understand you added it from someones idea, but why could you not have left the default selection the old way? This way it does not screw up other peoples workflows?

When on the fly this thing is now a nightmare. I don't understand why it was even needed but now more clicks more wasted time for me and others.

Make it so users can default to old style please, these are things you should be thinking when making changes. I understand you wanted to please someone but don't do it at the expense of others. You can leave it just change default default to old style, I will never have a need for this new way.

This is going to drive me nuts and many users have already called asking if there is a way to default to old.

This should be easy to fix.

kelly.young has edited your subject line for clarity: 2019 New mate constraint nightmare

[DRoam]

@kristin.jakuszanek, agreed that the current behavior isn't ideal and it should default to undirected.

Just wanted to offer you a quick tip in the meantime that should save you some hassle. You mentioned that you use the free rotate tool to try and get your part oriented the right way, and even after doing that it sometimes snaps into the wrong orientation.

Rather than doing that, you could take advantage of the new functionality and edit the axis constraint, then pick either the Opposed or Aligned solution to get your part to pop into the right orientation. Then you can change it back to undirected and say OK, and it will remain in the correct orientation. This accomplishes the same thing as using free rotate, but it's more dependable, and should be quicker, too.

I have to say though... the fact that you're saying you need to use free-rotate because your part is facing the "wrong" direction... this would indicate to me that there's a "right" direction, and therefore you should be using the Opposed or Aligned solution to enforce that..... just a thought. I won't presume to know your workflow. But this sounds like a situation where the new functionality could actually help your assembly be more stable.

[EDIT: to put it another way, your assembly should be fully defined. If the orientation of your cylindrical part matters, and some other constraint (like an angle or mate/flush constraint) isn't enforcing the orientation, then the axis-to-axis constraint should be... that's what the new functionality is there for.]

[kristin.jakuszanek]

Let me rephrase. 

When I say right way vs wrong way, this is only because it flips mates that I already have in place. I will typically do a face to face mate of the cylinders and then an axis to axis mate. When I do the face to face mate, I expect that to hold and not flip flush because of an axis, because I really don't care about what inventor deems the top or bottom of the straight cylinder. 

Face to Face mate flipped flush because of axis to axis mate   My frustration with the axis to axis mate occurs when I am trying to quickly move through constraining my cylinders, and it starts flipping all my pieces around to flush face to face mates, and it will flip other parts constrained on that axis, which is visually annoying.

I understand that I can click to the next mate type, which is the aligned, to the stop this issue, which is what I am sure the autodesk team is hoping, but it is unnecessary for my work flow, and causes an extra step regardless of how I move through my project. If it was just automatically set to undirected (or if we were able to choose how we want our axis constraints to default to) it would completely negate the issue and allow me to go back to my previous work flow where I didn't have to worry about knowing the direction an axis is 'pointing'.

Again, I understand the feature, why it was added, and how it can be beneficial, but I think the autodesk team fails to realize why it is causing more of a pain for certain users. I can see having it set as the default setting, allowing users to see and work with the new mate type, and see if it works for them, but I feel it should be up to the end user as to how they want the setting defaulted to, seeing as one work flow using directed axis can vary from a work flow in which axis direction doesn't matter. 

It's a neat new addition, don't get me wrong, but it has no benefit to me, and many others, it seems.

[DRoam]

---

@kristin.jakuszanek wrote:

When I say right way vs wrong way, this is only because it flips mates that I already have in place.

---

I'm on board with everything you're saying except the phrase above, specifically the part I underlined. Adding a new constraint should never be able to flip a mate/flush constraint that you already have in place. This shouldn't be possible, and I've never seen it happen. I may just be misunderstanding what you're saying...

Could you post a screencast of the bad flipping behavior in action so I can see the particulars of your situation?

[kristin.jakuszanek]

I had included an image of this in my post you replied to. It shows two cylinders, one cylinder has flipped inside of the other cylinder causing their faces to be flush with one another, as opposed to mated flat face facing the opposed flat face.
I can include additional images showing this a little more clearly, and with a little more of the constraints on the parts on Monday when I have access to my work computer again, if you’d like.

[Anonymous]

So you say overconstraining parts is more robust then.. Usually, when you screw two parts together, there are certain faces mating and holes matching each other. Then, whe you decide to screw it together mating opposite face (for whatever reason you have, it happens), you have to change three constrains - mate constrain and two directed axis constrains. Changing only one of them would end with one or two inconsistent relationships.

I believe situations, when you dont have faces matching and just want two parts not to flip on axis, are less common.

Thats why undirected should be default in my opinion (or at least it could be set up somewhere by user).

[Anonymous]

So you say overconstraining parts in assembly is more robust?

When constraining parts, I try to simulate real world constrains. Let's see an example:

Part is constrained, when two axis constrains and one flush/mate constrains are applied. If I decide to flip part for some reason, I just edit flush/mate constrain, switch one face like this:

That was not a problem when axis constrain was undirected.

Now, when constrains are "more robust", you have to change three constrains, because all three define orientation of part. That in my opinion is called overconstraining. Directed axis constrain doesnt help a thing, when you still have to determine position on axis using flush/mate constrain. 

My question then is, in what cases do you consider directed axis constrain more robust and you dont need to use another mate/flush constrain?

Thanks for your opinion

[Cris-Ideas]

Hi,

Not saying you do not have a point here there are however situations when directed axial mate I consider a better option.

This would be in cases when two parts (usually cylindrical) are physically (in real world) assembled in the way they can not flip and in the same time I am not interested in any flush or mate constrains of their ends, or in general not interested at all in applying any other constrain but axial mate.

Until now I was forced to apply 2 or 3 (depending on case) constrains to achieve stable relative position of such alignment.

So directed axial mate is valuable improvement, but in the same time my opinion is undirected should stay as default if default not possible to be selected by user.

Cris.

[SteveMDennis]

@Anonymous

I have come to understand much better how some people are comfortable with undirected and prefer it via this thread.

But having said that, undirected by its very nature is NOT fully constrained. There are two solutions that are equally valid, fully constrained means to me that the part is constrained such that it will not move.

So you have dealt with that since R1 by adding another constraint to fix those cylinders.

As I type this maybe "more robust" is a bit strong, because to FULLY constrain it you still need another constraint or you still have DOF available...

But again, this was done to address user cases where undirected was used (perhaps to James' point, incorrectly) and the cylinders would flip.

We are still reacting to this thread internally.

[josh.nieman]

If we can't have "undirected" by default, in some way shape or form.  

Ok, that may be slightly over stated  but come on... "still reacting to internally" -  this is ridiculous.  Unless you mean "posting screenshots to the break room bulletin board for laughs" in which case I empathize. 

Edited by
Discussion_Admin

[SteveMDennis]

I am not allowed by law to comment on releases that have not been released yet.

Sorry if a "lawyerly" type answer didn't sit well with you.

[Cris-Ideas]

Actually I think this kind of answer does not "sit well" with anyone.

It is basically not giving anything to anyone. And as such is completely useless. 

Autodesk is "reacting internally" to so many things that probably no one is even kipping track of them. And as we all know well this reacting can take as long as decade or more to produce any output.

So kip up "reacting"  in the meantime we will be getting older.

And as for commenting on new releases you very easily comment on them when you need to announce that something was fixed.  It had been many, many times when I was reading "this was fixed in latest beta".

And suddenly you are "not allowed by law".

Ridiculous.

Cris.

[mdavis22569]

While it might be in the beta ... 

That doesn't mean it will be released with the next update. There might be an issue delaying it from putting put out in the next update/enhancement. It's happened a few times that I recall from being with the beta group and an end user. You see it all resolved .... next enhancement comes out and it's not including.

So that's probably why you won't see the fixed in next release ... but will see it say fixed in the beta ... 

[DRoam]

I'm tired of hearing that "using the directed axis mate is bad because it over-constrains your assembly", or, "NOT using the directed axis mate is bad because then your assembly is UNDER-constrained".

Neither of these is necessarily true. It completely depends on the situation.

Here is an example of a situation where the directed axis mate is necessary to fully constrain the assembly,  but it does not over-constrain it:

In this case, a directed axis constraint is necessary to get the horizontal leg of the elbow aligned with the ball valve AND facing the correct direction. (That is, just as necessary as having mate/flush direction control is... yes, I could use an undirected axis mate with an Angle constraint, but how would you like for all face-to-face constraints to be undirected and have to apply an extra Angle constraint for every single one?) The directed axis mate is just the right tool for this job.

And here is an example where using the directed axis mate is NOT necessary to fully constrain the assembly, and in fact using it would over-constrain it:

(Sorry, the system won't let me embed the Screencast, so you'll have to click this link to watch it: Importance of a free-direction Axis mate)

In this scenario, the mate/flush constraint controls not just the position of the pin but also its direction, so enforcing the direction with the axis mate as well would over-constrain the assembly, and result in a sick assembly if the flush constraint changes direction, as shown in the video. And this is why Undirected was left as an available solution.

So, does using a directed axis mate automatically mean an assembly is over-constrained? No. (i.e. the second example). But is it sometimes necessary to get a fully-constrained assembly? Yes. (i.e. the first example).

The fact that the directed solutions were added, or that the undirected solution still exists -- neither of these facts is an issue. Please stop complaining that "directed" was added or "undirected" still exists. Both of these are tools, and are very useful in the right situations. If you don't need one of them, don't use it.

The only issue here is the default behavior (the fact that a directed solution is enforced automatically when creating a new axis-to-axis constraint), and that's the only thing anyone should be complaining about.

[3D4Play]

Thanks for that. It really ticks me off that Inventor uses an automatically enforced behavior that could and should be a user-defined default condition - especially after reading this thread, in which more posts seem to be asking for that, than arguing against it. I don't see anyone saying, "...if you made it selectable, I'd complain..." But, it's currently not a user-definable default, and it now forces (often interrupts) my workflow, when it didn't used to do so.

[Cris-Ideas]

Actually in the second scenario (from two posts up) when axial undirected mate and surface mate are applied this assembly becomes over constrained (strictly mathematically speaking) as surface mate and axial mate explicitly define direction of alignment of the parts.

In this case it is safe because axis of the hole and normal to face are aligned.

Strictly following the theory if we would like this assembly not to be over constrained one surface mate should be used and axis/point mate. This way assembly would  be constrained with "just right" number of constrains.

As a general case most assemblies are in fact constrained in a way that mathematically they are over constrained. Even if user understands this otherwise.

Other thing is inventor in fact does not validate solution obtained that leads to numerous problems with solve.

So in fact problems with solving are not related to a particular type of constrain but method used to solve constrain set defined in the assembly. I had many ridiculous solutions from inventor when fully legitimate set of constrains was supposable "not possible to solve".

But as it was said clearly, by most of the users here, in case of of this new axial mate "the only problem is not being able to set the default".

Well maybe not only problem. Second problem is that Autodesk is not able to react and just give us this option within a week. What would just make everyone happy. 

Instead they are "reacting internally", and are "limited by law".

Just give us the option to set the default. It is not difficult at all. You only need to make dialogue window to remember default, this is not hard to da at all.

It does not need testing in beta, it does not need NDA, it does not need analysing.

It is just brainless get it done thing.

So why you just refuse to make simple things simple?

Or expose this to API and some user will do that for us.

Cris.

[Anonymous]

I still don't get how you are satisfied with constraining pipes and elbow that can still be moved along axis.

I'm not saying that directed axis constrain is useless, just haven't had oportunity to be in situation when I would make use of it. I've read an argument someone uses it when constraining cylinders, but even in that situation I would prefer to make the cylinder either fixed or at least limit its stroke with minimum and maximum limit in mate/flush constrain.

[jletcher]

@SteveMDennis

 I don't like this remember last as the fix.

In fact it is causing more headaches,  now if I want a axis constraint but flush was the last thing I did but I go do a axis I don't find out till after I go try to get the axis at that time now I have to go back to the UI change it then I have to click my axis.

Please put this back how it worked in 2012 keep your options but make undirected the default and remove this remember last it makes thing 100 time worse.

I don't care to much for this remember last, makes a nightmare in the hole feature as well.

Please listen to the users please.......

[mcgyvr]

---

@jletcher wrote:

@SteveMDennis

 I don't like this remember last as the fix.

Please listen to the users please.......

---

I do.. 

As a user jletcher doesn't speak for me here

[jletcher]

You are also retired and no long use in a production mode, you would like anything I am against. just saying

[Mark.Lancaster]

@mcgyvr

You retired?  How does @jletcher know this and you didn't even tell me..