Flexibility not working properly in inventor - BUG that has been there for ever

[Cris-Ideas]

Hi,

I am trying to use flexibility feature but every time I give it a chance it fails me. There is a bug somewhere in inventor that makes it not capable of properly solving assemblies with flexible components.

This time I have run on to this issue with a very simple assembly, had been able to reproduce this buggy behaviour  tens of times, and had made video for you.

After unsuccessfully trying to post this I thought I will make another video for you so you could clearly see what is the difference when assembly is flat.

Here it is.

Here I have uploaded data set for you to play with :  http://a360.co/2fmTsvD

And in case you also think this is not working properly you can support idea to fix this: Fix flexible assemblies !

Cris.

[Cris-Ideas]

in case anyone thinks flexibility is working

# 220 bug

NONSENSE! and happens only because of flexibility

Also cross link to other thread with constrain solving bug that do not involve flexibility

https://forums.autodesk.com/t5/inventor-forum/constrain-solving-as-complicated-as-that/td-p/8172353/...

Cris.

[Cris-Ideas]

flexibility is not handled even in simplest cases.

It is just not working.

# 221 bug

Cris

[johnsonshiue]

Hi Cris,

Like I replied to you in the email, please suppress Angle:3 and it will work. The PosRep did not trigger the movement because the constraint was failing. The bug here is that the constraint error is not properly shown.

Many thanks!

[Cris-Ideas]

@johnsonshiue 

Hi Johnson,

Angle 3 is perfectly legitimate and meaningful constrain.

What is actually failing here is constrain solving in inventor. And it is not only on 2019 but this case is already logged for AI 2020.

I have already explained what Angle 3 does but I can do it again:

Angle 3 is there to ensure clamp is positioned so inclined pipe is on the out side of the main girder.
Without this kind of constrain 4 theoretical solutions are possible with this kind of clamps:
1) inclined pipe is on the outside in the bottom and inside in the top
2) inclined pipe is on the inside in the bottom and outside in the top
3) inclined pipe is outside
4) inclined pipe is inside

So if I add a constrain that fixes rotation of the clamp along axis of the horizontal tube than only one solution becomes valid. (the one I want).
I used angle to be able to flip this alignment easily if so desire, also Angle constrain takes off exactly those DOFs I want to fix.

If you examine geometry of elements participating in this piece of construction  you will see that if you are human it is obvious how this solution is possible. So If solution is obviously possible software should be able to find it.

Cris.

[johnsonshiue]

Hi Cris,

You have reported these cases to Product Support. Then you posted them again on the forum and sent emails to various Autodesk employees. Though these are good issues, the amount of time spent on duplicate efforts is not making sense to the product and the team. It is hurting our ability to help other users' issues.

Like I mentioned before, some of the issues are bugs, some of the issues are limitations, and some of them are related to the way you constrain the components. If these issues block your progress, I suggest you try other solutions. I am very sorry that we are simply not good enough for you unfortunately.

As long as you have access to Subscription Support, please leverage the support and the team can provide status update on any of your reported issues.

Many thanks!

[Cris-Ideas]

Hi Johnson,

Yes in deed I am doing all of that.

As I understand this: problems will be solved only if right people understand they are this are the problems. So I make everything I can to reach as many people as possible. Sorry for trying to be effective. But public/user pressure on a corporation is only thing that works. It is not my fault Autodesk is not doing enough to solve problems and makes its customers mad. 

As for the Angle 3 I have replied to you, and also I have replied here in public as what you say about Angle 3 has not sense.And since you blame me for most of the problems I have with constrain solving and constantly insist most of them are caused because of how I build my assemblies I will always argue with that.

Is you have any serious arguments to prove or show exactly what is wrong with constrains I define please present them.

I am always ready to analyse and implement changes if this is the case.

But as so far this story goes I do not recall a case of this kind.

And as I see it it is in my and other users best interest so you finally fix problems related to constrain solving as this is relevant for all users.

I have just tried very simple assemblies in AI 2020, and you know what: they fail, and fail all the way. Inventor does not even accept constrain sets that are obliviously correct. (you know the cases for sure).

And I am sure you are fully aware of the fact that this kind of software behaviour is not only "not good enough for me" but for many other users as well. So trying to make this problem look as minimal as possible by writing things "I am sorry we are not good enough for you" you can only fool your selves as people experiencing this problems understand their importance and extent.

Have you had a look on development forum thread related to flex , pos reps and adaptivity? What people are saying there?

"I hope you will finally address this issue"

"we tried to use flexibility but it turned out not stable and we dropped it finally and we are now discouraging our employees to use it"

......

.....

If you have anything related to Angle 3 please explain.

If you wish to continue political and personal debate I would rather  propose you did no do that, unless you want to waist time.

Hope you are still "here to help as long as I am Autodesk customer"

Cris.

[Cris-Ideas]

# 233 bug _ AI 2020

Cris.

[Cris-Ideas]

# 239 bug

Basics.

# 239 bug model

Cris.

[Anonymous]

Hi Chris and everyone,

I have experienced this pain with flexibility and adaptivity since back in 2013, and my personal solution to the issue has been to abandon their use. It's a pity, because if they worked like they should they would be pretty handy tools, and that's why with every new AI release I give them a try, but no, there's something very messed up.

At the beginning I thought it was my fault, that I did not understand well the feature, but now I know it's not the case. In my personal opinion I won't even call it a bug, as I think the problem is conceptual.

In the very simple case of an hydraulic cylinder with just one DOF (piston - rod sliding), if instead of putting it in the parent assembly, one puts the piston and the rod as separate non-flexible assemblies, everything works just fine. From a mathematical point of view there is absolutely no difference between the two situations, so it's difficult to understand why the "solver" does its job in one case and not in the other: just explode the flexible subassembly in a hidden fashion, recreate the constraints at the parent level, done.

Maybe it's a "structural" flaw deep inside the "proprietary" algorithm even since it's beginnings decades ago, and so difficult and/or expensive to fix, but really, Autodesk, there are thousands of engineers and designers that spend hours trying to make something you say it works but doesn't. And they end up with clumsy workarounds, as for example in the cylinder above, building a position rep with a certain stroke suitable to actuate a component in a certain position, instead of putting the component in that position and let the software calculate the stroke for you.

But, except for this and other small problems, great software, my compliments! 🙂

And my compliments to Chris too, for his yearly stubbornness and all the effort put in the topic!

[Cris-Ideas]

Hi,

in case you missed my other thread I would like to present to you something I did on the way fighting with flexibility.

This is theoretical solution for solving 3D Cartesian assemblies of rigid bodies.

Basically it proves there is (theoretically) no problem with solving any properly constrained assembly, na matter how many free DOFs it has.

So in my opinion there is no justification for constrain solving in inventor to be so buggy. Amount of money companies and people like me (that pay directly for their software) pay during years of using this software really entitle us to have software that actually does what it is said it does.

So if anyone is interested how any 3D assembly can be solved please read attached pdf.

There is example solved. The one from the beginning of this thread. The one inventor is not capable of solving.

Any feedback welcome.

Cris.

[Anonymous]

Hi Chris,

I'll have a look to your pdf, just give me a couple of days. It is certainly going to be very interesting, but also certainly nothing new. In fact, as I mentioned in the previous post, Inventor CAN ALREADY DO IT with non-flexible assemblies, and you must agree that the concept of flexibility doesn't complicate matters in any way...

But maybe I'm missing something, let me read the pdf first.

[Cris-Ideas]

Hi,

Flexibility should not add any difficulty for the solver, this is totally my point.

But there are many cases when inventor is not capable of solving assemblies with no flexibility and no pos-reps.

I have presented some examples here: "constrain solving -as difficult as tat"

So problem with constrain solving in inventor is bigger than flexibility alone.

Good and easy to follow indicator is DOFs analysis that clearly exposes how ridiculous results  inventor gets from constrain solver.

Cris.

[Cris-Ideas]

# 239 bug AI 2010

This was said to work in AI 2020

Clearly it does not.

So I think other cases I was tolled they work in AI 2020 probably also have issues.

We are still no where near flexibility being usable.

Cris

[ggreivulis]

Hello,
3 years have passed and still flexibility does not work properly. I have used it multiple times. In one dimensional movement it had worked, but as multiple mates are added it breaks most of the time only when placed in assembly as Cris had shown in videos. It is easy to duplicate these situations.

This problem had annoyed me for 2 years. I can work around it, but each time when I give it a chance I am disappointed in software.

I have 2 suggestions:

1. If Your support team can not make flexibility stable, then remake it from 0 or add a big danger sign "flexibility is not working properly".

Constrains are crucial part of designing, so please fix it.

2. And if you did 1. then create option to add constraints which motion range can be added, example angular constraint from 0 to 90degrees, distance from 10 to 50mm. Could been used for bolting, assembly opening closing parts etc.

I can imagine it would be hard to create and may be unstable, but it could be a good asset to Inventor, if You suceeded.

[Ray_Feiler]

"2. And if you did 1. then create option to add constraints which motion range can be added, example angular constraint from 0 to 90degrees, distance from 10 to 50mm. Could been used for bolting, assembly opening closing parts etc."

You can do that. Just expand the Edit Constraint dialog box.

[Cris-Ideas]

AI 2020 speaks for it self in terms of flexibility.

still no progress in basics.

Cris.

[Anonymous]

You made my Friday better Chris! 😁 .......if they never get this fixed then we will continue to be entertained at least! 

I for one will miss your input if the flexibility issue is truly rectified......

[Anonymous]

***update***

Never mind, I see I need to dive into this more.

Joints can be used and should be used in the sub assembly, but then it's the action of flexibility you need in the upper levels...

[johnsonshiue]

Hi Cris,

Many thanks for sharing the issues! We did try addressing the Flexible, PosRep, and Constraint related issues on 2020. But, we ran out of time and we could not find good solutions without causing more harm. You are on Inventor Beta. There will be a Beta build coming out soon. Please try it out. I believe quite a few issues you have mentioned on this thread and other threads have been addressed.

Thanks again!

[nannerdw]

Has there been any progress on this bug? Even the simplest flexible assembly fails to update until I click and drag on one of the parts.