Flexible Assemblies

[panzilte]

I would really like to know why when one uses flexibility in an assembly why the constraints fail? This is really annoying because one has to open the assebly or assemblies, suppress all the constraints then unsupress them to fix it. Why does this happen

[jlackey4740]

Are you saying that the assy you place in the top level assy has constraints in it. and then when you are placing into the top level, making it flexible, and trying to constrain it in the top level, it gives you an error? So then you have to go into the subassy and suppress the constraints in it?

Just curious, you at the Westinghouse here in Round Rock/Georgetown or whatever(kind of in the middle) Edited by: jlackey4740 on May 11, 2009 9:27 PM

[panzilte]

let's say you have a pneumatic or hrydaulic cylinder. a simple model of a cylinder includes two parts the main body and the rod. when contrained the rod should be able to move in and out of the cylinder body. now you take that assembly and drop it into another assembly and make the cylinder assembly flexible. it might be fine today or tomorrow but sooner or later that single contraint in the cylider assembly fails because the cylinder is flexible. to fix it you have to open the cylinder assembly supress that single constraint then unsupress it in order to fix it

[jlackey4740]

That's pretty odd, we use a similar set up with suspa gas springs, and constrain them the same way, but I've yet to see the constraints malfunction over period of time. The only time I see them act up is when the surface that they are mated to in the top level chnage and that constraint reference is lost, however, I've never had to fix the constraint in the spring assy istself. Are there any aspects of the two assys that are changing over time, size if the rod, cylinder body, the parts that they are being mated to or anything like that?

[panzilte]

other that the position of the cylinder rod nothing else changes... I just posted something here because a much more complex flexible assembly that I am working with is having problems and I needed to vent. I guess I need to get used to the fact that Inventor is a low quality piece (cheap) of software and that I just need to learn to deal with that fact.

[jlackey4740]

Ah, yes, many frustrations. Of course, I've only worked with Inventor, so I have no frame of reference as far as other programs are concerned, but I have my days as well. Well, hope you can figure it out, there are a few threads by Shastu that go into some detail with problems concerning flexible assy, although the problem he describes are different than what you have here, there may be some good advice from some of the other posters there that may help you find a solution.

Oh yeah. The only reason I had asked about the location of the Westinghouse you work at is because I work at a place about 5 minutes from there, so I was just curious. Hope if you get a chance to look at the other threads I told you about you find something that helps. Take it easy.

[Anonymous]

My experience with failures of this type almost always
indicates poor constraining technique. Too bad you feel this way about
Inventor.

--
Dennis Jeffrey, Autodesk Inventor Certified
Expert
Autodesk Manufacturing Implementation Certified
Expert.
Instructor/Author/Sr. App Engr. Tel. (260) 399-6615
AIP 2008 SP3,
AIP 2009-SP1 PcCillin AV
AMD 64 x2 3.0 Ghz, 8GB RAM GeForce 9800GT 512MB

XP Pro SP3, Windows XP Silver Theme

href="http://teknigroup.com">http://teknigroup.com

[ozstang65]

Which is why I have almost entirely given up on mating via part or assembly features, rather I will ensure that each part has work features that can be constrained to. This way, an entire part's features can be blown away (accidentally or on purpose!) and your assemly won't fall to bits.

[RonnyM82]

Thats sweet as but what do you tie the work features down to? I make sure that the first thing in do in any new part is create a work plane for my first sketch, that way i can pick the part up (usign the workplane) and move it whereever i like in 3D space without the sketch failing. This means i have to have the horizontal&vertical contraints turned off, otherwise things tend to turn themselves inside out. A bad habit of inventors i might add.

Maybe thats what is happening with the flexible assy problem???

[harco]

If you are using just axis/axis constraint then there is probably to much freedom in the flex assy.
The rod will occasionally flip over.
A more robust way, as mentioned in another post, is plane constraints. If the rod and cylinder are drawn centred at the origin then two constraints on two of the origin planes will allow sliding without flipping.

It's fun when your ram goes inside out!!

[mcgyvr]

I have seen it occasionally also. Even with something as simple as 4 screws that are a kit made flexible and constrained to 4 holes with the insert constraint only. Upon opening the assembly I will find the screws have all snapped to their "unflexible" state. Usually a rebuild all fixes the issue, but I have had to delete and redo the constraints.

I notice it most when the sub has a grounded component.

[Anonymous]

I have never had an issue personally with broken constraints
in flexible assemblies that I have created. I'm not sure, but it must have
something to do with the workflow used in developing parts and subassemblies.
Developing a good design workflow can take time to gain that experience. I've
just been fortunate to have had 23 years to practice, starting with 
non-autodesk software, then moving to Autodesk's early 3D software in
1988.

When considering how assembly constraints may be affected by
modeling changes, you must remember that a 3-D model is made up of faces. Each
face is assigned a unique ID when it's created. 3-D constraints can only
recognize faces, and attach themselves to a particular ID. If you modify a face
that has a constraint assigned to it, then often but not always, the constraint
will either break, or try to reassign itself to a different face. In other
words, the constraint can get confused and begin to start behaving badly. In
this manner, I have seen flexible assemblies work perfectly up to a point that a
part is modified.

My only advice is that when modifying specific parts within an
assembly where positional representations and flexibility is required, that you
revisit all of the constraints on the specific modified part. It is dangerous to
assume that no constraint relationships have been damaged in the modification
process.

--
Dennis Jeffrey, Autodesk Inventor Certified
Expert
Autodesk Manufacturing Implementation Certified
Expert.
Instructor/Author/Sr. App Engr. Tel. (260) 399-6615
AIP 2008 SP3,
AIP 2009-SP1 PcCillin AV
AMD 64 x2 3.0 Ghz, 8GB RAM GeForce 9800GT 512MB

XP Pro SP3, Windows XP Silver Theme

href="http://teknigroup.com">http://teknigroup.com

[GlennPoorman]

Any way you could attach a simple dataset that exhibits the problem you are seeing?

Regards,
Glenn Poorman (Autodesk)

[alewer]

I'll bite. Attached (I hope) is a simple dataset (2010).

Unzip, open assembly.iam
Activate the "flexible" rep
Click and drag the cylinder
Enjoy watching multiple constraints being violated without Inventor recognizing an error. If you want to really have fun, rotate the cylinder assembly while in flexible mode and see where it ends up, again while thinking all the constraints are healthy.

I can try posting a video if these instructions are insufficient.

I'm running 2010 64 bit on XP 64, Nvidia Quadro FX1700, but I have seen this issue on any system since positional reps were available (R11?).

[mikevick]

New question on flexible assemblies

I am about to create a large assembly and I'm decided whether to use flexible assemblies or not and i was wondering could you advise if it was a good or bad idea.  So the project I'm working on will have a framework as its master assembly. I need to place several modules or sub assembly onto this frame where they will then be constrained. the sub assemblies are just parts in random positions with no constraints. would it be ok to bring in these sub assemblies and change them to flexible and then constrain them to the frame. Or the other option would be to pull in the parts in dividually and constrain them to the frame that way. Any advice?

thanks

[andrewiv]

I opened it up in IV2011 and got it ito work by making the cylinder assembly terminated flexible inside assembly, and making cylinder assembly flexible inside cylinder assembly terminated.  I did this in the master positional representation and then just override the poisitional rep of the sub assemblies in the flexible rep.  Do not override the flexible status.  I have no idea why this works, but it did for me.  I hope I explained this well enough.

[BMiller63]

---

@andrewiv wrote:

I opened it up in IV2011 and got it ito work ...

---

just FYI, you're replying to a thread from 2009

[BMiller63]

---

@mikevick wrote:

New question on flexible assemblies

---

start a new topic for best results, adding to an old semi-related post causes confusion and typically results in poor results for your new question.

Be sure to state your version and provide some screen shots, sample files, etc for best results.

good luck