Community
Inventor Forum
Welcome to Autodesk’s Inventor Forums. Share your knowledge, ask questions, and explore popular Inventor topics.
cancel
Showing results for 
Show  only  | Search instead for 
Did you mean: 

Wow Creating An Assembly Is Nearly Impossible

30 REPLIES 30
SOLVED
Reply
Message 1 of 31
drauckman
2425 Views, 30 Replies

Wow Creating An Assembly Is Nearly Impossible

I went through a bunch of tutorials on how to draft and assemble parts and when going through them step by step everything looks really neat and works well.  Start with blank files and try creating something from scratch and that is when the laws of the universe no longer apply.

 

I am sure I am doing something wrong here but the problem is when logic and common sense no longer apply and each mouse click and drag is bound by some law that cannot be documented let alone repeated that is when I start getting lost.

 

My assembly is mainly comprised of pipes, clamps and pipe T-joints which are joined by rotational joints.  I am sure this has to be a tricky assembly task since I have witnessed numerous times my assembly flying apart in every possible direction into a tangled mess from just trying to nudge one part.

 

For example -> Currently I have a part bound by a single rotational joint and I ground the pipe this joint is connected to.  I have used this many times now to rotate parts relative to one another to prevent the arbitrary explode and re-arrange that occurs if you do not ground a part.  Now all of a sudden when I try to rotate this part I get a circle with a cross through it beside the mouse pointer telling me that this will not happen.  Why now?  For a sanity check I went back to a previously saved file and the same scenario works.  Why?

 

The best part of this software is the: "The assembly cannot be solved" dialog that pops up as soon as anything gets more complex than a single joint.  This dialog fascinates me since after it pops up and I cancel it I will spend a half hour trying to ground parts and nudge the pieces closer together.  Then I try the exact same joint and...... it WORKS!  It blows my mind every time.  So is there a internal timer that ensures you spend an appropriate amount of time screwing around before it will let you pass to the next level?  If I can ground/nudge the two pieces I want to join in about a half hour why can't the software do this, maybe I am missing something but I thought this was the point for software like Inventor?

 

Is there some tips and tricks I am missing.  I am new to this and I do kind of approach it from a this is how I would try manipulate it if it was in front of me.  I see a part that needs to be rotated so I try click and drag it into place, of course Inventor has no clue that the part I click on could possibly be the one I want to move so every piece not grounded will fly off into random directions.

 

Since I have been through the tutorials that come with the software and some others that I have found on the web is there any other resources to help figure out how to get this software to do what you want?  I could only imagine some poor soul that tried to pipe a plant or something in inventor, you could probably prototype it in real life a dozen times before you could get Inventor to solve an assembly with more than five pipe joints in it.

30 REPLIES 30
Message 21 of 31
JDMather
in reply to: drauckman


@tonofsteel wrote:
 

I sketched out the corner plate in AutoCAD and brought it into inventor. 

 


I recommend NOT using AutoCAD as your sketcher - soooo much easier in Inventor.

I recommend you put this project aside for a day or two and go through these -

 

http://home.pct.edu/~jmather/SkillsUSA%20University.pdf
http://inventortrenches.blogspot.com/p/inventor-tutorials.html
http://wikihelp.autodesk.com/enu?adskContextId=HELP_TUTORIALS&language=ENU&release=2014&product=Inve...


-----------------------------------------------------------------------------------------
Autodesk Inventor 2019 Certified Professional
Autodesk AutoCAD 2013 Certified Professional
Certified SolidWorks Professional


The CADWhisperer YouTube Channel


EESignature

Message 22 of 31
JDMather
in reply to: drauckman


@tonofsteel wrote:

So from what you are telling me I cannot quickly throw together a concept?  I

 


You?  I don't know.  But I can, because I first learned how to use the software before trying to do my own 1st concept designs.

 

Inventor is a professional program and deserves (requires?) a professional level of preparation.


-----------------------------------------------------------------------------------------
Autodesk Inventor 2019 Certified Professional
Autodesk AutoCAD 2013 Certified Professional
Certified SolidWorks Professional


The CADWhisperer YouTube Channel


EESignature

Message 23 of 31
JDMather
in reply to: drauckman


@tonofsteel wrote:
 

But if I did go and fix the dimensions and U-bolts I dont see how that is going to change the assembly. 


I am quite certain I could quickly assemble a concept using the not correct/not finished parts, and then adapt later, but I know how to use the program.

 

I recommed that as a beginner you build a disciplined rock solid foundation (look at the trouble you are already having with this simple assembly) before free-form conceptualization.

 

(just look at the title of your thread)

this is easy stuff once you have some experience.


-----------------------------------------------------------------------------------------
Autodesk Inventor 2019 Certified Professional
Autodesk AutoCAD 2013 Certified Professional
Certified SolidWorks Professional


The CADWhisperer YouTube Channel


EESignature

Message 24 of 31
drauckman
in reply to: JDMather


@Anonymous wrote:

@tonofsteel wrote:
 

But if I did go and fix the dimensions and U-bolts I dont see how that is going to change the assembly. 


I am quite certain I could quickly assemble a concept using the not correct/not finished parts, and then adapt later, but I know how to use the program.

 

I recommed that as a beginner you build a disciplined rock solid foundation (look at the trouble you are already having with this simple assembly) before free-form conceptualization.

 

(just look at the title of your thread)

this is easy stuff once you have some experience.



Wait, what is wrong with my assembly again?  Is it that the plate is not dimensioned and the U-Bolts dont fit correctly?  This is a simple assembly with the files I uploaded showing one connection that could not be made with no apparant reason why it should not be made.  All the joints and constraints are there to hold the shape that I want.

 

So what would you do to quickly assemble this concept?  I get that you are telling me I need more training and don't know what I am doing (true), but for this scenario what is it that I am doing wrong?  The training shows parts getting assembled and I am using the same techniques.

 

You did get me thinking though, I don't care at ALL about that plate and the U-bolts.  I don't care that they are not dimensioned, nor that they dont fit quite right.  I guess this is why I am trying to get past the part where you are telling me to design them further, I am concerned about the stuff that is going to be inside this frame right now.  Is it going to fit?  Will it have to be changed to a square?  Will it have to be based on a right triangle?  I don't know so I don't want to waste time designing something that has a good chance it is never going to be used.

 

So assume that I handed off this design challenge to another person.  What are the requirments at this point?  That all the pipes connect together at 120 degrees to one another all in the same plane.   Done, I no longer have to think about if this is going to be achieved using plate and u-bolts, a 2x4 with holes drilled in it, a 3D printed part, or even a custom injection molded piece.  Does that mean that I have to wait for another team to come up with a part that I can import, inspect for textbook conformity and then continue designing?

 

I think this is the part that I am missing, I don't care about that detail at this point at all, so why am I including it?  I started with a concept in my head, drafted the plate, extruded it based on the angle requirement.  I found the U-bolts would not work so I created joints and constraints to hold it where I wanted it for the time being.

 

What I should have done (yet to be confirmed, but from training videos looks possible) is create constraints based on the requirements and not even include the plate and U-bolts at all.  Then whatever solution is found can be dropped in.

 

I think the biggest thing is that there are too many remaining degrees of freedom at this point.  To make that last pipe connection every part in that assembly would have to move and/or rotate except for the one grounded plate.  This is too much for Inventor to resolve.  If I used my requirements above and constrained the joints between the pipes then they would be locked in place at the right distance and angle from one another so assembly would be much easier.  As it stands they can still slide inside the U bolts, I am sure if I constrained them all at a fixed offset so they could not slide and use a work plane to align to top and bottom I could fix this assembly.

 

Maybe I am missing something but your solution to this is to spend a bunch of time working on the plate and U-bolts and the rest of the problems will go away?  I don't understand why it matters that the U-bolt is not touching the pipe, or the holes don't line up.  The software does not care about this, I have constraints in place that should make these points moot.  Right now you can drag whatever you want wherever you want, the software does not understand how something looks, it is all held together by rules (joints and constraints).  The pipes always stay concentric with the u-bolts and tangent to the plate becasue there are constraints in place to hold it there.  Moving it around does not change the size or shape of the plate or the location of the U bolts, that is not the problem at all.

 

Now that I have read some more I am almost certain that the problem is that there are too many remaining degrees of freedom.  There should be work planes to align the top and bottom triangle pieces, and the pipe corners should be constrained to one another based on the requirements.  This is because the actual physical joint and the orientation of the pipes only have the contract of having a certain angle between them while lying in the same plane.  Both parts can be designed separate from one another as long as this contract is adhered to (and the pipe diameter).

 

 

 

 

Message 25 of 31
drauckman
in reply to: JDMather


@Anonymous wrote:

t
I am quite certain I could quickly assemble a concept using the not correct/not finished parts, and then adapt later, but I know how to use the program.

 


This is exactly my question though, how?  To make it easy I included the files and a description of what I am trying to achieve.  All the training shows examples but I am asking how for this very specific scenario.  (assemble the concept using the not correct/not finished parts as included)

 

To make it even easier just make that one connection in the assembly between the pipe and T joint.

 

I think that the answer to my question is that while building up the concept I left too many degrees of freedom open.  Now when I try to make that joint there is no way for the software to know exactly how I want to rotate / move every part to make the connection.  Some of the design requirements should have been added as constraints.  (I know the angle, I know the top and bottom have to be aligned with one another, the ends of the pipe will be 1 inch away from one another at an angle of 120 degrees etc)

 

The simple answer is that there is no gravity (or even friction for that matter) in Inventor so you cannot assemble things like you would imagine doing it on a bench.  You have to look at everything from a rule/constraint perspective.  If there is something that is known for sure then you need to include it as a constraint.  I know the ends of the pipes are going to be a fixed distance apart, yet I did not include this so they are able to fly off into all directions.

Message 26 of 31
JDMather
in reply to: drauckman


@tonofsteel wrote:
 

The simple answer is that there is no gravity (or even friction for that matter) in Inventor.....


There is gravity and friction (and far far more) in Inventor, but first I recommend you concentrate on the most basic principles.  Properly constrained sketches.  Are they a requirement? No, but I have found that fastest users are have developed very disciplined techniques.  On the other side of the coin - I have tried to train a lot of undisciplined users who never seem to get it.  They want to argue with me at every point.....


-----------------------------------------------------------------------------------------
Autodesk Inventor 2019 Certified Professional
Autodesk AutoCAD 2013 Certified Professional
Certified SolidWorks Professional


The CADWhisperer YouTube Channel


EESignature

Message 27 of 31
drauckman
in reply to: JDMather


@Anonymous wrote:

@tonofsteel wrote:
 

The simple answer is that there is no gravity (or even friction for that matter) in Inventor.....


There is gravity and friction (and far far more) in Inventor, but first I recommend you concentrate on the most basic principles.  Properly constrained sketches.  Are they a requirement? No, but I have found that fastest users are have developed very disciplined techniques.  On the other side of the coin - I have tried to train a lot of undisciplined users who never seem to get it.  They want to argue with me at every point.....


No gravity or friction when you open that assembly and use the assemble commands, sorry.  Maybe in simulation or some other tools but all the pieces dont fall down until they hit the ground, and when moving parts around there is no friction, they all slide freely equally well.

 

No offence but you have provided zero insight into anything other than:

- I need more training

- You are good at what you do

- You are able to do amazing things because you are Inventor zen

 

Which I pretty much knew already (but I did learn that you are good).  Other than that reading over your replies there is no insight to anything.  No answer to the question "how?" or "why?" in any way.

 

Although I will make mistakes I will not only get this assembled in Inventor I will actually build it to your dismay.  I will not be able to get it to assemble in its current form but in an hour or so.........

 

I get the training part but you also need to learn by doing.  I watched videos and read training material now I need to start getting things done.  You can read, watch and listen to all the training you want, it is not going to make you proficient in real world situations.  Its unfortunate that we can't just read a book and be an expert, it takes experience.  Am I going to look back on this assembly and notice a million things that I should have done different ten designs from now?  Heck yes.  But I will also have ten designs worth of physical prototypes, and ten designs worth of experience, not just a pile of books, classroom receipts and browser bookmarks.

Message 28 of 31
JDMather
in reply to: drauckman


@tonofsteel wrote:
 

I get the training part but you also need to learn by doing.   You can read, watch and listen to all the training you want, it is not going to make you proficient in real world situations.  Its unfortunate that we can't just read a book and be an expert, it takes experience. 


Yes, that is what I am saying.  Get experience with the very basics (like fully constraining sketches and not accepting fasteners and holes that do not line up simply because you are still in concept stage.  Once you have more experience then you can assembly things that aren't perfect.  (BTW - I had 20 years of shop floor experience including 8 yrs as machinist before I started using Inventor).

Examine the attached assembly - but I would not really do it this way as it is too much work.  Maybe it will give you some insight.

As you get more experience - look into Skeletal (master) modeling techniques and the Frame Generator.


-----------------------------------------------------------------------------------------
Autodesk Inventor 2019 Certified Professional
Autodesk AutoCAD 2013 Certified Professional
Certified SolidWorks Professional


The CADWhisperer YouTube Channel


EESignature

Message 29 of 31
JDMather
in reply to: JDMather

Examine the previous example I posted before looking at this one.

 

I started out with the intention of showing you how the Frame Generator technique would automatically adjust to changes in the skeletal master file, but then I discovered that the profile size you are using is not already in the Frame Generator library.  I could explain how to add it, but that is another paper.....

 

Even without the Frame Generator solution - this skeletal master technique demonstrates easier design (and especially editing as design changes).  I didn't know what would make more sense for your controlling dimensions - so I entered driving and driven (reference) dimensions.  These could be reversed.

 

Open the attached assembly.

Use the Insert constraint to place the 24" tubes in the Tees.

Use the Component Pattern command (with the browser in Modeling View) to pattern the exisiting components (select the first in the browser - hold the Shift key and select the last in the browser to select all).

 

(You might only do the triangle side and Mirror, then add the cross tubes).

 

Now examine my constraints - delete and see if you can reproduce (very easy).


It is far more impressive if we had actually used Frame Generator.

Research how to add profiles to the design accelerator.

If using Frame Generator - I would have only place 2 Tee connectors, mirrored, use the Point-to-Point placement method (to circle centers inside fitting) and pattern.  Any (logical) changes to the skeleton would update the assembly.


-----------------------------------------------------------------------------------------
Autodesk Inventor 2019 Certified Professional
Autodesk AutoCAD 2013 Certified Professional
Certified SolidWorks Professional


The CADWhisperer YouTube Channel


EESignature

Message 30 of 31
drauckman
in reply to: JDMather

Interesting, thanks JDMather!  I did not get to the attachments in the second post yet but the first one with the frame built I have been going through and I am learning alot just by looking through it.

 

I thought AutoCAD was good for creating sketches because that is what I am used to using for drawing 2D profiles.  However now that I see how you made the sketch and after going through some more tutorials I see that there are some ways of approaching it so that it is not so messy or busy.  When I brought in the sketch from AutoCAD and tried to dimension it there was a mess of details.

 

I opened a new part file and created the same sketch you did from scratch to get a feel for the steps that would need to be taken to build it up.  It is a diferent feel and workflow than what I am used to in AutoCAD, but once you get used to it a bit it works good.

 

One tutorial I watched had a part that was mirrored similar to mine but they did it all in the sketch.  I was thinking about this for awhile and I am wondering about the pro/con of doing it either way.  The main thing I could come up with is if you do everything in the sketch and then at some point later in the design you find out you need to add a tab/hole/slot etc. on only one side of the mirrored part that it would be easier to add to the extrusion by modifying the sketch directly?  You could do it using other tools I think but from what I see I would re-use design by taking the sketch as a starting point for creating another similar part.  But I guess it doesn't matter since you could just take the part as a whole and modify it in the tree.

 

I have noticed there are a few ways of achieving the same result and from all the things I read/watched it seems that there is some room for personal preference on how to approach solving the problem?  Of course some ways require less steps but others may be more intuitive.

 

I am going to look into your last post and do more research on the topics you mentioned.  Thanks for the posts, in the end it was the best help to be able to compare the two assemblies to see how it could be done better.

Message 31 of 31
JDMather
in reply to: drauckman

Over the years a number of people have started threads very much like yours - see

http://forums.autodesk.com/t5/Inventor-General/IS-INVENTOR-REALLY-USEFUL/m-p/1332811#M144094

  -Walt Jaquith  Especially read response #8 and every response by Walt in that thread.

#8 is probably the most significant post I have ever seen on this forum.

Every Inventor user should be required to read #8 before the software will launch.

 

"The master of constraints is the master of parametric design software."

 

It is almost always better to pattern (mirror is a pattern) features rather than sketches.

Sketch patterns also pattern the constraints which stresses the constraint solver (read - more likely to fail on edits, and who does everything correct the first time not requiring any edits?).


-----------------------------------------------------------------------------------------
Autodesk Inventor 2019 Certified Professional
Autodesk AutoCAD 2013 Certified Professional
Certified SolidWorks Professional


The CADWhisperer YouTube Channel


EESignature

Can't find what you're looking for? Ask the community or share your knowledge.

Post to forums  

Autodesk Design & Make Report