Tribulations of an Inventor neophyte

[Anonymous]

A couple of weeks ago I reached the forum for help with some revolved or sweep “solids” that did not have a surface.  Later on, I posted that I had solved my problem by creating a “tool” where I could place the offending parts and construct my paraboloid mirror.

One of you asked me why a newcomer choose such a difficult task.  I replied that was the problem that I needed solving. 

My experience has been frustrating.  While the first iteration worked:

• I could not replicate the pattern because it lacked an axis (later on I learned that I could create an axis at the intersection of two working planes);
• I had to redo the first iteration, and I wanted to have it fully constrained without relying on grounding pieces, and I spend days moving the parts and trying to constrained them, with thousands of “inconsistent” constraints;
• I ended up putting a ball and a half a sphere (at the two ends of the piece) but failed to remove the inconsistency, even when looking at the connections they seem to be OK and the measure of the distances of the axis was less than 0.5 mm (the part is about 1.5 m – so the tolerance was 0.0003) until it dawned to me that the ball/sphere leaves 2 degrees of freedom and by using a pair, I was removing all freedom (which is what I wanted), so the thing was inconsistent because I was using an expensive solution, so I accepted the inconsistency.  Nothing that in real life could not be solve with a light tap of a hammer, but I did not want to tap into the monitor!
• I am now creating the second set, and I am frustrated at the amount of time it takes to position the parts properly.

The books and the help can only teach you so much.  I have duplicated some examples and I have created simple parts to test hypothesis.  The forum is great, but searching on it takes a while to get something useful.

I am getting better, but I am also running out of patience.

[Xun.Zhang]

Hello @Anonymous,

Agree that learning something new is always difficult. However, have you played with tutorial a little bit?

In addition, some great video in youtube may helps.

It is hard to follow with specific ones in your end, it's better to have a full learning path before handling some real project.

Thanks!

[SBix26]

I'd be willing to bet there's a good robust Inventor solution to your problem.  Any chance you could post your files and show/describe what you're trying to achieve?  What version of Inventor are you using?

---

Sam B
Inventor Pro 2019.1.1 | Windows 7 SP1
LinkedIn

[johnsonshiue]

Hi! Do you mind referring to prior threads you posted about the issues? Like Xun and Sam indicated, there should be a solution to the problem. However, there are indeed some limitations in Inventor preventing certain geometry from being created. Without seeing an example, it is impossible to tell.

Please share the files here. I would like to understand the issues better.

Many thanks!

[Anonymous]

A couple of days ago I placed a message in the forum called “Tribulations of an Inventor neophyte” and I wanted to thank for the encouraging words.  I am not complaining.  I am not reporting a bug.  I have spent countless hours watching videos in Inventor and Utube and trying different approaches, sometimes creating simplified parts just to see how Inventor behaves, or whether I can attach something else to the part that could help.  As far as I can tell my problem stems from the fact that revolved and/or sweep parts do not have a face and the constraints are less useful then.  There ought to be a better way of dealing with Inventor.  I am enclosing several files for those that had requested them, all of them in “paraboloid.7z”. 

I am trying to construct a paraboloid mirror, 4 m of radius, by building three concentric circles, each formed by six mirror segments, with each mirror segment having an area of about 2 m2 (about size of current automotive glasses).  A structure with aluminum rods will provide the needed rigidity.  I am using Inventor 2018 Pro.  I am doing it for a class at Duke, attempting to show some undergraduates students ideas of what needs to be done.  I am not doing it for production purposes. 

The picture shows one of the attempts that I have used to create the first layer, just as an illustration (it is not yet finished).  I need to create two more concentric layers.  Each mirror segment ought to have about 17 pieces, but some are duplicates.  The file called placement.iam has most of the parts in the proximity of their intended position.

Mirror1.ipt is a segment of the mirror for the first layer.  It is a parabola revolved around the axis for 60 degrees.  It starts at 400 mm and ends at 2111.8 mm.  I have trimmed the mirror using an offset of 25 mm, to provide room for the structure in between the mirrors.  The mirror has several perforations to attach with a piece similar to Edge1.ipt which would attach the mirror segments to the structure.  Problem No.  1 – I cannot attach the pieces to the mirror.  The “Insert” constraint does not find the perforation.  It can find it with Mate, but not with Insert.

The structure is formed by a pair of parabolic rods (file Para_bar1.ipt) and two circular rod segments (Circu1 and Circu2.ipt), created by sweeping their respective profiles.  The rods are inserted into four boxes (Flat_box and Flat_box4.ipt) forming a structure where the mirror is located.  Problem No.  2 – it took me forever to get the structure in place and in shape, having to modify offsets and moving the segments to be in the proper orientation, angle, etc.  I grounded one component, place pieces in the correct orientation, ground them, the insert other pieces, move them around until proper orientation, ground, insert another, move, twist, reposition, remove inconsistencies, etc.  etc.  I even attached a straight bar to the parabolic rods to be able to visualize the angle and orientation of the parabolic bar.  I changed the diameter of the rods, thinking that maybe the curvature conflicted with the “flat boxes”, but that does not seem to matter. 

Decided to create a piece (the “tool”) – and placed some small balls on it, attaching half-spheres on the flat boxes, to ease the placement of pieces together.  It was easier, but keep on having inconsistent constraints until I figure out that Inventor was complaining that by having two ball-joint attached to a piece, the “normal” degrees of freedom of the ball-joint were removed.  I accepted the “inconsistency and was able to form the structure a bit easier.  The tool was flat and I could not use a circular pattern, therefore, I repeated the operation with another flat tool, but now following the shape of the mirror and then I succeeded in using the circular pattern to form the first layer of concentric mirrors. 

The main problem stems from the fact that the sweep rods do not have a face.  I created square ones that have a face, but using flush and mate does against the mirror depends on where I click in the mirror and if I rotate it, or move it, does not snap back where I want it when I do a local update. 

Thanks again for the interest and willingness to help.

Eduardo

 

kelly.young has embedded your image for clarity.

This is a public forum, please refrain from posting serial numbers or personal information
Please note when replying via email any signature used by your email client will be added to your post as with any other email.

Edited by
Discussion_Admin

[JDMather]

---

@Anonymous wrote:

  I am doing it for a class at Duke, attempting to show some undergraduates students ideas of what needs to be done.  .. 

---

I see numerous issues in your modeling techniques that would indicate to me from past experience that you have not been properly trained in using Inventor.

If I take time to help you resolve these issues - are you committed to learn how to use Inventor?

Yes/No

[Mark.Lancaster]

@Anonymous

Since this is a world wide public forum, never share any sensitive information here such as email address, serial #, subscription account number, business address, and phone/cell #. This is a world wide peer to peer network and such items should not be posted here for security reasons.

[Anonymous]

Thanks for the concern. I will be more careful in future.

[Anonymous]

I have learned a bunch of skills by doing, without having to take classes. I will persevere until I am satisfied with and then I will move to something else.
I am learning to play the piano without a teacher, and although it might be easier with a teacher, now I can read music and more or less hit the proper keys. My timing (or rhythm) stinks, but Çe la vie.
thanks

[JDMather]

I asked for a binary Yes/No response to my question.

I am going to be very frank here, so stop reading if you are easily offended.

Snip....  … never mind, I can't give you my honest assessment in a public forum.

Bottom line, Inventor is a professional program and deserves (requires) a professional level of preparation.

Students deserve a professional level of instruction.

Are you going to be attending AU this year?

[kelly.young]

Hello @Anonymous it looks like the part para_bar3.ipt is missing. In the future the easiest way to share files is with Pack & Go as .zip. 

Are you familiar with how to Derive Parts, create Adaptive Assembiles, or make Multi-Body Parts?

It sounds like you just moved the pieces in space to a close location, using Constraints to orient exactly will be paramount to your designs in the future.

It might help to create a master sketch to define locations for easier setup. If you want to create a screencast showing what you are trying to do that would be helpful.

Please select the Accept as Solution button if a post solves your issue or answers your question.

[Anonymous]

Not offended. No. I am not planning of attending AU. Thanks for the interest.

[Anonymous]

Hi Kelly,
Thanks for the reply. para_bar1 and para_bar3 are exactly alike The difference is the diameter of the sweep. Sorry for the confusion. You can substitute one part for another.
Pack and go produced a huge file because it included old versions and templates (it was 25 MB). Winzip gave me a smaller file.
Yes, I am familiar with multi-body parts, have looked at adaptive and derived parts. The multi-body approach I have used to simplify the number of parts that I have to position. The adaptive option would work fine if I was to change dimensions, but I am not planning on doing it. Derived parts will work when I have to place them in a circular pattern, otherwise I am going to overlap parts.
Yes, the approach is to place the part in the vicinity and try to constrain it. The problem is that some of the constraints do not seem to have a place to work on mirror, which is a revolved part, because I wanted to use the equation. If you look at edge and mirror1, I wanted to use insert in the holes in both parts to constrain them, but the holes in mirror1 are not available for insert, only available for mate. I even tried tangent hoping it will pick the curvature of para-bar, and it can constrained to the place I clicked on. I created a multi-segment parabola and extrude it. It worked for somethings, but I could not cut it at an angle because there were many "uneven" profiles.
Thanks again for the interest and help.
eduardo

[JDMather]

---

@Anonymous wrote:
1. Thanks for the reply. para_bar1 and para_bar3 are exactly alike The difference is the diameter of the sweep. Sorry for the confusion. You can substitute one part for another.

2. Pack and go produced a huge file because it included old versions and templates (it was 25 MB). Winzip gave me a smaller file.

 3. The problem is that some of the constraints do not seem to have a place to work .

---

@Anonymous

1. Parts with different diameter are not "exactly alike".

2. You can Skip the extraneous stuff in Pack and Go. (see image below)

or

2b. You can simply keep the relevant parts/assembly files in a Project Folder and right click on the folder to send to compressed (zipped) file.

3. Assembly constraints mimic the real world - so if modeled correctly - the assembly constraints should all be logical.

3b. To follow will be a series of videos on how to correctly and efficiently model this assembly.  Interaction will be required to achieve a successful outcome.  Back in a bit with Video #1 - Sketching 101, the Very Foundation of any MCAD Software.

[Anonymous]

JDMatter,

1.Thanks for the info about pack and go.  I obviously did not read all the screen.

2. Agree with two parts with different diameter are not identical.  I misspoke.  The difference in diameter does not affect the performance.  With a precision instrument, the two flat boxes should be slightly curved to accept the circular segments (circu1 and Circu2) and also the different curvature of para_bar, which is flatter at the lower end (where the showing work plane is) because the rate of change of the curvature of the parabola changes with the location.  I thought it might be part of the problem and I changed the diameter of para_bar to try to see if it made any difference.

Thanks again for  the help.

eduardo

[Anonymous]

JDMather,

First of all my apologies for misspelling your last name in my earlier reply.

I have watched your videos and I can see the difference between some one who knows  and someone that is learning by doing.   Your elegance and simplicity is shocking to me.  Just the trick of showing the origin and symmetry are great.  But the most important lesson was that I do not need to create sketch after sketch, but that I can sketch on the part. 

I really want to thank you for the effort you are putting on my problem.

regards,

eduardo

[JDMather]

---

@Anonymous wrote:

JDMather,

First of all my apologies for misspelling your last name in my earlier reply.

---

No problem there - I suspected that your first language isn't English.  In any case, my spelling isn't that great either.

You might want to correct these.

  I will create additional videos for your design as I find time.

Also, I hope that you discovered that there is sound narration with the videos - turn on your sound.