load a line in a block for structural analysis

[rprovideo]

Hi
I have a design I want to test. Its an aluminium block with several holes that are used to divert lines (ropes). I have some questions regarding the analysis in Fusion. Please excuse me if these are beginner questions, i am a beginner in this area. Tried to watch a few tutorials, but they only cover  their own designs.

This is what I have. Eventually, I want all the holes have ropes thru it, but I started with a single one.

1, I would use a dyneema rope, they can take up to 12 tons. I loaded this with 20kN, so 2 tons. Since there is no dyneema in the library, I used stainless steel, and it failed (stretched). Now I am using titanium, but it still streches a bit. Not a big deal, but is there something wrong with my model? I am not interested in the rope streching, ideally, i would set this material to infinite or something.

2, I fixed one end of the rope, and added the load on the other end. Also added the screws holding the block down. But if I dont add any more constrains, the rope just comes out of the hole:

now I can fix the middle part of the rope, maybe only on the x direction, and then it would stay. But part of my study is to find if the rope stretches the hole open (if thats the right word) and how much. If I fix the contacting part on the rope, that may never happen. Actually, since the rope is a bit smaller than the hole, if I fix it, it does not touch the side of the hole at all, and (I suppose) no forces will be calculated on my actual item.

3, Would it be better to have the force act on the actual inside wall? Like this

But this is not really accurate, most of the 2-ton load is on the rope (and whats on the other end), the block is actually taking the side load from the angle the rope is diverted. This is much bigger than the actual load (this is me tying the rope in the hole and pulling 2 tons, that will never happen, this is a diverter.

[rprovideo]

Okay, I got the extrusion. Thanks.

[TheCADWhisperer]

@rprovideo 

Now Rectangular Pattern the top Hole Face as shown below...

For a simple design like this it would be easy to pattern the sketch circle - but that is not parametric.

It is almost always better to pattern features (faces) in preference over sketch elements.  Especially as things get more complex.

[TheCADWhisperer]

@rprovideo 

Now pattern the bottom hole in one direction as shown below...

Mirror the bottom hole Faces as shown.

I should have noted that I always select the Origin Axis/Planes for direction in making the patterns rather than the part edges/faces.

The Origin geometry cannot be deleted so is more robust technique but the part could be edited causing a loss of an edge or face so that would be a less robust technique.  This modeling technique about the Origin is called the BORN Technique - Base Orphan Reference Node.

[rprovideo]

Thats a good point, thank you. 
I have all the holes.

[TheCADWhisperer]

Now add the 30mm Fillets to the top corners...

It is usually best (but not always) to add Fillets as placed features rather than as sketched elements.  Keep sketches simple.

Now add the 2.5 mm Fillets.

[rprovideo]

I think that line should be 24mm not 28mm to be symmetrical...

[TheCADWhisperer]

One of the benefits of history-based parametric modeling is that we can go back in time to make edits.

Drag the Timeline marker back to before we did the Rectangular Patterns.

Add the Fillets as shown (do as 5mm - we will change to 7mm later).

Drag the Timeline one step forward to the first Rectangular Pattern and right click Edit Feature.

Ctrl Select (CMD on Mac) the two top fillet Faces.

Repeat for the bottom

Edit the Mirror and add the missing Fillet Faces.

OK, this was a lot of work.  It might have been better to simply add all of these fillets at the end of the Timeline rather than going back in time. (In this case, in other cases it would make sense.)

[TheCADWhisperer]

---

@rprovideo wrote:

I think that line should be 24mm not 28mm to be symmetrical...

---

Where did I write 28mm symmetrical - post screen shot of my instruction.

(Shift Win S and Ctrl v in Reply on Windows, not sure on Mac how to do a screen capture - anyhow I do not see where I made that instruction.) I can edit the instruction once I see where I made mistake.

[rprovideo]

In here

but due to the wonders of Fusion, I just edited the sketch to 24mm dimension and now it looks perfect.

[TheCADWhisperer]

---

@rprovideo wrote:

In here

but due to the wonders of Fusion, I just edited the sketch to 24mm dimension and now it looks perfect.

---

We'll say I did that mistake in my instructions on purpose to test you. Yeah, that's the ticket.

🙄

[rprovideo]

good job 😄 

[rprovideo]

actually, 28mm is whats on my original drawing, but it then works with 7mm fillet and 28mm spacing in the pattern

[TheCADWhisperer]

@rprovideo 

In your original file there was an unresolved issue highlighted in yellow on the Timeline for this fillet.

Always STOP and resolve these issues when they first occur.

---

@rprovideo wrote:

actually, 28mm is whats on my original drawing, but it then works with 7mm fillet and 28mm spacing in the pattern

---

In the new file right click on the Fillet feature and change to desired radius.

Note the change in the geometry.  Make any other desired edits to match your Design Intent.

Attach your new file here.

I am going to be away from my computer for a bit.

My next set of questions are:

Q1. Why are you running this analysis - what do you want to know? (Give some deep thought to the answer.)

Q2. What is your definition of failure for this part? (Give some deep thought to the answer.)

[rprovideo]

yeah...I knew that will come around and kick me...Thank you

I added a hole with thread by selecting the origin point, then used a pattern on the feature to make the 5 mounting screws in the bottom

[John_Holtz]

Hi @rprovideo 

The file "R11 v16.f3d" has no simulation setup, so I do not know what is wrong with the analysis. But it sounds like you are making the model more complicated. Hopefully you are simplifying it in the simulation. After all, it is better to get results that are approximate than it is to get no results with a more accurate model.

This forum post has become more about modeling than the simulation. If you still have simulation questions, I suggest you start a new forum post so that readers interested in the simulation can find the information. (Be sure to add my using "@ my name" so that I know you created a new post.)

John

[rprovideo]

Thanks John. I'm not sure @TheCADWhisperer will continue, but I was able to do some simulation with his help. I have some questions, but as per your questions, I'll start a new thread.

[TheCADWhisperer]

@John_Holtz 

I intend to carry on this discussion into simulation once a robust model is created.

I assume @rprovideo will make changes to the model based on the results of the analysis and a robust - well defined model will help in that endeavor.

Next we can discuss idealization of the model geometry for the purpose of analysis.

I would keep all of this discussion within the same thread.

I will ping you when @rprovideo provides his definition of failure (or success).

[rprovideo]

Thanks @TheCADWhisperer. I did what you instructed to. I did meddled around with some simulations, but I am happy to see how you'd continue and learn some more things. 

[TheCADWhisperer]

---

@rprovideo wrote:

 I'll start a new thread.

---

@rprovideo 

I recommend that you do not start a new thread.

If you do, at least link it back to this thread.

[TheCADWhisperer]

---

@rprovideo wrote:

Thanks @TheCADWhisperer. I did what you instructed to. I did meddled around with some simulations, but I am happy to see how you'd continue and learn some more things. 

---

What is your definition of success/failure of the model analysis?

(I intentionally did not include information on this as I don’t want to bias your thinking.)

Why are you doing the analysis - what do you want to learn about the model and why?