TFI - CADTips Challenge

[TheCADWhisperer]

@Neil_Cross posted a fun challenge over on his channel this morning.

I just modeled this in Inventor and then gave a few minutes of thought to how I would model it in Fusion 360.

It is pretty easy to model in Inventor, a few minutes of thought is all I needed to decide that it wasn't worth my effort to attempt to model in Fusion, but I would love to see the technique used by someone with more Fusion 360 experience than I.

https://www.youtube.com/watch?v=kc5hpLR8edQ&t=153s

[TrippyLighting]

---

@Neil_Cross wrote:

.  Can you not measure the second angle in the Autodesk Viewer? 

---

I've not been able to measure anything in that viewer. Not in Safari or Chrome on MacOS or Chrome and Edge in Win 10.

The Angele can be calculated based on the dimensions already. The unbent X and Y dimension and the radii should all thats needed. @chrisplyler provided that formula in an earlier post. 

[michallach81]

Thx, I forget about the model in the viewer.

Nevertheless, I don't intend to redo it as curved, but I'm looking forward to the next challenge.

Good job with the challenge Neil, good job!

[michallach81]

My bad guys. I was so confident with me knowing the premises of Neils creation that I didn't proceed with any checking after I've done it.
The thing that made me wonder was missing angle on Neils drawing.
Now I know that I reversed the order of bending.
New entry:

Flat didn't change from the last.

[Neil_Cross]

So that was done entirely in Fusion 360? 

Is it a SM part? 

[michallach81]

Full Fusion360, I use SM for bending, but to say it's an SM part is too much to say. After both bands I've used Boundary Fill tool which result in regular body but parametrically driven by SM.

[daniel_lyall]

@michallach81  It would be interesting to see how you did your one considering how far off I am in mass.

[michallach81]

I can make a screencast after challenge 001 is over, how about that?

[daniel_lyall]

That would be perfect. 

[TrippyLighting]

That'll work 😉

[karthur1]

When I was modelling it, I noticed that the volume changed when I went from flat to formed.  In real life, that should not happen... so I am still a little confused WHY Inventor does it this way.  

Regardless, I made my submission last night.

Thanks,

Kirk

[michallach81]

Kirk you’re wrong. In principle volume on the inside of bends arc will be compressed, on the outside streched. We can expect that if the bending line is in the middle of thickness those values should balance them self and we should be same as flat. That ofcourse is not quite right, since area of such section is dependnt on circle area equasion which has a square root, so relation is not linear, nevertheless we can find a spot between end that will balance sides. In this case most of the part is below the curve which makes volume of curved smaller the of a flat.

[TrippyLighting]

---

@karthur1 wrote:

... so I am still a little confused WHY Inventor does it this way.  

---

You are using a modeling tool in inventor and what it does is correct and logical. This is a pure modeling effort with no considerations to manufacturability.

[Neil_Cross]

---

@TrippyLighting wrote:

---

@karthur1 wrote:

... so I am still a little confused WHY Inventor does it this way.  

---

You are using a modeling tool in inventor and what it does is correct and logical. This is a pure modeling effort with no considerations to manufacturability.

---

Indeed it's just a modeling challenge, it would open up a santa claus sized sack of worms if I was to take manufacturing (with only 1 correct answer) into consideration given I've got a global audience in all different manufacturing sectors.  So I tried to keep it simple! The flat form was provided on that drawing purely as a means to guide the modeller into how to potentially approach the challenge, and I think it would have been near impossible to work out the pre-bent sizes if all I provided was angles and compressed part dimensions!

I said this earlier but it's worth repeating, it's probs the first and last bent model I'll include in the monthly challenge!

[chrisplyler]

---

@Neil_Cross wrote:

I said this earlier but it's worth repeating, it's probs the first and last bent model I'll include in the monthly challenge!

---

Disregarding the problems mentioned, it has been a fantastic mental exercise.

Not having any automatic bending tools in F360, and attempting to devise an accurate way to achieve it manually, forces us to really, really, really think about the geometry.

I swept an R200mm arc along an R300mm arc to get a surface. Is that the shape that the bent part would take? No. That's the shape of the form you might press the part over. The actual part will line up with the edges of that form only at the four orthagonal side points. The corners will pull in further, because they will bend down and inwards, relative to the part's center point, with a radius of their own. The part's corners (imagining them without the fillets yet) will NOT fall at the corners of that swept form surface. A top view of that swept form surface will be rectangular, but a top view of the bent part will have slightly convex sides, because the corners bend down and in.

I gave up trying to model it correctly pretty quickly. I was short on motivation to overcome the obstacles. I got stuck trying to figure out exactly what radius I should use as a guide rail to translate those corners bending down and inwards. I wondered whether I should imagine all the "vertical" surfaces should become oriented radially after the bend or not, (and I don't just mean each hole, I mean EACH SIDE of EVERY VANE) and decided that was just WAY too complicated. But I did deeply enjoy the mental theorizing / puzzle solving efforts I went through. I would absolutely love to watch a video of you modeling it in F360 once the official challenge is over.

I'm not exaggerating when I say that I am very thankful you posted the challenge, just because of the thinking it got me to do. Whether I'm smart enough that the thinking was on the right path, or so dumb that I was way off base... either way it was fun fun fun.

[TrippyLighting]

---

@chrisplyler wrote:

---

@Neil_Cross wrote:

I said this earlier but it's worth repeating, it's probs the first and last bent model I'll include in the monthly challenge!

---

Disregarding the problems mentioned, it has been a fantastic mental exercise.

.... either way it was fun fun fun.

---

Spot on! 

[karthur1]

---

@michallach81 wrote:

Kirk you’re wrong. In principle volume on the inside of bends arc will be compressed, on the outside streched. .......

---

Exactly, I understand that and I totally agree with you. But what I am saying is that the VOLUME of the part should never change... no matter how we twist, pull or bend it.  I am not wrong about that.

If you want to discuss this more, we can discuss this in another thread. I dont want to high-jack this thread or mess up anything Neil has going with with this months challenge.

Kirk

[TrippyLighting]

---

@karthur1 wrote:

---

But what I am saying is that the VOLUME of the part should never change... no matter how we twist, pull or bend it. 

I respectfully disagree!

[karthur1]

---

@TrippyLighting wrote:

---

@karthur1 wrote:

---

But what I am saying is that the VOLUME of the part should never change... no matter how we twist, pull or bend it. 

I respectfully disagree!

---

If you disagree with me.... go grab a piece of sheet metal laying around in your shop.  something that you can bend with your hands.  Now find some scales and weigh it.  Next, bend it as many times as you want and re-weight it.

Did the weight change????  Of course, it did not change.

[TrippyLighting]

---

@karthur1 wrote:

---

@TrippyLighting wrote:

---

@karthur1 wrote:

---

But what I am saying is that the VOLUME of the part should never change... no matter how we twist, pull or bend it. 

I respectfully disagree!

---

If you disagree with me.... go grab a piece of sheet metal laying around in your shop.  something that you can bend with your hands.  Now find some scales and weigh it.  Next, bend it as many times as you want and re-weight it.

Did the weight change????  Of course, it did not change.

---

I did mention earlier that this was a modeling exercise with a modeling tool. This is not a sheet metal exercise and the tool used does not take real world material behavior into consideration. It does not have to.
The info Neil has provided including the geometry that can be inspected in the 3D viewer makes this absolutely unambiguous and clear.
If you take that piece of sheet metal how exactly would you manufacture the geometry you were presented with in the original project description ? 

[michallach81]

That flat grill did bend in untealistic maner.  This bend tool is a design tool, not a manufacturing one. That’s why it’s follow math principles not mimicing real life.