Complicated Loft problem not working

[ianhughes7UFVF]

Hi

I am using a loft to try and create a rather difficult shape. I original worked on one side and mirrored it but it comes out a little rough. So I decided I wanted to create a loft joining four sketches, but it requires the rails to be sorted better as the top needs to be flat but at the moment it turns out bumpy and the part looks like a love heart. Not the intended look.

This is how the part looks with two lofts:

As you can see the center portion is nice but the edges are obvious joined bits. I want this all in one.

This is how it comes out trying to make a loft using the four profiles:

Here is my sketch:

I assume its my rails that are the issue.

I would greatly appreciate advice here.

Regards

Ian

[TrippyLighting]

Trying to build this in a single loft is probably not going to yield the result you are looking for.

Can you provide an image of how this is supposed to look like ?

[ianhughes7UFVF]

Hi

Here is what its aiming at:

As you can see rather difficult. It I have to stick with the version I did thats fine, but hoped it could be done in one big loft.

Regards

Ian

[TheCADWhisperer]

First thing I noticed is underconstrained sketch geometry, missing dimensions, tangent...

[davebYYPCU]

Two Rules of thumb that work for me, when investigating how the Fusion Loft will work,

All Rails must connect all Profiles,

All Rails must have G1 or Tangent segments, where Profiles don’t need this condition.

Your heart shape top comes from not using the top rail, and more likely not Tangent connections to the curve at each end.

Might help....

[ianhughes7UFVF]

Hi

I started to think about the top rail but wasn't sure how to draw it. I tried a front curve but not sure if its meant to be a straight line.

Regards

Ian

[TrippyLighting]

Here's one attempt that offers a lot of flexibility on creating such objects.

It creates a quarter panel based on three simple sketches and then mirrors and stitches this into one body that can then be thickened. You can easily use this to create 4 different quarter panels by using more simple sketches to define the rails.

[ianhughes7UFVF]

Hi

Thanks for all the directions and it sounds as though what I am after is pushing the limits somewhat of a loft.

So with what I have is making clearer rails like I have made in red lines going to help, or would I need a fourth one like the green one to achieve it.

All would it still not work.

Regards

Ian

[TrippyLighting]

---

@ianhughes7UFVF wrote:

Hi

Thanks for all the directions and it sounds as though what I am after is pushing the limits somewhat of a loft.

So with what I have is making clearer rails like I have made in red lines going to help, or would I need a fourth one like the green one to achieve it.

All would it still not work.

Regards

Ian

---

The limitation is not in the loft, but in your modeling skills.

You are using the wrong technique, it's that simple.

Did you read my post at all (the one right before your last post) and understand how the attached model was constructed ?

[davebYYPCU]

Try again, the top red line does not reach all profiles, and

the green line does not reach all profiles.  

In the example you provide Loft will work for the 3 inner profiles only, with all 4 rails.

There is an optical illusion on the right hand end of the sketch, if similar to the left hand end, then it can be half, with mirror.

Might help.....

[ianhughes7UFVF]

Hi

I realize my skills are not good, and I am struggling all the time with this. So I apologize for that.

I did see you post. I find that seeing a screencast works the best for me. Maybe I am thick, but I need to see something stepped through to be able to copy it and then I can do it.

I initially made the waist in one quarter and it worked really easily. But mirroring it produced too many angles. As you saw from the picture of the original, its a very rounded shape but needs flat top and the sides tighten up.

I did look at your picture, but it didn't seem to help with what I am trying to do. Sorry.

Regards

Ian

[TrippyLighting]

---

@ianhughes7UFVF wrote:

I did look at your picture, but it didn't seem to help with what I am trying to do. Sorry.

Regards

Ian

---

In that case, as Iv e recommended to you a number of times, you need to NOT continue but take a step back and try to understand basic concepts. I've seen numerous of your threads and participated in some.

You can continue to follow step by step instructions serving you a solution on a slower platter but following these without really understanding why a particular workflow works and another does not will leave you exactly where you are right now with no progress.

I had attached a model to my post so you can step through the timeline and see step by step how this is done. If you still cannot understand how that is done the you are trying things way beyond your level and - again, you need to take a step back.

The model you created does not reflect the geometry in the photo run posted. I personally would do this with a T-Spline but I would not even begin to show how that would be done before you grasp some surface molding techniques

[TheCADWhisperer]

---

@ianhughes7UFVF wrote:

I realize my skills are not good, and I am struggling all the time with this.

---

This is a complicated project.

Can you Attach examples of some of your easier projects that you completed (simple prismatic and cylindrical geometry) in practicing skills leading up to this complex project?

It might be useful to take a step back as @TrippyLighting  suggested and have the experts here diagnose some of these easier projects.  I have found that in diagnosis - it is often observed that some "poor" techniques need to be "unlearned" and replaced with more robust, logical and predictable techniques - even in these easier projects, before tackling a complex project.

You have free access to the equivalent of thousands of dollars worth of expert help - standing by and waiting to help.

The first thing they will be checking is to ensure that you are not attempting to build a "house-of-cards" on a foundation of quicksand.

[ianhughes7UFVF]

Hi

It sounds like I am annoying everyone here at the moment. I sincerely apologize for this as this is not my intent.

My mistake is I have jumped in the deep end and been asked by fellow hobbiests to create projects that are extremely hard to do.

I have learned alot from all the advice. Six months ago, I was so inexperienced that I had no idea how to use lofts, project, use sweeps, and a whole host of things, that I feel alot more comfortable using now. I am extremely messy with my projects and that I appreciate is my problem.

I don't seem to have any simple projects unfortunately. They are all hard, which is again half the problem.

So once again, sorry guys, I really don't set out to be annoying, just trying to learn, the hard way it seems.

Regards

Ian

[TheCADWhisperer]

---

@ianhughes7UFVF wrote:

Six months ago, I was so inexperienced that I had no idea how to use lofts, project, use sweeps, 

---

Forget Loft, Sweep, Revolve, Extrude... ...for a moment.

First you should be careful to fully define each and every sketch element.

Blue, unconstrained geometry should keep you awake at night.

Rather than asking a question about Loft, ask first how to fully define the sketch.

[chrisplyler]

It goes wonky if you try to do it all at once because of your blue (unconstrained) corner rails. I have circled the parts that look tangent, and put arrows where they obviously are not tangent. The arrows at the sides don't matter too much, but the arrows at the top cause the top surface not to end up flat. You can imagine now why when you make it with a single loft, the top surface doesn't end up flat, right?

Instead of arcs in those upper corners, I would use a spline (or a conical curve, if you feel you can get a curve you like better) that is set tangent to the straight top. In this way, the loft's side surface should flow smoothly into a flat top surface.

And the lower side circles LOOK tangent to the eye, but they're still producing a slight variance in the surface around the bottom of the hole, so you should constrain those arcs tangent to the underside line also, forcing them to be exact.