MORE alignment/joint conundrums...

[ipmcc]

I feel like 90% of what I post here is about how I can't align parts, but here we are again. I have a device with a tilted face, and then another component that will slide onto it, at that angle. F3D attached.

Consider this design:

 

The interesting view might be this: 

There is (or should be) 0.25mm/2 = 0.125mm as a buffer between all these planar/parallel faces.

These faces are not congruent to one another -- there is a cut in the rails of the cylinder. How can I align the sliding tube, cuts and all, to the rails it will slide onto? It should be placed perfectly between the two 'rails', and the 'rails' (on the stand) should sit between the two 'guides' (on the cylinder).

Align & Joints show all the standard points, but those points are not useful due to the cut in the ridges. Similarly, I can painstakingly align two faces to be coincident, but that's not what I want. I want them to be 0.125 away from the other.

I swear: Aligning things that are designed to actually be produced (i.e. they need to have a small amount of space between them) feels dang near impossible in Fusion. I have been using this package for over a year now, and I still just cannot wrap my head around how to align parts in real-world production-ready ways. I've tried Align, Joint, As-Built Joints, and Constraints and I just haven't been able to figure out how to make any of those features do what I need in situations like this. Usually, I just end up eye-balling it and calling it 'good enough' but that's crap.

Should I be modeling 0.125mm 'shims', doing face-to-face joints against those shims, and then hiding them? Nothing in real life is 100% flush, so this seems like it should be something that virtually everyone needs to do. Even if I were willing to accept face-to-face align/joint between the rail and the parts they slide in, I still haven't been able to make it work in the part. I'm at my wit's end. Is this possible?

Regards,

Ian

@ipmcc - this post has been edited due to Community Rules & Etiquette violation.

[kacper.suchomski]

Hi

You can optimally use joint and constraints interchangeably.

[TheCADWhisperer]

@ipmcc 

Ruthlessly simplify - The BORN Technique (Base Orphan Reference Node)

The Origin Planes exist before all else and cannot be deleted.

[ipmcc]

First? Thank you! Very helpful indeed.

I can see when you zoom in at the end that there does appear to be the desired gap, but I'm confused as to why/how. The last constraint you added in the video looked, to me, like it should join the front surface of the faceplate to the inside-front of the slot on the cylinder, but it doesn't seem to do that. On several re-watchings, it looks like that last constraint is actually constraining the angle between the faces (to be 0°) but not making them flush. Interesting...

I think what's troubling me about this approach is that these relationships feel effectively coincidental. In other words, they achieve the end goal, but in a way that feels non-intuitive to me, and that does not express design intent. It uses the base-plate as a 'bridge' of sorts -- these joints and constraints largely (except for the last one, discussed above) describes the other parts' relationships to the base-plate and less to each other.

Using what you posted as a starting point, and bearing in mind your comment that, "You can optimally use joint and constraints interchangeably.", I went through and recreated the relationships you showed with all constraints, and then again with all joints. Joints were much more fiddly in terms of UI, but it was possible, in the end. But again, all solutions were effectively predicated on referencing the base plate, and never once did I seem to have the chance to enter the offset values that describe the most fundamental relationship in the model. Again, it achieves the desired result, just not in a way that captures the design intent.

If I need to intermediate these relationships using other relationships to the base plate, it weirdly feels like using invisible shims and flush joints would paint a more clear picture of design intent. But then there are a bunch of shims in the design.

I suspect I'll need to spend some more time figuring this out.

Thanks again,

Ian

[kacper.suchomski]

---

@ipmcc wrote:

I can see when you zoom in at the end that there does appear to be the desired gap, but I'm confused as to why/how. The last constraint you added in the video looked, to me, like it should join the front surface of the faceplate to the inside-front of the slot on the cylinder, but it doesn't seem to do that. On several re-watchings, it looks like that last constraint is actually constraining the angle between the faces (to be 0°) but not making them flush. Interesting...

---

In the video you can see that the first constraint (axis) is a mate constraint and the second is an angle constraint - this is symbolized by the icon in the dialog box. The angle constraint does not add adjacency between faces.

---

@ipmcc wrote:

I think what's troubling me about this approach is that these relationships feel effectively coincidental.

---

My order wasn't random. It's a result of my familiarity with these tools.

---

@ipmcc wrote:

In other words, they achieve the end goal, but in a way that feels non-intuitive to me, and that does not express design intent. It uses the base-plate as a 'bridge' of sorts -- these joints and constraints largely (except for the last one, discussed above) describes the other parts' relationships to the base-plate and less to each other.

---

Much of design is counterintuitive. Design relies heavily on logic. Logic and intuition are two different states. Relying on intuition in the design process is often counterproductive. We should read instructions, learn, practice, and use tools according to their intended purpose, not by intuition.

---

@ipmcc wrote:

Using what you posted as a starting point, and bearing in mind your comment that, "You can optimally use joint and constraints interchangeably.", I went through and recreated the relationships you showed with all constraints, and then again with all joints. Joints were much more fiddly in terms of UI, but it was possible, in the end. But again, all solutions were effectively predicated on referencing the base plate, and never once did I seem to have the chance to enter the offset values that describe the most fundamental relationship in the model. Again, it achieves the desired result, just not in a way that captures the design intent.

---

Connections and bindings have different capabilities and are designed for different scenarios. The only way to learn them is through study and practice.

Both commands have some intelligence—they recognize potential solutions; but it takes some learning to understand what these tools can predict.

A natural consequence of multitasking tools is their complexity. There are commands that have a single purpose; but there are also those that offer multiple possibilities - and consequently, require conscious manipulation and control.

---

@ipmcc wrote:

If I need to intermediate these relationships using other relationships to the base plate, it weirdly feels like using invisible shims and flush joints would paint a more clear picture of design intent. But then there are a bunch of shims in the design.

---

No, you don't have to—it's a natural workflow to constrain all the components; usually in the order they would be assembled in the real world. But there's nothing stopping you from constraining the small face of the angled plate to the flat face of the cylinder instead of to the large face of the base plate.

---

@ipmcc wrote:

I suspect I'll need to spend some more time figuring this out.

---

Yes, but do it consciously. Abandon intuition and focus on studying theory and practicing with examples and exercises.

Good luck

[TheCADWhisperer]

@ipmcc 

First Principle - any object in space has 6 initial Degrees of Freedom (DoF). 

Three translational and three rotational.

As we start to assemble components we are removing DoF in a way that can be logically documented and duplicated.

In the virtual CAD world one strategy is to use the BORN Technique where the Origin is a Base Orphan Reference Node.
This reduces the assembly instructions to as simple as possible.

In the video I mention the importance of understanding DoF at a first principles level for assembly Constraints/Joints in Dynamic Simulation of mechanisms in motion.

Here is one example Planar Joint (with Contacts).

[TheCADWhisperer]

https://youtu.be/2ro-JzWUbQY