Component moves relative to assembly when master component is used

[jandyman]

This question is complicated enough that I don't know how to include a simplified model, so I'm including a more complicated example. Although I'm getting better and better at components and assemblies, I've somehow created a situation where when a model includes an external component, one of the components in the included component ends up moving relative to the parent component. I'm sure it has something to do with the joints and origins, but I don't know how to approach troubleshooting, so in addition to a solution, I would appreciate pointers to any training material that will help me approach similar problems more effectively in the future.

I've attached two files - the included component and the including component (although it seems that the including component export may have included the included component as well). If you look at the included component, you can see that the "arms" are in logical positions:

However, when the component is included into the other component, the arms move to an illogical position (seemingly locked into the origin in the original design):

Like I said, I'm sure it's because of something I did with the joints and origins scheme, but I'm not educated enough to understand it quite yet.

[TrippyLighting]

Lets first fix your very first sketch from being not fully constrained!

Blue lines should keep you up at night!

My recommendation would be to fix all not fully constrained sketches before joining anything. It is very easy to destroy a design by  accidentally dragging around on a sketch object even when not actively editing the sketch.

That is particularly important as you're not just modeling but cutting wood/metal based on your designs!

[TrippyLighting]

The structure of your assembly also could use a couple of improvements.

This body in the Master Assembly should actually be its own component and be grounded to its parent.

I think the issue you are seeing is because the Master Assembly also needs to be grounded to its paternt, the top level assembly. 

Also, patterns would drastically help in reducing unnecessary sketch work.

You have 4 arms, and a number of modeling features all at .75 distance.

[jandyman]

---

@TrippyLighting wrote:

The structure of your assembly also could use a couple of improvements.

This body in the Master Assembly should actually be its own component and be grounded to its parent.

I think the issue you are seeing is because the Master Assembly also needs to be grounded to its paternt, the top level assembly. 

Also, patterns would drastically help in reducing unnecessary sketch work.

You have 4 arms, and a number of modeling features all at .75 distance.

---

I'm still struggling with this, and grounding suggestion isn't helping so far. I will continue to investigate, but one comment above puzzles me. How is the Master Assembly not its own component? It can't be independent of the Arms and Axle, as the sketches that make up the foundation of the "housing" body are based on projections from the Arms and Axle. That's pretty important, and central to what I'm trying to do here. I get the feeling that understanding your comment may help me modify the workflow.

As far as patterns, I do use them in the sketches. I haven't used them in the assembly/solid domain, I'll try to learn how to do that. Got a video I could watch?

[TrippyLighting]

---

@jandyman wrote:

---

Got a video I could watch?

---

Here you go. Let me know if my rambling actually helps 😁

[jandyman · Accepted solution]

---

@TrippyLighting wrote:

---

@jandyman wrote:

---

Got a video I could watch?

---

Here you go. Let me know if my rambling actually helps 😁

Thank you! Now that I've got my original problem solved, I will watch it and adjust my workflow.

So as to what was causing my original problem, it was that the Joints between the Arm and the Axle had gotten stuck into the base component instead of the subassembly! And this happened even though the subassembly was the active component when the joints were created. Why? Well, it's a bit of a mystery, but it appears to be related to creating joints "between two planes", where the planes were construction planes, even though those construction planes were defined in the subassembly. As a general discovery, it seems that if you create a joint that references something in the root component, Fusion 360 will silently put the joint in the root component as well. That doesn't completely explain what happened, but I think it is somehow related.

Once I redefined the joints carefully so as not to reference construction planes or anything in the root component (watching out for origins selected accidentally), the joints were created in the subassembly properly, and everything behaves as intended. And I learned a lot about troubleshooting this sort of issue, even though it was time consuming.