Join components as if welded

[juliangall]

I am modeling a stair balustrade so that I can use simulation to determine that it will resist the necessary forces.

There is a handrail (circular pipe) at an angle (going up the stairs) and a number of vertical posts (square pipes).

The posts are instances of a component as they are the same. I want to fix the posts rigidly to the rail so that I can apply forces in my simulation. My problem is I have not modeled any joining surfaces between the posts and the rail. When the balustrade is constructed, the tops of the posts will just be welded to the rail. All I want to do is have Fusion 360 recognise that there is a rigid joint where my post and rail components intersect.

Is there an easy way to do this or do I need to create my "weld" in Fusion 360 so I can make the joint?

Thanks,

Julian

[jeff_strater]

It's a bit hard to tell without the model or a screenshot, but my first recommendation would be to create the rail and the posts in place, and then just use a Rigid Group to make them all rigid with respect to each other.  Then, just ground one of the components, and the whole structure should be rigid.

This is from the modeling point of view.  Or, are you asking about simulation here?

If you can share the model, you'd get plenty of suggestions.

[juliangall]

I tried creating a rigid group but this didn't work in simulation. You can see from the screen-grab that my post isn't formally joined to the rail. This isn't necessary for the visuals of the design or for instructing the person who will make it. I just want to lock the two components together so they act as a single structure in simulation.

I can't just cut the top of the post using the rail as that would require having more than one component for the post - the rail is sometimes horizontal so the orientation isn't the same with each post. 

I can't cut the rail to make sockets for the post as I may need to move the posts, or add more to adjust the forces.

Thanks,

Julian

[TrippyLighting · Accepted solution]

---

@juliangall wrote:

I tried creating a rigid group but this didn't work in simulation. You can see from the screen-grab that my post isn't formally joined to the rail. This isn't necessary for the visuals of the design or for instructing the person who will make it. I just want to lock the two components together so they act as a single structure in simulation.

I can't just cut the top of the post using the rail as that would require having more than one component for the post - the rail is sometimes horizontal so the orientation isn't the same with each post. 

I can't cut the rail to make sockets for the post as I may need to move the posts, or add more to adjust the forces.

Thanks,

Julian

---

The joints found in the Assemble menu are purely used for assembly modeling and have no effect on FEA (finite Element Analysis) simulation.

To simulate this properly you either need to model the proper geometry so the area under stress can be properly meshed by the software, or define a contact between the components involved in the simulation. For example you can define a ring contact for a bolted connection. I am no specialist for simulation, but I'd guess for your case there might not be a contact type available that would reasonably well approximate the situation. Maybe @Anonymous can help out.

In general, however, I feel you are missing out on a couple of key concepts, in Fusion 360.

If the geometry for each post is different because the actual joint between these 2 components is different in each case then you are correct that you need a different component for each post,  because it is a different part.

However, that does not mean they cannot share common base geometry that can be modified very easily for each component.

The trick here is to create an empty component and then copy paste the body from the base component into it. That creates a new copy paste feature in the timeline. Everything. you do to that geometry in that new component happens in the timeline after the copy/paste feature will be unique to that component. If you want to make changes to the base component, then travel back in the timeline to before that first copy-paste  operation and make the change. That change will propagate through then timeline to all copies..

If you cut the rail using the posts, then that also creates a timeline entry. If you then move the post, that will move the cut in the rail. 

The above assumes you are using a timeline based design.

[Anonymous]

As @TrippyLighting has said, joints don't really have any impact on the simulation workspace. It is contacts that you care about.

There are no welded contacts in f360 but you could use a bonded contact. Some people will literally model in a fillet weld bead because you will run into a few issues:

1 - there will be a stress concentration at the intersection between the joints so safety factor will show low. Those of us that have done it for a long time know when to ignore them and when they are a real problem.

2 - different weld types will transfer stresses differently. Especially if it's a stitch weld or something similarly inconsistent.

Regardless, for a good simulation you need accurate geometry to properly represent the system stiffness. I recommend making the model match the cut and contact of the real part and then using a bonded contact. Sometimes it's acceptable to accommodate a gap like this with an offset bonded contact (or a pinball region if you know ANSYS), but this is not one of those cases.

[juliangall]

@TrippyLightning thanks for all that information. Very helpful.

The key information I was missing is that simulation does not use joints. That wasn't obvious to me - I assumed the whole model was used in the simulation. It appears that a bonded contact is what I need, so I'll experiment with that.

And thanks very much for the brief tutorial on components and timelines. More experimentation needed with that too.

Thanks again.

Julian

[juliangall]

Thanks @Anonymous . Your reply really helps fill in my understanding.

In my example the force is applied horizontally to the rail but from a structural point of view, the critical part is at the base of the post, where it bends most. My "weld" holds the rail to the post the but the force will never be enough to break it before the post fails at the bottom. That's why I just wanted an unspecified rigid joint.

Following your advice and that of @TrippyLighting, I'll experiment with rigid contacts.

Apologies that @TrippyLighting gets the Accepted Solution. Both your posts were really helpful.

Julian