Can't figure out how to move components

[Anonymous]

I'm just learning Fusion 360 and I'm finding that the traditional group, copy, paste and move operations are not working like I would expect.

First, I'm trying to move components and they all move together. In particular, I want to move two of them into a curved slot.

I've figured out how to use align to get the piece to rotate the correct degrees, but everything moves. By isolating an object, and selecting "move component" I was able to isolate it, but I'm worried I'm still doing it wrong. My final product looks like:

Second, I have a set of planar joints -- just the pieces I'm cutting out -- joined to a grounded piece of stock. I can move 4 of the 5 components like I expect. The fifth piece moves a group of three components at the same time. I can't figure out how to move this component independently. I've searched the help on this one and can't find it. I also don't understand the "golf flag" on top of the joints. I clearly need to read up on this, but am not sure where to get started.

Here is a screencast of me struggling with both of these movement/isolation issues. Any help greatly appreciated.

[TrippyLighting]

Please share your design. Export as .f3d and attach to the next post. This is easier to show on your model in a screencast.

Also, thanks for posting a screencast!

In short, the piece in your design you are trying to move into position should not be moved into that final, assembled position, it should be joined.

You  don't need to activate a component if you want to move it.

A single click on a "piece" in the  viewport selects a body.

A double click selects a component.

[Anonymous]

Thanks -- I have the file at home and can't upload it here. Any thoughts on the group of planar joints that move together and I don't want them to?

[michallach81]

One more thing I can add to what Peter already sad, on your screencast when you repositioning triad manipulator (Set Pivot), you've forgot to thick the green check mark to confirm the change.

Another thing is that you should play with move command, because there is a difference between moving component and body.

Michał

[Anonymous]

Thanks! I missed that. Any thoughts on the joints that I can't move separately?

[TrippyLighting]

Simply post the file as soon as you get home (Ia mat work at the moment as well 😉 and I'll get to it.

Looking at the browser structure there might be some general potential for improvement and I'd like to show you a way to do that to help you move forward with yur designs without the tool - Fusion 360 - getting in the way and rather aiding that process along very nicely.

Your file is small enough yet has enough coplexity to make a little tutorial from it, provided, of course  I have your permssion to do so.

[jeff_strater · Accepted solution]

Hi @Anonymous,

The main problem that you are encountering (everything moves together) is because of what happens at 0:21 of your screencast:  

You have changed Move from moving components to moving bodies.  What this does is to move the location of the body inside of your "spar (2)" component.  So, it affects all instances of that component.  You probably want to use "Component" move in this case.

This section of the help:  bodies and components is useful in understanding the differences between bodies and components.

But, stepping back a bit:  Personally, I would not use Move for what you are trying to do here, because Move does not establish an associative relationship between components.  So, if your side pieces change you will have to reposition each of your spar components all over again.  Instead, I would try to use a rigid joint for this purpose.  That will establish an association that will update with geometry changes to either component involved in the joint.

Jeff

[Anonymous]

Attached is my file. I was able to figure it out, and watched all the learning videos. Which helped a lot. I also like the idea of using joints. (Even though I can't figure out how to keep some from moving together.)

[Anonymous]

What an excellent answer. I learned so much. You summarized everything very well, and gave me a link to learn more. Wow. Thanks.

[Anonymous]

You have my full permission to use the file.

[Anonymous]

One last question that remains is how to break the link between my joints. I've found through experimentation that this only happens when I land two joints on the same location in the grounded component. It is like I have a 'parent' that moves everything. Any hint on how to fix this would be most appreciated. I've looked through the docs and can't figure it out.

[jeff_strater]

Hi @Anonymous, sorry for the delayed response.  I'm not entirely sure what the link between joints that you describe is referring to.  Could you possibly record a screencast of what you are seeing?  I took a look at the model you shared, and it doesn't have joints in it (since it was before you started adding joints, I think).  If you want to share a more recent version of the model, that would also be helpful.

thanks,

Jeff

[TrippyLighting]

I loked at this desing at the beginning of the week and attempted to make a littel tutorial but threw a few attempts away.
There is so much to be improved on this design.

1. The Main Sketch that drives everything is dangerously underdefined und in many places unconstrained.

2. All components were created by first extruding bodies from this skeleton sketch and then these bodies were used to create components.

3. Nothing in the design is properly joined and there is no base component that is grounded.

I believe the "joint" Lauren is referring to is not a joint. It relates to the two move-body commands later in the timeline. She very likely clicked on a "part" in the viewprt and selected move, which moves the body within the component. As the slat-components were copy/pasted the effect of that move is that all bodies in all slats move at the same time. Hopefully I get around to this this coming weekend.

[fulcrumusa]

---

@jeff_strater wrote:

But, stepping back a bit:  Personally, I would not use Move for what you are trying to do here, because Move does not establish an associative relationship between components.  So, if your side pieces change you will have to reposition each of your spar components all over again.  Instead, I would try to use a rigid joint for this purpose.  That will establish an association that will update with geometry changes to either component involved in the joint.

Jeff

---

That brings up an interesting question about joints, which I have struggled with - How do you make joints between components which are laying flat against each other?

Using the example in this thread, the natural thing to do would be to define a rigid joint between the two flat surfaces (the side of the long board and the flat of the insert groove). However, those geometries do not have joint origins that lend themselves to that. One could create custom joint origins but then it's a matter of creating them exactly at the right place and orientation so they produce the correct component orientation and alignment.

Doing what I am describing is a breeze with "As-built" joints because you can join components but "As-built" joints do not maintain their relationships. For example, if one of the sides of the aseembly is moved, the attached componets will not move with it.

[mickingaamco]

From what I can see in your browser, all of your parts are bodies, not components. Bodies cannot be moved as they don't have their own origin. Convert them to components, then they can be placed where you want them. You will probably want to place them by adding Joints, which is similar to mate, align, etc. in other programs.

[jeff_strater]

this is a really good question @fulcrumusa.  Your point about as-built joint being the obvious way to do this, but having downsides (the joint is depdendent on the position of the components, not the geometry) is a good one.

What I usually do in these cases is either play around with component visibility during the joint command, or use Select Other.  It can be clumsy, but gets the job done for me.

Here is a screencast:

Hope this helps.

Jeff

[fulcrumusa]

Thank you, @jeff_strater. I see what you mean. I did manage to make some progress but then everything ground to a halt. It's easy to join components which have straight lines and flat faces against each other. However, I my case the adjacent features are not normal to each other. I tried playing with various joint origins, angles and offsets but I could not make it work.

I've attached a link to my design - http://a360.co/1O8pcev. The components which I am trying to join are the horizontal extrusions and the 45degree angle brackets. I've also attached a screencast showing the parts involved.

I would appreciate some help on this. It might be nice to also create some videos or knowledge base article on joining irregular parts.

[TrippyLighting]

Lookijg at your desing that are two things I note:

1. Nothing is grounded so everything can float. Every assembly must have at least one grounded component.
2. You use linked components excessively. There is really no need for that. You should only use linked components if it is mature in it's design AND you ever anticipate a minor design change. Unmodified purchased parts don't need to remain linked at all! When breaking the link make sure to do it right after inserting the part, before saving the file.

[fulcrumusa]

@TrippyLighting, thank you so much for taking the time to create that video and explain what's going on! I really appreciate it!

Why does the joint origin need to be create prior to creating the rigid group? Does the rigid group prevent the creation of the origin somehow?

[TrippyLighting]

I deleted the existing rigid group in order to create a new one, this time including the joint origin.

It might be possible to edit an existing rigind group joint and add something to the select, but I am not at my Fiusion station at the moment to be able to to try it out .

I could also just have left the original regid group intact and create an new joiunt between the joint origin and , as, the aluminum bracket.

Either way, the point in this case was to to join the joint origin in some way to one of the parts in that rigid joined assembly.