Trouble with joints

[yoshimitsuspeed]

How would I get the wheels in this model to roll inside the groove?
Basically I need to emulate gravity pulling them against the contact sets. The problem is that joints always need to refrence a joint origin. This isn't the first time this has been a major inconvienence to me. So I can't figure out how to use a joint to do what I need. It sure would be cool if we could have applied forces as an option. Basically same as simulation where you could choose to apply gravity or x force to the ladder in your chosen direction.

[TrippyLighting · Accepted solution]

Looking at your model, I am not sure what you intention is.

I would assume that the bearing blocks are installed fixed on the latter, basicallty with the same vertical distance as the horizontal rails ?

Are you only hoping to show horizontal movement of the ladder along the rails ?

[yoshimitsuspeed]

Correct, the bearing block is attached to the ladder. The wheel in the bearing block will roll in the groove in the rail.
My hope was to use joints and contact sets to hold the wheel in contact with the groove so I could push it back and forth without any vertical play.

[jeff_strater]

There is really no "gravity" in Fusion.  You'd need a dynamic simulation product to really get that.

However, there is "rest position" in a joint.  I sometimes use that to simulate a spring.  And, if combined with Contact Sets, you can get close.

Here's a screencast:

Jeff

[yoshimitsuspeed]

Yes I know this and it's convienently simple when you are joining two blocks with two points coincidentally in the right location to associate.
I am asking how one would apply that to my model or any other situation where things aren't so perfectly laid out.
Or if there are any other methods of achieving the objective.

[yoshimitsuspeed]

My point though was that it would be much simpler if we could just apply a gravatational pull in a sutuation like this instead of needing to figure out how to make the joints do it and wasting a ton of time trying to figure it out.

[TrippyLighting]

I've taken a good long look and your assembly and have restructured it. Trying to simulate the physics of this asembmly is not what you want to do if I understand it correctly. You just want to see the sliding motion of the ladder along the rails.

You've diligently first created components and then activated them with the exception of the first one. I've tried to roll back the timeline to correct that but ran into troubles. However, with a littel workaround you can still correct that. 

I simply rolled the timeline back to after the first combine action and deleted everyhing after it.

Then I deleted the empty component that you forgot to activate whe you first started the design and seaved the design as "ladder". Then I can re-import it into another design and have a nicely structured component.

Your assembly does not have many parts but still, it's not a bad idea to determine what can be assemblies and subassemblies. This will keep the Browser lean and helps navigating it.

I created a Ladder assembly that contains the ladder component and a Bearing Block assembly. the bearing Block assembly contains the Bearing Block and the wheel.

Lok at the attached firle and see if that makes sense. Agan, you don;t have to do it that way - I don't always do this either - but in  experience a desings approag the stage of completetion, they keep accumulating parts.Nuts, Bols, etc. 

As such it heps to think a littel ahead in terms of what canbe an assembly and subassembly etc.

I deleted als the joints that you had in your design and completely re-assembled it. With the exception of one joint I choose to make all other joints rigid. Unless you really want to simulate the rolling motion of the wheels ther is not reason to give it a revolute joint. Giving it a rigid joint is actually of advantage in this design as you'll see in a moment.

The only moveable joint you really need to simulate the horizontal ovement of the ladder is one sliding joint.

Here is where it helps to use a rigid joint for the wheel as I am using it's profile sketch to create the sliding joint betwen the ladder assembly and the top rail.

I greated a little screencast on how to greate that joint as it's easier to show than to explain.

The re-structured assembly is also attached for yu to study. 

[yoshimitsuspeed]

Thanks for the input. I hadn't tried to do sub assemblies si I'll remember that in the future. THis was just meant to be a quick and dirty example that I could send some pics and a render or two to a customer to show them how something someone else had designed in another CAD program would look rendered in F360 so I definitely could have spent more time on layout and design had it been an actual working project.

Do you have a link to that video? I am interested in many things. Like how you placed the bearing blocks with a rigid joint in their locations and not just origin to origin.
Also curious about how you did the slider joint although this definitely seems like a sub optimal workaround. It seems like it should be pretty easy for f360 to do this properly with spinning bearings and as realistic as possible. In this scenario it isn't really any disadvantage having them not spin but it still seems a little ghetto to need to do it this way.

Thanks for everything you did though and I would love to see exactly how you went about it.

[yoshimitsuspeed]

Ah cool, I just found where you had placed the joint origin. I hadn't tried playing with that so I didn't realize you could place them refrenced off of other origins.

[TrippyLighting]

Here is a screencast that explains e few of the joints.

[yoshimitsuspeed]

Awesome thanks

[yoshimitsuspeed]

I always hate it when I get a notification that something has been accepted as a solution when it's really just a partial workaround. It makes me feel like AD is going to look at this and think that this is an acceptable solution that no longer needs to be addressed.

[TrippyLighting]

I'ts not a workaround, but a practical solution that requires littel computing resources. You'll learn to apreciate that when it comes to more compex assemblies.

If it provies comfort to you, I don't think this cane be done with Solid Works, or Solid Edge either. While mating works dirrferntly there it would be similar in function.

[yoshimitsuspeed]

It is a workaround. One that will work well enough in this application however it still has it's shortcomings. It should still be possible to use a radial joint on the bearings so they function as they should. 
It should still be possible to have the bearings rest on the survace of the track and follow it. For example the track would likely be slightly bigger than the bearing. In this model that little variance wouldn't matter but that still doesn't make it right and in some situations it would matter.
It seems like it would be ideal to use contact sets in this application and just have a good way of holding the part in the groove.

[kb9ydn]

---

@TrippyLighting wrote:

I'ts not a workaround, but a practical solution that requires littel computing resources. You'll learn to apreciate that when it comes to more compex assemblies.

If it provies comfort to you, I don't think this cane be done with Solid Works, or Solid Edge either. While mating works dirrferntly there it would be similar in function.

---

You could do something like this in SWX by setting it up as a motion study.  There you can simulate gravity and physical dynamics of objects as they interact.  I'm not even remotely an expert on motion studies but I've played around with them enough to know that the vast majority of the time it's not worth the effort to set them up (to me anyway).  They get really complicated really quickly and if you don't get everything constrained just right, your parts go where they shouldn't or worse the simulation won't solve and you can't figure out why.

This particular example looks simple enough that it could be done without too much work, but then because it's so simple, what would be the point?

I have often wished for some kind of spring mate though in SWX.  There are times when it would be useful for example to have two planes or faces remain parallel to each other but either attract or repel.  This (in theory) should help with simulating basic types of assembly motion.  It wouldn't simulate any sort of physical dynamics of objects (like mass or collisions); it would only solve motion based on the mates applied.  I think with just this additional feature you could do a decent job simulating things that currently are nearly impossible, like roller chains.

C|