Fusion
I would suggest you post a complete design so we can analyze it. I've seen VEX assembly designs before and often the structure of the design is insufficient and the performance penalty is the result of less than ideal workflows.
Not really, unless you want to put it on Dropbox and share a link to that. Or you can zip it and try to attach that.
On a mac you should be able to simply pull the .f3d file from the Finder into the respective field in the post.
@kentchad - I'm not sure if this is the problem or not, but since I have run into this before, I thought I'd suggest it. I get that message sometimes, but for me, it happens if I mistakenly use the Insert Photos icon (because I use that a lot):
F3D shows up as not supported, because it is not an image.
Instead, use "Reply", not "Quick Reply", and scroll down a ways until you see this:
and either drag the F3D into this widget, or use the "browse" button to browse to the file, and F3D should definitely be supported by that process.
Jeff Strater
Engineering Director
I looked at the top level assembly and saw a number of red and yellow icons. I ran a Modify->Compute All.
There are 29 warnings and 10 errors in the design. Fixing those alone will increase the performance of the model.
There are other things that can be done, e.g. completely avoiding the use of the position capture feature.
The entire assembly could be structured better into sub abs sub-sub assemblies.
Not all of the sub assemblies would need to have a timeline.
A lot of the square holes in the vex profiles aren't going to be used. I'd delete those to reduce the amount of geometry Fusion 360 has to deal with. Some solid body patterns , for example all the little rolls in the wheels could be replaced with component patterns. Again, reducing the memory footprint.
@TrippyLighting wrote:
The entire assembly could be structured better into sub abs sub-sub assemblies.
Not all of the sub assemblies would need to have a timeline.
So what do I mean with the above ?
For example this little bearing block assembly consisting of:
the bearing block, 2 screw and, 2 nylock nuts, is used at least 18 times in the assembly. Each as it's own individual component, joined individually. That's at least 5 joints per small assembly.
However the bearing block is static meaning the design of it isn't going to change. The thickness of the vex profile is also not going to change, so the 5 components should get into a little assembly design without a timeline. There they can be moved into position using the move and align tool and then assembled with a rigid group. This will eliminate 18*4=72 joints from the timeline of the overall assembly.
There is a lot more potential for this in all of the designs.
Let me know whether or not this makes sense.
olks that work wityh Fuiosn 360 in profesional environments. @kentchad wrote:
I found a few YouTube videos on workflow - is there anywhere else you would suggest going to learn more about how to do this more efficiently?
No. A lot of my suggestions are based on the experience gained through analyzing hundreds of Fusion 360 designs posted by other users. Perhaps at some point in time I will create a tutorial, but a good comprehensive assembly tutorial is a monster project.
Also, I think form an engineering standpoint, there are a number of problems with the overall drivetrain design that need to be addressed sooner, rather than later;
- The wheels float horizontally on the shafts, so spacers of sorts are needed to prevent the wheels from rubbing on the rails.
- The large gear visibly interferes with the wheel. Is that intended?
- If not,
- another spacer is needed to maintain a constant distance between the large gear and the wheel.
- Then the problem arises that the overall spacing between the rails is too small. The hole spacing on the VAX rails is .5", so the rail spacing needs to increase from 2.5" to 3".
- Then the shafts are too short 😉
- Also, I question the selection of wheels, or to be more precise not all 4 wheels should be equipped with the little rollers. If all 4 wheels have the little rollers, the vehicle will roll down a slanted surface sideways without any means to stop it.
- If the interference was intended to stop the little rollers from moving, then that should only be the case for either the front, or the rear wheels but not both sets.
- If not,
The base frame with the wheels is the foundation of that design. My recommendation would be to address all the little problems, incl. how to actually affix the motors and encoders to the rails. Once that foundation is functional, then move forward.
@kentchad wrote:
That makes a lot of sense. He's still 12 so he has a LOT to learn.
That is one big, complex assembly for a 12 year old! My respect to the young man!
Don't hesitate to ask more questions.
Here's another suggestion and a rather important one. Consider patterns and symmetries and how you can use those to your advantage. For example the current wheel base frame (the two profiles with the wheels between them) is symmetric about the ZY plane.
Instead of separately building two independent assemblies, you could build just one and then mirror it. Unfortunately in Fusion 360 that would not be a good workflow. BUT ...
If you could design this in a way that is also symmetric about the ZX plane then instead of mirroring it, you can use a circular pattern. That, again would dramatically cut down on assembly work and the number of timeline operations.
Another thing I noticed that in the beginning of the project the rails were moved and then "locked" into place using the position capture feature.
In the attached file I first created a dimensioned sketch and then joined the rails to that sketch.
I also used a user parameter to define the WheelSpacing. This makes it very easy to change some basic dimensions without having to hunt down a particular joint in an assembly.
