Top down methodologies 1

[daveykbelgium]

I have a lot of experience setting up large projects with top-down methodologies. I now want to use mechanisms too.

Who can I speak to about top-down methodologies with mechanisms?

What about a question like - how easy is it to set up mechanism when moving parts, dependent on one another, are in two separate assemblies? I have seen no examples like this yet.

Does anybody know how to set up two side plates, say 4 free rollers, wrap a belt around the rollers, and drag the belt, moving the rollers?

Another nasty in static top-down assemblies - if the top-level assembly of a product is so big you can't retrieve the master rep (I have done two projects this big), how can you re-order, restructure and use group functionality?

I'm about to go over to Wildfire next month, so all my experience with top-down methodologies is with software up to 2001.

Looking forward to a few replies.

Happy Easter

Dave

 

[mloew]

Dave,

I do TDD with mechanisms all the time. The typical approach I take is to create a static representation of the mechanism as a single part skeleton (only using the minimum datum geometry necessary to detail the components for the mechanism). For example, for a planar four-bar type mechanism I would sketch the mechanism (quadrilateral) and put points on the vertices and axis through those points and normal to the plane of the mechanism. I would then create published geometries of the geometry required for each link (one of the curves and the points and axes associated with the curve). I do this for each of the 3 links that moves (i.e. not the ground link). Each of the three moving links then would need an individual part that only consists of a copy geometry feature of each of the pub geometries. In assembly mode, I assemble the original static part skeleton with a CS0 to ACS0, and then each of the three linkage points with pin joints (mechanism constraints). This same technique works for far more complex mechanisms as well, it is just illustrative with this example.

The assembly becomes a mechanism skeleton and preserves all of the mechanism functionality along with the parent/child relationships needed for TDD. You can instrument the design to introduce BMX functionalities to optimize the function of the mechanism as well. I do all my engineering work this way. Eventual solid parts that get developed “hang” on the mechanism components. It all works very well.

The only real difference with Wildfire is the ability to have access to the drivers (servo motors) of sub-mechanisms thus eliminating some limitation of having to include all of the eventual mechanism in the same assembly.

I hope this helps. Please feel free to follow-up with additional questions and comments.

Best regards,

Matthew Ian Loew


Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

 

[mloew]

Dave,

Was any of this helpful? I am interested in your feedback.

Best regards,

Matthew Ian Loew


Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

 

[Guest0527211403]

Matthew

Kudos on that, I couldn't have described it better myself.

One thing to add:

I try to automate the drawing process as much as possible using drawing templates, so, while creating published geometries, I create coordinate systems in the skeleton model which set the x, y and z directions for that part. I then use this new coordinate system reference while importing geometry into each of the linkage components.

This ensures that parts are not hanging out in space when you open them up individually, and that you can use standard views on drawing templates.

It's a bit of extra work in the beginning, but there is a definite payoff in the long-run. The trickiest part in the beginning was learning the different ways to make coordinate systems, and which way the x- y- and z- axes should point. I have to orient my model and confirm it with a 'right hand rule' check (you'll see me with my fist on the screen with my thumb towards me).

You may sometimes have to build some datum geometry around the sketches Matthew spoke of previously in order to have the correct references for a coordinate system. I recommend resizing the datum planes and axes so that they do not clutter up your screen too much (in WF2, this is in the creation dialog box. I cant remember how to do it in 2001). Assigning similar names to the datum features will also help you create rule based layers to ease up the display as well. In a very large assembly skeleton, it would be a nightmare to work on if you did not do this.

Hope this can help

Best of luck
Mark

 

[mloew]

Mark,

Thanks for the comments. I too use a local coordinate system for placing the copy geometry feature. The target part then indeed contains a nice reference coordinate system to enable further operations downstream.

One thing I have stumbled with is in how to create prints with geometry based on pub geoms/copy geoms as the do not contain any dimensions. I am forced to use created dimension in the prints (yuck) or use annotation features in the parts/assemblies. Any better ideas?


Best regards,

Matthew Ian Loew


Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

 

[mloew]

There is a similar thread on mechanism skeletons at MCADCentral that has been revived.

Best regards,

Matthew Ian Loew


Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

 

[daveykbelgium]

Hello Matthew and MArk

I've just got back from holiday so sorry for not replying any sooner.

I will digest what you've written and get back to you in a few days.

Many thanks for your replies

Dave