Hoping for a little perspective or pointers to documentation.
Recently chose Shark LT for design work on some folding benches and tables. I’m very comfortable with drafting and 3D visualization but until now have not needed the 3D CAD precision. Rotation, tapers, (and potential CNC or water jet production) pushed me over the edge.
Where I’m having difficulty is with associative history, in modifying the primary 2D geometry that drives an extruded (0.750”) 3D shape that is then “rotated”, “translated”, and “mirrored” into place within the primary object. With a simple 2D->3D part, it is a joy to use and a huge timesaver. However I’m having trouble grasping the boundaries or limits of the editing and adding primitive objects. Indeed using “undo” in this area is fairly unpredictable, occasionally generating multiple copies of objects(!) and 'Undo' in this context has generated the crashes I’ve had to date. V9.0.4, BUILD (1161) (Is there an user operation logging feature, so I can see what I’ve done?)
I’m trying to identify a collection of primitive objects and manipulations that are trustworthy on further manipulation as the model is extended and new requirements are discovered (e.g. slots, holes, edge manipulation) . Associative history (correct terminology?) is superb, but I’m finding the learning curve rough and having particular trouble finding trustworthy editing techniques.
For example:-- adding interior circles and bezier curves (cutouts) works, and edits well
-- Rectangular polygons do not work with “Profile Surface_nnnn > Add Curve”, however if they are inside the boundaries (edge chain?) prior to the extrusion, they work, and edit as expected. Huh?
-- Adding a closed bezier curve seems to work, but creating a sharp corner(s) is an art form not yet discovered by me. Thought I tried all modifier keys during creation. Dragging a “vertex” onto an “endpoint” lacks suitable snapping or precision. And then how do you extract the “vertex” if you later need it? I'm guessing one can with "Inspector>Data", but finding the "vertex" point in question is rather hard past trivial bezier curves.
-- adding a fillet and then converting to a bezier allows editing a corner. Finding a way to convert a fillet or bezier back to a sharp corner (many edits later) is not so obvious.
-- can’t find a way to split an outside boundary line into two lines and preserve the model. “Segment to Curve” removes dependencies. Changing a “Line” to “Control Point Spline” seems a bit overkill, and managing the Vertex to create sharp corners is a quite tricky and error prone.
-- occasionally the Part_nnnn turns into an “ACIS solid”, losing all associative history, but I’m not sure why. Probably removing a defining feature, but it happens rather silently, so I’m not sure what (and when) I’m doing wrong.
-- The scariest part of tripping into an “ACIS solid” is that “undo” often does not work, thus is not possible to get back to a defining model. Obviously saving often helps. Just seems rough compared to other parts of Shark.
Are my expectations too high?This seems fairly rough sledding compared to other aspects of Shark LT. I went with Shark LT over renting Cobalt on the monthly plan (monthly is very attractive when exploring something new with an unknown future) but product releases and features appear to dramatically slow, after Tim left. And I like supporting the innovator, but the associative history is really rough and hard to discover.
My workaround:I’m starting to adopt the concept of putting the defining 2D profile into it’s own hidden layer, then editing the 2D original, then re-doing the extrusion, rotation, translation, and mirroring (about 1 minute with practice).
As I’m new to Shark LT, am I missing something obvious? What should I be reading? Is this a well understood CAD strategy (I can read about somewhere)? Professionally I’m a software developer, so I’m used to developing a model and planning ahead. Both Tim and the software seem very well organized. The base "Profile Surface" editing techniques seem quite hard to discover.
I’d be happy to build a model and video to help others, but discovering the core ideas, and their limits, is much harder than imagined.
Cheers, - Phil
Edited by user Thursday, October 8, 2015 12:14:36 AM(UTC)
| Reason: Not specified