I work on a machine that was designed under solidedge and we Let's move it to SolidWorks 2018. To avoid having to redraw everything in 3D, I thought about opening the solidedge part with solidworks (which gives an imported body) and then used the featureworks with feature recognition in order to reconstruct a tree. Then I open each sketch of the functions in order to fix the sketch (it's to go faster, instead of putting back relationships and dimensions)
Then I record my piece.
I think it's probably possible to create a macro that can process in bulk all the documents in this file.
The problem is that I'm what you could call a novice, a blue and even A noob in macro
Do any of you know if it's playable? And do you have any leads? I just tested the macro recording function, but I really don't know how to use it, I'm really in the fog despite this damn heat wave.
Already a very average function for a part, so to extract a multitude of parts in batch to make a machine, I think it's going to be hell. And even if by some miracle it worked and the structure of the part was more or less correct, imagine the day you have to modify a part of the machine...
For me the best solution is placed between your screen and your chair, and seems to me the only viable solution for production.
Not the expected solution, nor the fastest, but the only one that can give you a correct result. (according to the cartoonist!)
I quite agree with you on the principle. Except that this is an existing machine that has already been built 3 or 4 times. So I assume that he should not have only a few percent of the parts in modification during assembly and development. In addition, given the size of the machine, we are talking about more than 1000 hours of work. We even considered leaving the parts in imported body and only if necessary a modification rebuild it in 3D.
I did the same thing from Topsolid to Solidworks on 2-3 important machines as well (over 5-6 months), a few years ago and personally too much risk with the feature manager, I never managed to get a correct part (except extremely simple part, then 1000...) On the other hand, transforming the machine and superimposing what you redraw, yes it's even mandatory for me. And yes it's long (less than studying it by 0) but at least it's reliable. Anyway, the macro must be feasible but will you be faster to make this macro, possibly make it reliable, and then test it? (Be careful, quite complex macro, see even more, very good knowledge of vba or .net mandatory)
Thank you for this feedback, we already have a subcontractor rebuilding a subset with the featureworks function. I'm going to pay attention to this subset in order to check that everything overlaps correctly in order to detect any anomalies.
