As you know split lines can be very difficult to remove from a part if a design changes means they are no longer necessary. This happens to me with plastic part design occasionally when a split line meant for draft reversal or stepped partling lines etc changes.
A method I use to remove the split line feature with lot's of children (and if there are simpler ways please let me know!) is to rollback before the split, offset a surface from the face the split line originally split, then roll forward, edit split, then replace solid face(s) with offset surface(s). This after this change the split feature will have no children and can be deleted.
There are probably lot's of scenarios where this wont work, but for me it's helped a few times.
Thanks for the tip. I can see it being especially useful for intersection and silhouette split line features
When using your technique, do features referencing the split faces need to be redefined to get rid of dangling errors? Sketches on planar faces, edge fillets, sketches that reference edges and verteces from the split faces, etc?
For years I have used the following technique for sketch based split lines - edit the sketch used for the split line, and redefine all the lines to 'construction'. The split feature will fail, but that's good (I call it sabotaging the feature, a trick that has a lot of applications outside of split line). I can then redefine any children then delete the split feature. However, if your technique doesn't require any rework/repair of the kids I'll have to give it a spin.
Good tip Zander. There is another way also without having to rollback. At the point you want to get rid of a split line, and if it's a solid body, find a way to turn it into a surface body by deleting a face somewhere using the delete face option. Now make a new face (or copy the face before you deleted it and use it) and then knit it back to solid. Knit automatically removes splits in most cases.
I object to this behavior in knit and I'm working to make it a checkmark option in the knit command, but in this odd workaround it works for you.
Related to the split line topic - I've been experiencing trouble with doing reverse drafts on portions of faces that have been split by the split command. ie. I'm relieving an undercut - creating a seal-off so a split line is created on the face. I want +1=B0 draft on one portion of the face and -1=B0 draft on the other portion. Often times the draft command is unpredictable in this situation and will usually consume the split line hence drafting the entire face one way or the other. Draft seems very sensitive ie. I needed a parting line draft created in an area that needed to be duplicated on other side of the symetrical part. One side would draft beside the split line, but the other side would consume the split line as described above no matter what I tried.
1) with Mark's suggestion you will be leaving the split feature in history - this approach just gets rid of the lines. A decision to use it depends on style. I spend most of my working day in rollback so features are where they are 'supposed' to be, not just adding stuff to the bottom of the tree. While rolled back I also remove irrelevent features, instead of just cutting them away ( or otherwise working around what they left me). However, that is just style, and I cannot be so arrogant to say that other styles are not valid (OK, I can... but I'll lay off the 'best practice' opinion). If you are most comfortable working at the bottom of the tree, Mark's suggestion will be a life saver.
2) Add on tip: REMEMBER to hit 'try to form solid' in the surface knit feature when you are doing this. If you first knit, then thicken in a second feature by 'creating solid from enclosed volume' you lose a lot of really good functionality. I wouldn't mention it if I haven't seen it a bunch of times. Here's the deal - if you double click a solid face that was made from a knit-to-solid, you still can access the dims for the feature (critical if you want to do equations or undo experimental changes to sketch dims
- if you edit the sketch, you lose undo, but if you double click the feature and make the change, you can undo) If you double click a face made from thicken-boss-from-knit-feature, you no longer can access those dims (at least through 2006); that thicken-boss becomes a hard parent for everything that follows, and you lose a lot of editing flexiblity (and i think drawing flexiblilty with inserting dims from faces). It shouldn't work this way IMO - a face should always remember what feature was used to make it in the first place, but we play the cards we are dealt. For folks following mark's tip, you'll be happier if you make the solid out of the knit feature.
Hear!Hear! and thanks, Mark! How many times do I use split to control loft start tangency (within the 60 deg limit, which I would also like to have as a checkmark) , only to lose that prep-work when I knit? Thanks yet again for working for us, man. Ed
Good point to bring up that undo is now operable on sketch changes in
I don't use 2007 yet on real work yet. Please note that's why I made a point to mention in my post that my experience was 'at least through 2006'. However, I would challenge your notion about 'impressionable minds' here. I think there are just folks (perhaps more careful readers?) looking for ideas to try out.
Personal to Daisy - Dude, you have to pick a beef - must I put 'through
2006' after every item I write, or are you going to bitch about my posts being too long because I have to layer on exceptions after exceptions because of a guy like you? Be a MAN - pick one and go with it. I look forward to your decision.
Ed PLEASE don't shorten ANY of your posts. They are the most thought out (and thought provoking) of the group.
I need to say that I hope that there are some impressionable minds here. I hope we all have the ability to still be "impressed". I think what Ed (and Matt, Wayne, Paul, TOP, Zander, Dale, Mark, have I missed anybody) is (are) trying to do is to get us to THINK for ourselves. If I see something here and say to myself "I think I'll try that", That's a good thing. Now sometimes I try it and it doesn't work for me but I have, at the very least, gained some knowledge. And sometimes, I'll admit, I can't get my little pea brain around is being talked about, but I try to store away for later when I say "Hey there was something on the NG about this a while back".
All this to say to Daisy (sorry I won't get your reply) is the LAST person on the planet you should be attacking is Ed. He's has (arguably) given more to this NG (read ME) than anyone else here!
Hi Muggs, I assume your speaking of flowerpot? I put him in my killfile as soon as I came across him. I don't know about everyone else of course, but my life is way to short to waste reading that kind of drivel!
There is a function that will geometrically remove the split in a simpler way, though. If you do a Delete Face using the Delete and Patch option, selecting a face to one side of the split, it also gets rid of the split line.
Of course you have to weigh if having all of that feature history lingering is really what you want. There is sometimes a point where expediency outweighs style, given that the results are the same.
Another good suggestion. The thing that gets me about delete and patch is sometimes it is really fast, and sometimes it takes forever (OK, just feels like forever ). I've never been able to see much rhyme or reason to it. Any ideas/insights/patterns to delete+patch times that you have gleaned from your experience? When removing a split line I would figure it would always be instantaneous. Ed
Good point Matt, I'll make sure to get that into the spec to get this option in knit and perhaps off by default in Delete face. As you probably know, the delete-patch command is just a macro to the sub-routines of face-untrim and knit solid.
BTW guys, if you haven't make sure that you're getting OER (online enhancement requests) on things like this - nothing impresses development more than mulitple customer hits on issues.
why bother? the idea is there why keep repeating it? same with bug reporting... surely the peolpe at SW have some idea of the worth of these things themselves. if not put up a wish list and have users vote on it
and despite all the requests and suggestions over 'years' we are still waiting for the new forum and FAQ etc etc...does anything really actually happen that customers ask for or is it all a fop?
and...why can't we have our curvature comb outline back? FFS Mark you do ID stuff...
it doesn't actually make any difference what we want we get what marketing tell us we want. while you might be doing your best it is obvious some your fellows have a piss poor attitude. my few cents
Neil, the enhancement requests are important because developers are not exposed to the design enviroment at all. They are busy working on the software not designing parts etc.
Therefore they cannot understand how important some seemingly small bug or feature is to people who need to perform that function 150 times a day. eg. Arrowhead on notes and ballons not 'sticking' to edges in
The only piss poor attitude here is yours, I can only assume you don't carry on like that with people in your real life, so why be that way here. Posts like that do absolutely nothing to solve problems or enhance the software. Making it personel is actually not contructive and ditto on getting emotional.
So you believe that SolidWorks is doing everything possible to improve the quality of the product. There is no doubt you are respected amongst your peers. So why kiss up to SW in defense of their obvious short comings?
Does not the Beta program discover and report a majority of these problems prior to release. Why can't the developers institute all of the solutions before release and actually work like intended? From my reference, SW has not made any significant progress towards improving the release quality of the product. There are certainly plenty of marketing types that would like to convince us otherwise.
We all get emotional or we wouldn't be human. Suppose that is also why some will defend the company regardless. How many times has SW said that quality is a high priority and the users will continue to see improvements with each new release. Three cheers for SolidWorks!
Users pay for software and service. We should expect a product that meets and exceeds our expectations. Or is that just another marketing bullshit line.
Don't guess what I believe because you'll probably be wrong! Seriously though, I'm the one who started a thread recently proposing a halt to all new feature developement until long standing bugs are fixed. I don't believe sw is doing everything possible - that wouldn't be logical... I do believe that you shouldn't talk to people on usenet any differently than you would if they were standing in front of you. I don't think writing endless streams of edgy nasty posts in every thread saying 'where the hell is my f'ning curvature comb border' will achieve anything.
I don't think I've defended anyone or any company but I've met many of the actual people who work at solidworks and know them to be thoughtful and commited to the software. At the same time, I get frustrated by long standing bugs that never get fixed (arc length dimension with dual units). I also recognize that a software developer who isn't under a deadline to send out 60 pages of drawings will not understand how much time is wasted by some minor bug, so I do let them know, I call my var, I send emails and talk to people at conventions - but I don't shout or act rude because I try to treat people the way I like to be treated. If you think you can justify agressive nasty behaviour when it comes to software I think you don't have any 'real' problems at all
- like starving or dying for example. This isn't the end of the world, it's a cad program, it has bugs but it's also great and fun to use and has greatly improved my productivity and work enjoyment over the past 8 years or so.
It's a big complicated program and codebase - it will likely never be bug free and never work exactly the way the you want it to. So take the good with bad. If there is more bad than good for you try some other software, there are loads of them out there.
Also, try to figure out what positive effect yelling at someone has on solving a problem. Have you ever had success coming up with solutions by yelling?
"Do I contradict myself? Very well, then I contradict myself, I am large, I contain multitudes. "