Mates breaking?

Just opinion:

1. If I was evaluating moving from one program to another, I would pick the seasoned production version, not a near Beta like SWks 2006 where one can expect glitches for another few SP releases at least. (I use SWks 2005 SP4.1).

1. Some subtleties in mates can cause problems and Matt Lombard's web site has notes on these things that go wrong, and this news group has gone over some of the ways in the past.

Matt's site has links to lots of other SWks user's web sites, too.

Thanks for the link Bo, I will check it out.

With regards to using SW2005 for production use, I certainly agree. I'll get a hold of the VAR and see if I can evaluate the previous release. In all honesty, I wasn't aware how fresh SW2006 was until receiving it in the mail and reading up on some of these forums.

Thanks again for the advice.

-Gareth Conner

Gareth

Try hitting 'Ctrl + Q' This will force a rebuild of everything whether it needs it or not.

Rich

If a mate shows up in Red there is a mate diagnostic tool that will help you find out why. In SW you have to be very careful not to overconstrain an assembly.

I'm not clear on what you mean by mates in a top level assembly ripping apart a lower level assembly.

The rigid/flexible switch has to do with subassemblies that are underconstrained so that they can move like a mechanism. In the past SW required a subassembly to have a separate configuration for each instance of flexible in the top level assembly. In 2005 that requirement went away.

CTRL-Q does a forced rebuild. It regenerates the feature tree from top to bottom.

Someone suggested trying an earlier version of SW. That will not likely happen since once SW releases a new version the old one becomes unavailable. Even if SW2005 SP3 were more stable they wouldn't sell it to you.

Thanks for the reply,

I did find the mate diagnostic tool, but was unable to glean anything helpful.

I mentioned the rigid/flexible switch just because I am using some linkages in the assembly that require the sub's to be sovled as flexible (to similate the mechanism motion). I figured that if I mistakenly switch one of the sub's to flexible, and the sub had been underconstrained, that the mates in the assembly could move parts in the sub. However, I did confirm that the offending parts that moved were located in a rigid assembly.

I'm sure this is a case of operator error, as I've only been getting my hands dirty for a couple of days. The bit that threw me was that just closing and re-opening the file made the issues disappear. Initially I had been quite cavalier about declining the VAR trainging sessions (7 days @ $1800), but perhaps that was foolish. Without a manual, and only outdated 3rd party books, I sense that I'm fumbling around more than I should.

I've discovered you're correct about obtaining older SW versions. This is perhaps a shame, yet I'm becoming more convinced that the issues I'm having are of my own doing :) I have found many good things in SW, but there is a lot left to uncover!

Thanks again for taking the time to reply.

Best regards, Gareth Conner

You might also look into your local technical schools. Although you would be working with SW at least 6 months behind current release, the pace would allow you to absorb much more than VAR training. I don't know where you are located, but consider a large user group meeting or SW World for further training.

This is sometimes limited in assemblies. In particular, subassemblies frequently aren't rebuilt (even with ctrl-Q) until the next time the parent assembly is loaded. This is why some mate errors don't crop up until a reload, or are resolved when the assembly is next loaded. However, it can be done via a macro. I stumbled onto this a while ago. Sometimes it bails me out.

'forces rebuild of top assembly and all subassemblies

Dim swApp As SldWorks.SldWorks Dim Model As SldWorks.ModelDoc2 Sub main() Set swApp = Application.SldWorks Set Model = swApp.ActiveDoc Model.ForceRebuild3 (False) End Sub

This looks like what CTRL-Q is supposed to do. Are you saying that sometimes it will work when CTRL-Q will not?

WT

I'm not sure what the difference is, but ForceRebuild3(False) is distinctly different from Ctrl-Q. ForceRebuild3 updates incorrectly parts with equations in Linear Patterns. I notice this most often in bolt patterns, where the linear pattern has an equation for the number of instances. ForceRebuild apparently evaluates the equations with the default configuration values. The only recourse is to open the part, rebuild the part with Ctrl-B, and then return to the assembly. This (thankfully) fixes all configurations, but the part won't update correctly in place after ForceRebuild. Ctrl-Q apparently leaves the bolt pattern unmolested, or it evaluates the equation correctly.

I've not noticed the interference with equations. At least, not in conjunction with ForceRebuild3(false). I get problems similar to what you describe without it. Equations are a bit odd. They don't seem to take part in the history of the feature tree.

To answer Wayne's question, yes. This goes farther than ctrl-Q in assemblies, causing the subassemblies to rebuild.

Dale, Thanks a bunch! I'll give this a try next time the problem occurs. Ctrl-Q does not seem to go as deep as closing and opening a file, this macro should save some time!

Thanks again, Gareth

Dale Dunn wrote:

That makes it sound like it's not a case of operator error. Mates are not one of the more robust areas in SW.

It's a little late to be giving you this advice, but I am very fond of using the training as an evaluation tool. SW is pretty easy to pick up, at least as far as the basics go, but there are bound to be areas you miss in your self-training.

A good VAR will set you up with earlier versions if you want them. Ours did.

Jerry Steiger Tripod Data Systems "take the garbage out, dear"