1. Home
  2. Solidworks CAD Design software
  3. Flexible subs

Flexible subs

SW2004, SP2.1

I just found a problem that I believe is related to SW2004, but I only have one service pack available right now to try. The short version is that with a flexible subassy that has patterned parts in it, the patterned parts would not move with their parent part when it was moved.

Here's the bigger picture. I have a lift that goes up in the air 30'. It has a tower that is bolted to the floor, and a carriage rides up & down on it. That carriage assy has arms that go in & out, so I had made the carriage assy flexible to be able to drag the arms while in the main assy. So, with 2003, I could drag the carriage up & down normally and also drag the arms in & out because of the carriage assy being flexible.

When I converted to 2004, I didn't immediately open the whole lift, but instead was working on drawings of subs. So, when I finally opened the whole lift, I found that most (not all) of the patterned parts in the carriage assy would not go up & down with the carriage as it was moved either manually or by postioning with mates. The strange part was that upon opening the carriage assy, the parts were out of place there also, which made no sense at all since none of the patterns were dependent on the upper assy.

I could make them move by opening (editing) each pattern and closing it, not even changing anything, but it wouldn't stick. I tried choosing another edge for the pattern, blowing away the pattern and recreating it, suppressing stuff, cycling through lightweight & back, large assy mode & back, all to no avail. At first I thought it was maybe only the ones that were patterns of patterns, but then disproved that.

So finally I found the lightweight switch, updated per the above proceedure, and now all is well. I had tried all of this on more than one machine, and had created a test assy, and all results were the same, so that ruled out my machine, and also my lift assy.

Take a look at flexible subs and see if they still work - let us know.

WT

Reply to
Wayne Tiffany
Loading thread data ...

I think this was an issue in SWX2001+ as well, but I don't remember if it behaved the same in 2003. (In other words, it may have been like this since the introduction of FSAs)

Have not worked with FSAs lately, but IIRC, a rebuild (maybe a cntrl-Q) would fix the issue.

Reply to
Arlin

Multiple CTRL Q's, rebuilds, etc. did nothing. It used to work just fine in

2003, SP5.1, and I just tried it in 2004, SP0. No workey there.

WT

Reply to
Wayne Tiffany

Wayne,

A few questions -

Are your patterns that won't move Derived or Local?

The components that are not moving are they in the top level assembly configuration or in a subasssembly configuration?

So who is the one that goes 30 feet in the air to see if this works?

Regards, Marie

Reply to
mplanchard

The patterns are manual linear patterns - none derived, and not all one direction or 2 directions. They are also all in the carriage assy, which is one level down in the lift assy. But what's interesting is that a few patterns are not misbehaving, and I couldn't tie it to patterns of patterns either. On the junk test assy I made up, I just put a sub (with a manual bolt pattern) into another assy and if the sub is defined as flexible, the pattern will not stay updated in the top level. I can send you my test if you wish.

The lift is the one I was working on that I showed you at SWW. Hopefully I'll make it down to Nissan to climb to the top. :-)

WT

Reply to
Wayne Tiffany

I had a similar problem with patterned features not flowing dynamically with the assembly but not flexible. I was animating an assembly and the component patterns were causing interference because they were not dynamically moving with the assembly, thus causing lockup and me pulling out my hair because I didn't at first know what to look for. VAR told me that it is working as designed I still think it should work as NEEDED. I ended up just mating each component individually because there were only a few. This could cause major problems for somebody who derived other mates off of the pattern though, especially if the patterns were not bolts. The thing that gets me is when I would drag in the assembly everything, as I recall, moved as expected.

Corey

Reply to
Corey Scheich

Saw this the other day on a tiping trailer assembly one of our users was working on. Problem was exactly as you describe.

We were able to get around it because the offending sub assembly diden't really need to be lightweight.

He was still on SW2004 SP1 we also tired on another machine with SP2.1 and got the same result. Al

To reply via email remove SPOIL from address

Reply to
Al C

Ok - I tried this out at spin 0 - no need to send me the file.

In my opinion at the top level - things work fine - but go one flexible level down and the bolts stay at the bottom location and every thing else moves to the top shelf.

I do agree that this happened a few releases back.

Will have to wait and see if it was fixed at a later spin

Are you calling this one in?

Regards, Marie

Reply to
mplanchard

I say this behavior several weeks ago in a very simple assembly. FSA have a lot of problems. I'm convinced that they produce application instability as well. When combined with limit mates, things get ugly.

Jim S.

Reply to
Jim Sculley

I sent it to my VAR, but I don't think he has tried it because of teaching classes. I'll push him again. Thanks for trying it.

WT

Reply to
Wayne Tiffany

Component patterns and mates again. I think I've seen this in a previous release as well. I'm kind of superstitious about a few things, and flexible subassemblies is definitely one of them. I think in the end, this is the same question as your other component pattern issue, but I think you're right that this one is a bug, where SW got caught playing that fast and loose game with the history dependent assembly items.

matt

"Wayne Tiffany" wrote in news:c1ih20$1iuke0$ snipped-for-privacy@ID-201804.news.uni-berlin.de:

Reply to
matt

We have seen this in 2003 also. The only way we could correct it is opening up the sub at the same time and doing a ctrl-q. It then updated the sub and the main assembly. This among many other reasons is why I discourage our users from using flex assemblies. The biggest reason is because of the huge performance hits that we see using them.

SolidWorks needs to save this information where it pertains and not back tho the sub. I have had many discussions with our VAR and SolidWorks themselves on this. Hopefully they will change how all of this is handled in the future.

Reply to
SWuser

PolyTech Forum website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.