Assembly Rebuild Times 4

[Eltron]

Hello again, Boys and Girls. I was just wondering if there is a way to check the rebuild times on an assembly. In a part you can check the Feature Statistics, but there isn't an option for assemblies. AssemblyExpert just tells you the number and type of parts. Maybe someone has a macro?

Dan

http://www.eltronresearch.com

Dan's Blog

 

[JMirisola]

Don't you have a stopwatch?

Jeff Mirisola, CSWP, Certified DriveWorks AE

http://designsmarter.typepad.com/jeffs_blog

 

[handleman]

My guess is that rebuild times for an assembly would be sort of difficult to calculate. Solving mates, with their potential for circular tanglements, probably means that the rebuild time depends on what you changed. Also, it depends on how many components/subassemblies have to be rebuilt for the assembly to rebuild.

Just shooting from the hip, though.

-handleman, CSWP (The new, easy test)

 

[Eltron]

Right, but the same can be said for parts. I mean, if you hit the "Recalculate" button on Feature Statistics you'll get a different rebuild time even if you don't change anything. Just looking for something similar in assys.

Dan

http://www.eltronresearch.com

Dan's Blog

 

[Eltron]

I got pointed to a macro on fcsuper's site that will work.

Find it here:

http://sw.fcsuper.com/index.php?name=UpDownload&req=viewdownloaddetails&lid=72

Dan

http://www.eltronresearch.com

Dan's Blog

 

[CorBlimeyLimey]

A macro listing the individual rebuild times of components in an assy has been requested several times before. It would be a very useful tool when trying to discover which component in an assy is causing a long save-time.

 

[handleman]

So... which do you want, individual component save times or rebuild times? When you have an assembly open, SW keeps track of which components of the assembly have been modified. If a component doesn't get modified, it doesn't get re-saved when you save the assembly. The same logic is applied for rebuilds - When you rebuild the top level assembly (even Ctrl-Q) it only rebuilds components that SW determines require rebuilding. There is a sort of "super Ctrl-Q" available in the API that force-rebuilds all the referenced documents as well.

-handleman, CSWP (The new, easy test)

 

[handleman]

This macro will force-rebuild each referenced configuration of each part file in the active assembly. Suppressed and lightweight components are ignored. It starts at the deepest level of the assembly and force-rebuilds all referenced configs of all documents at that level. It then goes up one level at a time and force rebuilds again. Each subassembly at this level should contain fully rebuilt components, so only the current level is force-rebuilt. There is an option to force-rebuild twice in a row if desired.

I never use the "in use" configuration setting for any assembly components, but if you do I think this macro will likely change that.

The macro keeps track of which configs of which parts have already been rebuilt, so no config gets rebuilt twice, even if it's on a different level.

There is also an option to save the components (after rebuild) and report the save time.

The results of the rebuild/save are presented in Excel format.

-handleman, CSWP (The new, easy test)

 

[handleman]

Not sure if anyone's tried this one yet, but I thought it might be handy to go ahead and report the file size of each component as well.

I also turned screen updating back on in Excel. Turning it off doesn't make it run a whole lot faster, and with it off it sort of looks like it's not doing anything.



-handleman, CSWP (The new, easy test)

 

[CorBlimeyLimey]

handleman,

I hadn't got around to testing the first one, but that last one is excellent. It will really help (and already has) when trying to find out why an assy is taking so long to save.

Thank you.

 

[Tobin1]

handleman

Very nice - as usual .

Thanks

Tobin Sparks

http://www.nov.com

 

[Russell67]

Wow, this is nice. How can this be modified so i can specify the depth to traverse. Thanks

 

[handleman]

Probably something like this:

-handleman, CSWP (The new, easy test)

 

[Russell67]

This is outstanding. I ran this yesterday and was quickly able to identify a few parts that had a very high rebuild time. I was able to open those parts and looking at the Feature Statistics see what the problem was and by tweaking the design, either with an equation or deleting useless features i was able to shave off 4 seconds of rebuild time in my assembly(per part reduction of 1/3 rebuild time). Does not see like a long time but it also helped to reduce the file size and some of these parts can be used several times in an assembly.

Now with the ability to set the depth i can definitly see this as a pre-plm insertion process.

Awesome tool. Thanks

 

[handleman]

Glad you like it, Russell! One fairly small caveat about specifying the depth, though. If you specify a depth, but some components that are deeper need a rebuild, I think the subassembly at specified depth will report a longer rebuild time because the components will have to rebuild before the assembly can rebuild. That's the reason that the macro starts at the deepest level by default - by the time you reach any assembly each component of that assembly will have been already fully rebuilt.

-handleman, CSWP (The new, easy test)

 

[Russell67]

With the new file i have noticed that. However i have seen that the correlation between the rebuild times reported and the times posted by the Feature Stats is close enough.

When i run this on a large assembly i look for the high rebuild times then open and evaluate the parts to see how minor tweaks can be done to speed up the rebuilds.

With this info i can go back to the designeer and give him some pointers for things to avoid in the future. Also me as well to just overall better modeling practices.

 

[fcsuper]

handleman,

I'd like to do a small article about this macro. I think this is the kind of thing a lot of people have been wanting for a long time. please contact me. Thanks!

Matt Lorono
CAD Engineer/ECN Analyst
Silicon Valley, CA
Lorono's SolidWorks Resources
Co-moderator of Solidworks Yahoo! Group
and Mechnical.Engineering Yahoo! Group

 

[handleman]

Done. That is, as long as you are still a buzy person.

-handleman, CSWP (The new, easy test)

 

[fcsuper]

I am. and actually have been too. talk soon.

Matt Lorono
CAD Engineer/ECN Analyst
Silicon Valley, CA
Lorono's SolidWorks Resources
Co-moderator of Solidworks Yahoo! Group
and Mechnical.Engineering Yahoo! Group