Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

  • Congratulations waross on being selected by the Tek-Tips community for having the most helpful posts in the forums last week. Way to Go!

Mold Wizard Performance Issues

Status
Not open for further replies.

PheasantPlucker

Mechanical
Jan 23, 2003
38
0
0
US
NX 4.02.2
Dell Precision 370
2 MB Ram
Win Xp
Does anyone else in the group use Mold Wizard,and have they had performance issues? I have trouble adding components, where is seems to cycle thru and update every component in the the assembly file. Sometimes it takes 10 minutes to add a standard part.It also gets worse as you add more parts and it is also worse in some directories than others.
Any help on this would be greatly appreciated.
 
Replies continue below

Recommended for you

I'm quite new to NX and moldwizard myself, but have a few designs under my belt now. In one job I created a really ugly assembly, I had a lot of bodies wave linked in a not so clear structure. Minor changes would cause a long cascading update throughout the model that lasted a couple minuets (4x3Ghz, 8gb, new high end system). In more recent designs I have been careful with the assembly and linking structure, and seem to be avoiding this problem.
 
Pheasant,

Don't know much about mold wizard, but perhaps taking out any other variables might help. Try copying the whole thing to a local drive and setting load options to partial loading, from folder with interpart data turned off, and faceted reference sets preferred over bodies. It may not be very practical but that would be your benchmark for the fastest case. Turn back on the options you would normally use one at a time to see if that reveals the cause.

If the directories that are worse are on networked drives then the answer is probably line speed. Also on native systems deep directory structures when searched with the load options set to "Search Directories" will be much slower.

Best Regards

Hudson
 
Hi,

Try to set costumer default (default spreedsheet aplication)to xess.

It probably run better.

YOu should try it anyway

 
Interesting suggestion.

I may be wrong; but for Windows based machines wouldn't changing that setting may be equivalent to turning off all spreadsheet interaction altogether?

If mold wizard needs to use spreadsheets and or references excel in the background then I wonder if is isn't just your excel installation affecting performance. I don't know that it does, but I found recently that the anti-virus software and installing a HP wireless printer affected Excel and the Desktop Explorer performance dramatically on the home computer. If Excel is in the mix and always starts too slowly then you may want to look outside NX for your solution.

Now if this is plausible then it is plausible that changing that setting to xess would also do the trick, but perhaps at some cost.

Best Regards

Hudson
 
Thanks for the replies. I tried switching the spreadsheet to xcess, but didn't see any improvement.
All the directories are on a network,so I tried the local drive with no better results. The Top level directory in Mold Wiz has better speed than if I add a part to the misc subassembly or the prod sub assembly.
I've also reinstalled NX4, but it doesn't seem to have helped either. Every time I do something in MOldwiz, I can see on the status bar all the things updating. It seems every link of every part needs to update before I can continue. Very frustrating...
 
I wonder if you might have a flaw in the way you structured your components and linked bodies. In the model I had trouble with, almost all components in the assembly would update if I changed something simple like the length of a screw (similar to what you describe).

When I looked into it I realized that when I made the mirrored cavity I threw a big pile of components (including thier false bodies) into one mirror part. All those mirrored bodies in one part meant that if anything affecting any one of them changed, that whole catchall component would be flagged as out of date, then updating every downstream feature, pocket, ko pin subtraction, screw subtraction, etc (nearly the whole mold).

By splitting it up and making a mirror part for ko pins, a mirror part for screws, a mirror part for slides... only the actual changed components (any maybe a couple closely related components) would update. If you change a ko pin diameter, the screws don't update.

I got caught out on this one because of the huge difference in regeneration order and method from ProE, my old system.
 
We have 4/5 guys that use it for plastics and castings. They also have found some performance issues, especially in regard to the "updtes". We are also trying to use it over in TC, that brings a whole other set of issues. Is anyone using it in TC? What I find most disappointing is the structure of the std parts library. Some components use different attributes names and some are not filled out. It lacks consistency. I saw the same problems with that way back in 1998. It seems that they just keep tacking things on and never addressing the product as a whole.
 
Status
Not open for further replies.
Back
Top