topic Re: Assembly Circular Pattern Lock Rotation Bug (2021.1) in Inventor Forum

Assembly Circular Pattern Lock Rotation Bug (2021.1)

gcoombridge — Mon, 15 Feb 2021 06:21:13 GMT

Hi All,

Assembly circular patterns with locked rotation checked are not updating correctly in 2021.1 when related geometry is changed. Rebuild all does not jog it into action but if you double click the feature it will update. Circular patterns without locked rotation don't do this...

Screenshots and sample assembly below (using ilogic form to update the diameter):

Re: Assembly Circular Pattern Lock Rotation Bug (2021.1)

johnsonshiue — Tue, 16 Feb 2021 23:45:38 GMT

Hi! I believe this is a bug or a limitation. The issue here the Base Point (reference point) was not selected. If you select the center of the source disc, it will update accordingly.

Many thanks!

Re: Assembly Circular Pattern Lock Rotation Bug (2021.1)

gcoombridge — Wed, 17 Feb 2021 01:45:25 GMT

Thanks @johnsonshiue! I was too quick with my bug label.. I had this issue with an ilogic assembly when I was cycling whether the rotation was locked on or off based on a boolean parameter:

oPattern = oDef.OccurrencePatterns.Item("BasePlatePattern") oPattern.LockRotation = BasePlatePatternFixed

I guess I just need to define the base point in there too (looks like PatternRadiusPoint in the API help.

That being said it is an odd workflow because the pattern executes correctly without the base point - it just doesn't update... It would be clearer if the dialogue required a base point to be defined.

Re: Assembly Circular Pattern Lock Rotation Bug (2021.1)

gcoombridge — Wed, 17 Feb 2021 01:46:48 GMT

An improvement may be that the base point defaults to the origin point of the first pattern element if one is not defined to prevent the update issue.

Re: Assembly Circular Pattern Lock Rotation Bug (2021.1)

johnsonshiue — Wed, 17 Feb 2021 02:00:28 GMT

Hi! You did not do anything wrong here. To me, any unexpected, unreasonable behavior is a bug. A bug is a bug. No matter how we call it.

The thing is what we can and should do about it is a separate discussion.

We have every intention to fix any bug. However, it does come with cost. Sometimes a bug, even the most serious one can be trivial to fix. Quite often a bug may be a bit involved and there could be other reasons causing such behavior. Fixing it may introduce unknown consequence, prone to leading to unintended regressions in other areas.

If we can start from scratch for every bug fix, it would be ideal. It would be super easy. But, we have to deal with existing paradigm and tweak carefully. It is indeed very hard to make everybody happy.

Many thanks!