Why this SW2004 error?

Hello.

Can anybody explain why the following happens in SolidWorks 2004 SP1?

I have a semi-complex part with a Variable Fillet applied on one edge.
No errors, no nothing. If I save the part, open it again and only EDIT
and after that CANCEL the variable radius feature, the feature gets
red. The error is "... radius of the fillet is too large to fit the
surrounding geometry...".
Can anybody please help me in this regard and explain the reason for
this error or, most probably, what am I doing wrong?

Any help will be greatly appreciated,
dUMZ

 

Kind of hard to tell without seeing the part, but you might try opening up
the part and running the part check on it. If it says it is OK, try ctrl-q
(which forces a rebuild) and run the part check again.

Under Tools/Options/System Options/Performance is "Verification on rebuild"
checked? I think it defaults to off, so your performance is faster, but it
will catch errors that otherwise might be missed until later.

Jerry Steiger
Tripod Data Systems

 

Like Jerry says.

The only things I would add to that would be to run Tools, Check. This may
point out a bad spot in the geometry. In fact, you might want to roll back
before the fillet and run the check. Is the part imported or all native?

Other things you might try is to cut away the part bit by bit and see where
the fillet fails.

There could be other things wrong too, like funny things can happen
sometimes if you use a "0" value in the middle of the VR fillet, or if the
transition between values is too abrupt.

Also, and now that I think of it, this sounds more likely, it is possible
that the fillet was at some point dependent on the geometry of a feature
that is no longer there or that the edges the fillet is dependent on have
changed because of a split face, edited sketch or some other change? There
could be something going on like a "dangling" condition with one of the
control points. Are all of the control points specified? What happens if
you delete and recreate the feature? Do you have dimensions controlled by
equations or link values?

There are a ton of possibilities, including something wrong with the
software. I couldn't put my finger on it without seeing the model.

good luck

matt


(dUMZ) wrote in @posting.google.com: