mirroed geometry - rel 2001

The first thing that will cause a reoder failure is dependencies to features

This has me confused, now. How did you do the mirror? Copy, translate, move, etc.? I don't know, Dave, it might have worked just as well to redefine the default datums and reverse the red/yellow direction of the mirror plane and the other reference plane. Well, I don't know, maybe that would have made your part "upside down".

Anyway, did you do the mirror as 'dependent' or independent. That might make a difference. If dependent, the original side is driving. If independent, you'd think you'd get dimension references on both sides. See if you can switch from dependent to independent and see if that makes any difference.

David Janes

It wasnt a Copy/Mirror command.

Think the features would have failed.

I used Feature/Mirror Geom (not Copy). I mirrored all the geometry existing at the time. I cannot redefine it. It shows up as a 'merge' feature in the model tree. There was no "dependent" option.

Seemed like a good idea at the time.

Dave

: : I used Feature/Mirror Geom (not Copy). I mirrored all the geometry existing : at the time. I cannot redefine it. It shows up as a 'merge' feature in the : model tree.

Oh, oops, I remember that one. Just a dumb copy of everything in the part to the other side of a datum. You can't even pick which features to mirror. Yeah, I thought they got rid of that revs ago. Now I even see it in Wildfire ~ 'Edit>Mirror'. They call it a merge feature but it's more like a single feature iges import. What would happen if you used insert mode before the mirror, Copy>Mirrored the geometry you need, deleted the Mirror Geom and rerouted the rest of the features that were dependent on the Mirror Geom. Wouldn't that give you more leeway in dimensioning, more shown dimensions on the other half?

Dave

I can't get that to work now. All the features fall over if I delete the original mirror merge, and they would have to be redefined. It's too complicated. I just wish the feature worked properly in the first place!

this may be too late to help, but here goes...

I agree that the mirror function does not work well in Pro-E. What does work as expected is a surface mirror.

Create a surface quilt of the geometry to be transformed by using seed and boundry or solid surfaces as references. Then use Feature|Create|Surface|Transform

To create a mirror of the geometry. Then create a solid from the quilt.

To avoid small holes, or slivers, first merge mirrored surface with the original surface.

BTW, if you roll back the model to a point before the surface creation, you can add features, and then redefine the surface feature to make sure the features are updated. Resume the mirrored feature, and this will update also.

Depend>

Chris,

I've had the most success with the methodology that you've described above. There's really not a perfect way to achieve what the original poster wants to do, but I feel that overall, your suggestion offers the user the most flexibility regarding downstream model modifications.

Pro/E has always kept mirror-copied features' section sketches coincident with the parent feature that they were mirrored from. So when users downstream attempt to perform a Feature-Redefine-Section-Sketch operation on a resulting mirror-copied feature, they immediately become confused about the section sketch's location and orientation. It would be nice if the application would mirror the section sketch whenever the user chooses 'Independent', and leave it in the original location when they choose 'Dependent' during the mirror-copy operation. Same as today for the latter. Also, if the user chooses 'Dependent', the system should issue an error message stating that the 'mirror-copied feature's section sketch cannot be redefined because it was mirror-copied using the 'Dependent' option. Please delete and re-create'. Or something similar to this.

S.T.

: >>: : >>: I used Feature/Mirror Geom (not Copy). I mirrored all the geometry : >

: > existing : >

: >>: at the time. I cannot redefine it. It shows up as a 'merge' feature in : >

: > the : >

: >>: model tree. : >>

: >>Oh, oops, I remember that one. Just a dumb copy of everything in the part : >

: > to the : >

: >>other side of a datum. You can't even pick which features to mirror. Yeah, : >

: > I : >

: >>thought they got rid of that revs ago. Now I even see it in Wildfire ~ : >>'Edit>Mirror'. They call it a merge feature but it's more like a single : >

: > feature : >

: >>iges import. What would happen if you used insert mode before the mirror, : >>Copy>Mirrored the geometry you need, deleted the Mirror Geom and rerouted : >

: > the rest : >

: >>of the features that were dependent on the Mirror Geom. Wouldn't that give : >

: > you : >

: >>more leeway in dimensioning, more shown dimensions on the other half? : >>

: >

: > I can't get that to work now. All the features fall over if I delete the : > original mirror merge, and they would have to be redefined. It's too : > complicated. I just wish the feature worked properly in the first place! : >

: >

Truth be told, 'Edit>Mirror' or previously, 'Feature>Mirror geom', was meant to be used as the last feature of a model, one built as half a model with bilateral symmetry accounted for in the design of the first half. Limitations should be taken into account before trying to use this feature: * As Dave pointed out in the first message, shown dimensions appear on only one half of the model ~ the modelled half, not the mirrored half. * It is also likely that curve dimensions will appear only as radiuses, not diameters. * G2 tangency between the halves is a crap shoot, even after taking such measures as using datum ribbons. * Finally, the Wildfire function works on surfaces but not, apparently, on solids.

David Janes

Thanks, but its a complex part and I don't want all the pre-mirror features to become a single feature. If I revisit this model again in the future I may recreate it from scratch, and never touch the feature/mirror command ever again! Don't know why PTC can't make the dimensions appear on the mirrored half though.