Hi @TimF Thank you! Sorry for the amibiguous description.
What I showed is only one half of my model. If the other half is also shown, hope you can see what I'm trying to do - removing the gap at the center line. It seems that the AFB need to be created across the center plane after morroring?
I hope that AFB can be created before mirroring, because there are still a lot of things to add before mirroring.
What difference does it make whether the blend is put on before or after the mirror?
You still should model PAST the mirror plane then trim everything to the mirror plane right before you mirror. You DO NOT want to model UP TO the mirror plane.
You have an issue with the half that contains 2 surfaces. The one on which you did a Match Edge is NOT TANGENT up to the mirror plane. Fix that then see if you can do what you want with the AFB prior to mirroring.
What difference does it make whether the blend is put on before or after the mirror????
If the blend is created across the center line after the mirror, there would be no gap in the blend at the center position. Is this correct?
Take a look at what I did here. This is a quick fix - meaning it might be better to fix the symmetry issues with the surfaces that are closest to the plane of symmetry rather than using the Edge Symmetry command - I haven't used Edge Symmetry enough to know for sure whether it's a reliable tool in every situation. It may work here because you're using 0.5 for your G2 tolerance - if I recall correctly, the G2 was coming out around .27 or so at the highest.
Notice how I extended the surfaces before applying the AFB. I do this because I've been burned too many times having trimmed the blend's adjacent/tangent surfaces to the plane of symmetry then applying the blend only to discover that the blend drifts across the plane and doesn't come right up to it and causes either a deviation in continuity or it just won't Sew without a small gap. The deviation in the surface can still happen using this technique but I believe the larger surfaces were clean enough that the blend didn't get skewed/wonky when trimmed to the plane of symmetry.
I had to really go overboard with the Face Analysis display resolution - all the way to the top to get clean looking reflection and radius (Gaussian) results. Go ahead and check all edge continuities - I just checked across the plane of symmetry.
Hi @TimF Thank you so much for teaching the technique of extending and trimming to achieve continuity across the symmetry plane!
The relavtively high G2 error doesn't matter, because G1 at the symmetry plane ssems to be enough.
In the past I prefered to use Englarge to extend surfaces. Here I realize that your adoption of Extend Sheet should be the better choice for this case.
As @Maurizio suggested, I need to decrease the tolerance for my whole model. Recently with many troubles caused by tolerance, I began to pay attenion to the tolerance issue.