Trying to create a face fillet in a surface model - keep getting this message. No idea how to trouble-shoot to avoid/correct the condition. Help is no help. Thanks.
Matt Graves - Somewhere in the part face surfaces there is either a lip or a gap, so basically there are more then two connecting faces. One thing that will get you good answers is to either upload the part or add pictures to your thread..
Right now I would be happy with some pictures of surfaces that aren't smoothly connected - I can't upload the part, and I'm reluctant to show pictures. Thanks.
Can you turn the Surface into a Solid???
Understand that we can't see, we can't really help, could you just do a sample??
This is the best I can do. Trying to fillet edge between blue faces and the faces above it. I can make this in a couple other ways but what I really want is a constant width fillet, but all face fillets are off the table because it thinks something isn't joined smoothly.
Again, I'm not looking for help on this particular model as much as I'm trying to trying to understand the principal of the problem (this comes up a lot) so that I can avoid the situation in the future; the fact that an example is so hard to come by lumps this error in with "the loop is bad", or "geometry conditions" - it's a technical problem that can't really be explained well.
Ok - did you knit all the surfaces? I just thinking there is a Gap or an Overlap somewhere for that not to work. This is a good question for Paul Salvador ??
Surfaces are knit together. No obvious gaps or overlaps. I suspect a bad face or something, but stringent geometry check is telling me everything is good.
Select all the surfaces and select the Surface Thicken Feature, what happens.
It thickens. I'm actually moving on - variable radius fillet will work to get it done. Like I said, I can get around this particular instance, but I just wanted to see if someone could provide an obvious example of what the error is talking about. Thanks for your help.
Matt Graves wrote: Surfaces are knit together. No obvious gaps or overlaps. I suspect a bad face or something, but stringent geometry check is telling me everything is good.
Matt Graves wrote:
..interesting.. no errors... maybe look at the usual suspects,.. shared vertex/edge.. and maybe turn on "curvature" or "zebra stripes" which could help locate visual oddities (red)?
Matt Graves My experience tells me that the problem is in these places. Avoid small faces that are completely removed as a result of rounding. Avoid building surfaces with lots of faces.
I try to understand the error messages if it is not possible I redo the process again with different dimension. This is the most common way I do. Because most of the error messages are cryptic.
..sometimes there's no message or the standard not able to mesage.. but lately in 2019 and 2020 I've seen errors which as Matt Graves mentions have no surfacing problems?
That mean you say that error messages are become reliable with enhanced versions. If so I will take this into account.
Retrieving data ...