Holding tolerances in V6


(Mikep) #1

I have been using Rhino going on 15 years and I have never had an issue with building surfaces using the sweep two rails command. I have always been able to build using two rails that have complex curvature. I use blend curves often when trying to get complex radii that can’t always be built well with the fillet command.

I have just started using V6 and I am really hating it. For some reason, it absolutely not hold ANY tolerances with blend curve and sweep two rails command. There are also many times, as I am designing and reworking surfaces that I have to rebuild a surface and then retrim. When I try to rebuild a surface now in V6, it will not even remain close to the original shape no matter what I do. On the first project I have worked on, I have had to go back and redo more work than I ever have before, just because of this issue. One more thing is that when I blend curves, the curves will not always go to the ends of both curves being blended. They will actually blend but move up or down the curves slightly to the point where they don’t even want to join.

All this being said, is there a setting that I am somehow missing either because there are new tools to set tolerances or I am just forgetting something settings that have always been there but may have different defaults than previous versions.

Any input would be appreciated since this is costing me a great deal of time and a lot of frustration trying to get projects done in a timely manner and with good surfaces.



I think we need some specific example?

(Pascal Golay) #3

Hi Mike - as Jim says, an example would be very useful in trying to sort this out - if you cannot post an example here (public) please send to tech@mcneel.com, to my attention.



(Mikep) #4

I’ve taken some basic examples to show what I am seeing.

In the first set of images I have taken a surface that I need to rebuild to trim differently. In the past i have always created a border form the surface as shown in image one. In image two, I rebuild the surface. In this example it shows the UV set at 6,6 and degrees at 3,3. I have tried this using UV points of varying degrees to see if I can do anything to hold that surface tight to the original. Again, I have been doing this for a long time and have never had an issue with earlier releases. I then take that border and, say, edit a corner, then trim the surface, I get my new surface. I have kept that border as a reference in this example. I create a new border using the new surface and do a curve from intersections to show how the original and the new border vary.

In the second simple example, if I take two curves with cross sections and sweep those rails, the surface is not holding tight to those curves and it results in a bad surface. I just used simple set of curves here but I know from past experience that I never had an issue unless I was trying to really get two curves that had tight corners to sweep and then I’d get tortured surfaces that are pretty ugly. In this case I, again, took curves used in the sweep command ad then curves built from the resulting surface and did an intersection to see the difference. The last image shows how much these curves vary.

The bigger problem is when I export as i have to do for a vendor that uses NX. Step and iges files, even if I dial my tolerances way up to make sure I have closed surfaces without gaps, almost always result in at least some small difference that means the vendor has to do some extra work to stitch those surfaces back together.

Like I said, maybe there is some setting that I have just neglected to adjust to get these tolerances to hold. I’ll be honest. I am self taught and have picked up some bad habits over the years and maybe it is just coming back to bite me but I have never had this hard a time until I started using Release 6. Any help is appreciated.

Other issues also involve blended curves (making sure I have picked the right ends so I don’t get curves blending from the wrong ends) and getting the resulting blends not match up to the ends and extracted isocurves extending beyond the actual surface.

Thanks for any help