OSTexo,

I also thought Michael would have pitch in but I guess he's buzy with the new beta release, it's perfectly fine with me.

I appreciate your tutorial and your answers, if it works fine for you why change anything. But it could be a bug also, I just find having to do this kind of extra step is not very "MOIish" to say the least.

By the way, I happen to have an old Machinery's Hand Book 23 revised edition page 1481 "screw design profile" that I could send you if still needed.

Thanks for your time,
Felix

Hi Felix, could you maybe give me a quick description on what the extra step was, it was ShrinkTrimmedSrf is that correct?

What was the command being used right after that, was it the new Flow command or something else?

In the first v3 beta Flow was sensitive to the size of the underlying surface so it could help to use ShrinkTrimmedSrf on it, but this is changed for the new v3 beta which only uses the visible portion of the surface for Flow so it's basically like having ShrinkTrimmedSrf built in to it automatically.

That's the only thing that I can think of other than control point manipulation where ShrinkTrimmedSrf would be particularly significant.

- Michael

The question was on the rotation of the cylinders seam edge before boolean difference, to move the seam edge out of the other seam edges to prevent (I'll coin my new phrase here) "SEAM SLITHER", where two seam edges that may skim in and out of each other can cause boolean issues, or create segmentation that can cause the filleter to fail...

Oh yeah, yeah that's kind of a good practice overall because the more things you have running into and overlapping each other tends to make things more difficult for surface/surface intersection calculations.

So anytime you have a chance to reduce the complexity of an intersection by possibly rotating some particular seam edge out away from where it makes things more complicated is a good idea.

It would certainly be nice if it wasn't necessary, but setting things up with less complexity usually makes for less chance for things to get confused about 2 barely skimming things that you are trying to intersect.

Also it's not great for the filleter to have a bunch of complex diced up edges rather than a more simple edge structure. Filleting is a pretty sensitive area so it's usually a good idea to try and give it a more simple edge structure to deal with if it is possible to do so.

- Michael

Hello FelixPQ,
Can you scan and send? I'm sure there are some neat tidbits of information in that guide.

Hello Michael,
Thanks for the explanation on the seams, I'm not completely nuts (yet).

Hello Burrman,
Copyright that quick.

Hi Michael,

yes, ShrinkTrimmedSrf was one of the 2 points I raised. I look at the video again and ounce he did all the boolean diff, he separated the resulting solid even if every thing looked fine, use shrink and rejoin everything back together into a solid. Since he activated show points, it was obvious the shrink command had an effect. If I remember correctly, this was threaded part of the screw and he did the same procedure on the head of the screw before using union on the 2 parts.

I just wonder if these additional steps where really needed and or useful in some way. Having the control points on make's it obvious that shrinking provides a nicer underlying structure limited to the natural or actual boundaries of the object and it may be a good enough reason to do it but does it help down the line with other operations?

As for the other point I raised about why he rotated the cylinder, this was answered and I understand better why now.

I have a bad memory so I wouldn't know if these kind of detail or things to whatch for are mentioned in the docs somewhere, if not then maybe it could be a good idea to group all those little whatch for thing and maybe have some suggestion on how to prevent problems down the road.

Thanks,
Felix

Hi Felix, re: shrink trimmed surface -

> but does it help down the line with other operations?

No, not typically - under normal circumstances it should not make any difference if you shrink surfaces or not.

It could occasionally happen that there would be some difference due to a bug or sometimes shrinking can actually kill off some badly formed area of a surface like some area where it's self intersecting or jumbled control points or something like that. If that happens, then it can make a bigger difference, like for example if you tried to do an offset and some not immediately visible area of jumbled control points will make the offset go crazy trying to track along a surface normal that is bouncing all over the place, but if that area happens to get removed by ShrinkTrimmedSrf, then doing an offset after that could produce a better result.

One side effect of shrinking is that it will reduce file size by some amount though, because it discards some extra areas of surfaces and that reduces data size by some amount.

Another side effect though is that you won't be able to do an "untrim" (delete trim boundaries) to recover the full original surface that was being cut up - sometimes being able to recover the original surface can help when trying to repair objects.


If you want the smallest possible file size, then I guess that's one reason to do it...


Normally I would not think that you would see any difference from doing it or not - the main exception is the Flow command from the first v3 beta which uses the full underlying surface. That's not the case with Flow in the latest v3 beta anymore though.


- Michael

Hi Felix, one other thing I should mention is that showing control points itself (not necessarily shrinking surfaces) can be useful as a sort of diagnostic function to try and get some more information about the structure of a surface.

For example if something is not behaving right one of the things I will usually do is to view the control points of the surfaces to see if I can find any warning flags like strange clusterings of points or stuff like that.

Don't know if that applies to what you are asking about...

- Michael

Sometimes it can also be used as a workaround - if some operation is having some difficulty handling a particular surface calculation shrinking it can sometimes change it enough to happen to avoid the particular problem.

That doesn't necessarily mean that it's always _better_ to do a shrink - it's just that it can possibly change things enough to make calculations behave slightly differently, sometimes maybe avoiding a problem sometimes maybe running into some other problem.

But if something is not behaving well, it can be worth a shot to try it to see if that particular thing then behaves better.

- Michael

Michael,

so basically, file size, could remove potential issues with some command and it can't be reversed. Hopefully, it can't cause problem of its own either.

Thanks,
Felix

A really good tutorial, especially for beginners. I had some spare time today and had a go. Here is a render from Modo (25secs):



It's a 10mm bolt. I guessed the head size as a I couldn't be bothered to look it up, so it looks a bit small.

Spooky realistic Steve.
I like the Galvanized material.

Hello SteveMacc,

Great model and render, the material choice is right on.

its a pleasure to watch a great moI tut with audio

wish there were more

thanks

Hello,

A tutorial on knurling is in the works, it should be out shortly.

With you a bolt becomes an true artwork, an admirable objet.

You have the precision of one watchmaker or clockmaker; same if I nearly missed to sleep during your tutorial; a little too technical & too long for a lively mind full of impatience.

Hello,

A knurling tutorial should be up tomorrow. It's shorter than the screw thread tutorial.

Ok a bit late but I wanted to ask this ,

What I don't understand, in the beginning of this tutorial, you tell us to make 20 copies of the 1 inch line.
After some more steps you delete all the copies and never use them. What is the meaning of those 20 copies? :-)



Peace,
Rich_Art. ;-)

Hello Rich_Art,

It was an easy way to determine thread pitch. By using this method you have an easy way to get the pitch right on and will use that to determine the other parts of the geometry. You don't have to spend your time figuring out or inputting to many values, you just need to know how many threads per inch / mm you want to model. Hope that makes sense, I'm not so good at explaining things.