Generating relief on top of torus

[sketchymick]

@unknownuser said:

@sketchymick said:

i am more_likely_than_not misinterpreting what you're saying here but if the 'rib' your referring to is the smaller 'rope' going twice around the torus -and- by 'solid' you mean a solid group (sketchup solid) then jean has already accomplished that..

No you're not misinterpreting it. I was just curious as to whether there was an elegant solution, rather than having to manually stitch up the ends (Johnny's method). Of course the pragmatic way is to spend 30 seconds to stitch it up like this, but the purest in me is searching for the combination of facets that would avoid needing to do this. It's pure curiosity.

[gilles]

Here is a way:

---

---

---

---

---

---

---

---

---

---

---

---

---

[Jean Lemire]

Hi Gilles, hi folks.

Very clever.

Best regards.

Brillant.

Salutations.

[gilles]

You are right, I forgot one step.

Thanks to clarify.

[cotty]

@gilles said:

Here is a way:

Nice and detailed solution!

[sketchymick]

@gilles said:

Here is a way:

Nice !!!!

[jeff hammond]

@gilles said:

Here is a way:

[gilles]

Hello Jeff, had a good night?

[pilou]

Seems Follow me and rotate by Wikii make something similar ?

[jeff hammond]

@unknownuser said:

Seems Follow me and rotate by Wikii make something similar ?

[attachment=0:26zw3j75]<!-- ia0 -->far.jpg<!-- ia0 -->[/attachment:26zw3j75]

hmm.. maybe component stringer then curviloft can do the trick too?

(can't check right now because I'm on a bus but doesn't component stringer have a rotate option?)

[jeff hammond]

@gilles said:

Hello Jeff, had a good night?

ha.. yeah right.. 3hrs sleep

[pilou]

@unknownuser said:

but doesn't component stringer have a rotate option?

Component Stringer by Chris Fullmer
Yes it has, it's even its principal quality
It's only default is the no intuitive position of the Axis default

[gilles]

surely Rhino Muhahahahah.

[jon bondy]

I created this in what I assume is the normal fashion: started with a torus, drew a line that wrapped around the torus twice, removed one line segment, created a surface orthogonal to the line end, drew a circle, and used a follow-me to create the "pipe". I then attempted to stitch the ends of that pipe together with a series of manually drawn lines to create triangles. I gather that this attempt failed. The original instructions said to use Skin to stitch the pipe together, but I could not get that to work at all.

So, I cannot see how to fix this helix. And I gather that the original torus also has a defect, but I've not had a chance to tend to that.

Again, any hints would be appreciated.

Jon

---

James Torus 1b2.skp

[jon bondy]

@jean lemire said:

Hi Jon, hi folks.

See this SU file for ideas.

Jean: this is what I attempted to do, but I was unable to close up the helix successfully.

Jon

[jeff hammond]

@gilles said:

surely Rhino Muhahahahah.

after Jon gets all his issues sorted out, maybe I'll put up a little vid showing a couple of ways

[jeff hammond]

@jon bondy said:

I created this in what I assume is the normal fashion: started with a torus, drew a line that wrapped around the torus twice, removed one line segment, created a surface orthogonal to the line end, drew a circle, and used a follow-me to create the "pipe". I then attempted to stitch the ends of that pipe together with a series of manually drawn lines to create triangles. I gather that this attempt failed. The original instructions said to use Skin to stitch the pipe together, but I could not get that to work at all.

So, I cannot see how to fix this helix. And I gather that the original torus also has a defect, but I've not had a chance to tend to that.

Again, any hints would be appreciated.

Jon

it's hard to say where you went wrong but the helix will close.. your latest uploaded drawing has multiple faces on top of each other.. if you're stitching by hand, it's best to go into it with a systematic approach then you just have to be a robot instead of creating a mess.. (but like i said, it's hard to see where you went wrong or why all those extra faces were in there..

personally, i think i've done more than enough hand stitching in the past so i opted for a plugin to close the gap ..
i deleted all the stuff in that segment then selected both raw circles and used curviloft..
http://sketchucation.com/forums/viewtopic.php?t=28586

i also deleted the center path line, grouped the helix and it's registering as a solid (no holes or extra faces etc..)

jamestorus.skp

the torus (which is hidden in this model for anyone wondering where it is) looks good to me.. if you're concerned about it not registering as solid, it's because of the circle path line running though the middle of it.. delete that line (or remove it from within the component) and it too will be a solid..

.

[jon bondy]

I tried Gilles approach, and when I got to step 11, the two parts did not line up. I modified the component by re-rotating the top until it looked OK, but I doubt that I got it spot on. The resulting torus looks nice, but is not solid at all. Boy, those explode and group steps take 10-15 minutes!

I was also not sure exactly how I was to do the smoothing in step 12.

Jeff: curviloft made quick work of joining the ends of the pipe together (thanks!), but the resulting group is still not noted as being solid (it just says "Group (1 in model)"). Ironically, I already had curviloft, but I have so many tools that I can't keep track of which to use when!

Thanks for all of the help!

[jeff hammond]

the curviloft created segment will be in its own group immediately after using the plugin.. so, you'll have to explode it and get it within the rest of the helix's contex in order to make it one group.. also, make sure the centerline path isn't in the group because that will prevent it from being solid..

a solid group will only have a manifold (water tight) outer shell with no extra geometry floating around inside

[jeff hammond]

@jon bondy said:

I tried Gilles approach, and when I got to step 11, the two parts did not line up. I modified the component by re-rotating the top until it looked OK, but I doubt that I got it spot on. The resulting torus looks nice, but is not solid at all. Boy, those explode and group steps take 10-15 minutes!

here's a slight variation stolen from gilles' method that might give another way to show what he's done .. this starts at around step 6 of his tutorial..

the number of segments around the torus probably vary between gilles' and this one meaning the angles will be different but it doesn't really matter for the example.... the main circle i'm using for the torus has 48 segments..

also, you mentioned earlier about not knowing how to do the soften/smooth steps… select all the stuff you want to soften then right-click -> soften/smooth (or via the menus… Window -> Soften Edges