Full Version: When using Flow, target surface splits

Hi All,

I'm just practicing the Flow command and although it's worked several times before, in this example once I click on the object, then base surface and then target surface, it seems to 'trim' the target surface in 2:



I tried redoing the surface a few times but each time it splits it.

What am I doing wrong?

Many thanks,
Gunter

Attachments:
Dome exp.3dm

---

Hi Gunter,

re:
> in this example once I click on the object, then base surface and then target surface,
> it seems to 'trim' the target surface in 2:

It's because your target surface is made up of 2 surfaces that are joined together, it isn't just one large single surface.


> I tried redoing the surface a few times but each time it splits it.

Could you please give some details and a file for how you tried to redo the surface?

The reason why it would be split up kind of depends on the method used to construct it. It could be something like your construction curves are also split into multiple segments instead of being one single segment, or segments that have a slight crease between them instead of being smooth where they touch.


One other thing to note, if you make the base surface hug the object more closely then you could use the Projective mode for flow in this case, projective mode can use a joined surface as the target instead of only a single surface:





- Michael

Image Attachments:
gunter_projective1.jpg  gunter_projective2.jpg 

---

Hi Michael,

Thank you for your quick reply!

I created the surface by sweeping to half ellipses around the ring rails:



I don't think there's a split there.

As for projection, happy to use that function but in the screenshot you attached, it looks like the object changed its shape when it flowed on to the target surface, is that right?

Many thanks,
Gunter

Attachments:
Dome exp.3dm

---

Hi Gunter,

re:
> As for projection, happy to use that function but in the screenshot you attached, it looks
> like the object changed its shape when it flowed on to the target surface, is that right?

Yes, but the main purpose of the Flow command is to change the shape of an object.

- Michael

---

Hi Gunter,

re:
> I created the surface by sweeping to half ellipses around the ring rails:

When you make a closed surface there is what is called a "seam edge" at the surface's starting/ending point.

In this case you are going to want the seam edge to be at the bottom area instead of the top area. You can make that happen by rotating your circle rail curves by 180 degrees around their center point before doing the sweep.

Or also there's a plugin here that you can use to relocate the seam point on a closed curve:
http://moi3d.com/forum/index.php?webtag=MOI&msg=10125.8

The seam edge of the sweep surface will be located at the seam point of the closed rail curves.

- Michael

---

Hi Michael,

Thank you, I installed the plugin and it worked! Apologies though, I can't say I understand the seam concept and when to know to use it. I had a look at the thread where the plugin in was posted, but that went over my head, Im afraid.

Many thanks,
Gunter

---

HI gunter

请在操作步骤的的基础上增加类似这样的工序。(Please add a process similar to this on top of the operational steps.)

这是因为moi的join和boolean union无法在所有的情况下都形成surface在这篇讨论中：http://moi3d.com/forum/messages.php?webtag=MOI&msg=11426.1 有相应原因的说明。
This is because the join and Boolean union of moi cannot form a surface in all cases. In this discussion There is an explanation for the corresponding reason.
因此你需要在完成圆环的缝合曲面后，提取边缘和曲线，再一次进行扫略，让曲面变得完整。
Therefore, after completing the stitching surface of the torus, you need to extract the edges and curves, and then perform another sweep to make the surface complete.

Image Attachments:
zz.gif 

---

Thank you so much Suiyun!
Gunter

---

Hi All, so I've now reached the point of flowing the lattice work on the solid ring:



The next step I'm trying to figure out is how to substitute the top of the solid ring with the lattice so one can see right through it. I tried various Booleans but none produce the results I'm looking for. Essentially, I'd like the lattice shape to be the top of the ring and seemlessly merge/blend into the rest of the bottom section of the ring.

Please could someone explain how I can go about this?

One more thing! In the next screenshot, you'll see there are gaps in the actual flow. I tried to use the 'Uncurve' script to matche the width and length of the ring curves so they flow accurately along the edges of the ring but that didn't seem to work. Any guidance on this would be fantastic, thank you!

I ran out of attachment space, so here is the 3dm file:

https://drive.google.com/file/d/1EUmq0Yrnm36hqdAlBodhls02OCctsn-q/view?usp=sharing

Many thanks indeed!
Gunter

---

HI gunter

LIKE THIS


moi无法进行太多对象的布尔运算，因此最好不要使用孔洞进行布尔运算，而是将整个装饰复制一份，隐藏，并对整体进行布尔，删除多余部分，显示。
（Moi cannot perform Boolean operations on too many objects, so it is best not to use holes for Boolean operations. Instead, copy the entire decoration, hide one of them, and perform Boolean operations on the whole, delete the excess parts, and then display all.）

Attachments:
Bulbous lattice ring ok.3dm

Image Attachments:
zz.gif 

---

Hi Gunter,

Is this what you want?



I deleted the outside faces of the solid lattice so all that remained was the holes (surfaces):



Then I selected the solid ring and choose the hole surfaces as the cutter. Had to wait maybe 30 seconds for the boolean to finish, but that left me with this:



And then I deleted the highlighted fragments.

- Peer

Image Attachments:
after boolean diff.png  cutter.png  finished ring.png 

---

Suiyan, thank you! That's very helpful indeed!

---

Hey Peer, Yes, this is exactly what I'm after, thank you very much!

---

别客气，我想你该对最终效果有更清晰的认识。就是到底哪些部分应该被布尔，哪些部分被保留。这种孔的造型最好的方式应该是在还是平面时拉伸那些孔的封闭曲线（双侧），让这些柱体高度完全能穿透戒指圈的目标实体，然后对它们进行流动，使这些柱体流动至戒指圈的目标体→布尔。

这可以最大程度的简化并且避免布尔出现错误结果，虽然这次的布尔很顺利。

（Don't mention it, I think you should have a clearer understanding of the final effect. Which parts should be Boolean and which parts should be retained. The best way to shape these holes is to stretch the closed curves (on both sides) of the holes when they are still flat, so that the height of these pillars can fully penetrate the target entity of the ring ring, and then flow them so that these pillars flow to the target entity of the ring ring → Boolean.
This can simplify and avoid errors in Boolean results to the greatest extent possible, although the Boolean process went smoothly this time）

---

Thank you Suiyan, yes, this is exactly what I tried to do and it helped.

The other thing I experimented with was rather than flowing the lattice, I extracted the diamond shapes from the lattice and extruded those. Then Flowed those onto the ring and then Boolean difference.

It's fun to try different methods!

Thanks again!
Gunter

---