Jump to content

joining roof faces


Recommended Posts

I don't do any math at all and can build very complex roof structures...I can't understand why bringing roof planes together is so difficult. Using the geometry and haveing a basic understanding of objects in 3 dimensions (i.e. opposing planes) should be more than enough to solve planer collision issues in VW because they have allready done the math. It's up to you to make sure that your roof works....I kind of like it that way.

Maybe I'm wrong...but I think the time will come when you will be able to click one surface and tell it which plane to collide with and the system will cut, trim and add objects to roof plane collision. In the mean time put your calculators down, work with the objects you have and let VW do all the math for you.

Just my thoughts...

Pete A

Link to comment
  • Replies 51
  • Created
  • Last Reply

Top Posters In This Topic

Thank you Robert. I tried this but the results are less than desirable. I created two hip roofs that crosed each other. The Clip action cliped at the eave of the roof and not at the valley.

I had hoped that the roof would cut at the valley line.

I am using 12.5.0, not 12.5.1.

Link to comment
As long as one does hip roofs it is indeed very easy. I've never designed one. Adjusting individual roof faces in complex situations (eg asymmetric roofs) is anything but easy.

I would have to say that more than 75% of the roof structures I do are "Asymetrical" (dual pitch). This is where the power of VW really plays out. Hip, Gable, Polynesian, Gambrel, Mansard, Pavilion-hipped, Pagoda, Shed...and all done using only what is available at present in VW....no math calulations from the user.

In any asymetrical roof face collision you will find the hip, valley and ridge can be found using the geometry of the planes in VW.

And it is also true that the method Robert suggests to assit in locating collision vectors using the clip operation works to locate many other derivitave type non-conventional roof structures with 3D solids and roof objects working together.

Link to comment

Well, maybe. I can't figure out how to easily create any modern, European roof shapes quickly. I couldn' care less for Mansards or Polynesians.

By asymmetrical I don't mean a hip roof of any type.

EDIT

Let's see if I can get an image here.

roof_problem_1.png

Ahh - for once it worked.

So, without a calculator I can't figure out how to match the two roof faces.

Edited by Petri
Link to comment

Petri, your first example is, as I see it, an issue of design rather than CAD. The second example can be fairly painlessly solved.

Roofs.JPG

Create the roof faces of appropriate slope (step 1). Convert a copy to lines (hidden) (step 2, red lines). Add the missing hip (step 2, yellow line). Create polygon and use boolean operations to modify roof faces (result = step 3).

Link to comment

The correct answer is to change the location and plan angle of the axis line (and/or z-level?) so that the two determining points (of the brown roof) are retained, the grey roof is reconfigured, but not split.

And yes, it is doable - with a calculator. We have three known points, so the plane can be determined.

Link to comment
Petri, your first example is, as I see it, an issue of design rather than CAD. The second example can be fairly painlessly solved.

Create the roof faces of appropriate slope (step 1). Convert a copy to lines (hidden) (step 2, red lines). Add the missing hip (step 2, yellow line). Create polygon and use boolean operations to modify roof faces (result = step 3).

Your argument was, however, that one can create any roof with the roof tools.

Link to comment
Your argument was, however, that one can create any roof with the roof tools.

Petri, actually I didn't make that claim. The above illustration shows exactly what I and others have been saying. It uses the roof tools (both roof object and roof faces can be useful), hidden line conversion, and boolean operations. But not (and this is the important point!) a calculator.

Equally important is the point that started this thread: if this can be done manually, it could be done by software. Joining roofs faces using the "Connect/Combine" tool would a great improvement.

Your first problem, as clarified, requires calculation of a horizontal spring line, not parallel to the wall, as a basis for the skewed roof plane (since all roof objects start from the assumption that the eave is level). This problem is "asymmetrical AND irregular!" Once the plane is created, you can edit it to fit the building edge, resulting in an angled eave line. If I understand you correctly. In this instance, getting the correct initial plane orientation would not be facilitated by the ability to Combine/Connect roof faces. One approach to a CAD solution would be to create a NURBS plane on the 3 determining points, then intersect it with a horizontal plane to determine the "virtual eave" of the roof face.

PS: BTW, thanks for the VectorScript solution to my layer-switching problem. I haven't had a chance to try to implement it - I think I basically have a mental block about using VS! But I will get to it someday - have to get back to work now . . .

Link to comment
Your argument was, however, that one can create any roof with the roof tools.

Petri, actually I didn't make that claim.

Pretty much so.

Equally important is the point that started this thread: if this can be done manually, it could be done by software. Joining roofs faces using the "Connect/Combine" tool would a great improvement.

Precisely.

This problem is "asymmetrical AND irregular!"

Very regular to me and many others. Any architectural magazine will have buildings with roof forms like this - but much more complicated, of course. Even this very simple shed-sized example is too much for VW's roof tools.

One approach to a CAD solution would be to create a NURBS plane on the 3 determining points, then intersect it with a horizontal plane to determine the "virtual eave" of the roof face.

Actually, these examples do not require anything that complex. They are simple planes. We have three points wityh fixed x/y/z -coordinates, then n points with fixed x/y coordinates. High-school maths. The difficulty arises from the silly assumptions of the roof tools: that eaves are horizontal and that one knows where the tilt axis is & what the angle of the roof is going to be.

Going through the gymnastics you describe is not helpful or feasible in the design stage.

Link to comment

Petri, to quote from my original post above:

"When you convert the roof to lines as Peter suggests (I generally convert a copy to lines and edit my roof faces, rather than do total reconstruction), you will see that there is no line at some valleys. But you can construct that line based on the other lines you can see. Then you can correctly edit the shape of the roof face polygon."

Tell me how that is different from the worked out example.

"Irregular" is by no means necessarily unusual or undesired! Just a description of geometry.

Creating a NURBS surface and intersecting it with a horizontal plane can be done in under 1 minute. The reason I use NURBS is not because of the free-form shape, but because it is easier to intersect the planes with tools provided by VW.

Link to comment
Tell me how that is different from the worked out example.

Probably it doesn't.

"Irregular" is by no means necessarily unusual or undesired! Just a description of geometry.

Of course. The roof tools are designed in such a manner that they can't be used to create irregular and asymmetric roofs; thus, no roof tools for someone like me.

Creating a NURBS surface and intersecting it with a horizontal plane can be done in under 1 minute.

Does this in creating a roof with roof tools?

Link to comment

Exactly, Mike. Now, I don't expect an object or a tool to do anything and everything. If a cheap tool works for 80% of users in 80% of their projects, we have a winner.

Nevertheless, the "roof object" and associated "accessories" are an embarrassment and liability in Finland (and, I believe, in most European countries). EDIT I refrain from speculating Japan and Asia - but I have a hunch...

With the development costs of something completely useless, NNA certainly could have developed a tool for creating the geometry of roof faces from 3 known x/y/z coordinates and any number of user-definable x/y coordinates.

Unfortunately, dormer windows (useful for less than 30% of VW-users) have been the priority. If there's any irony in this, it must be the percentages: the majority subsidising the minority.

If wall styles were not available only in Architect, I would never encourage a Finnish architect to buy anything but Fundamentals. I really have to stretch my credibility and twist my integrity as they are not stupid enough not to realise that they pay for useless roofs, doors, windows and stairs, not to mention canned wall styles, dreadful textures, hopeless symbols and so on.

Edited by Petri
Link to comment

Petri, I might use more generic 3d objects to model a non-traditional roof. The VW roof object is only useful because the parameters for a "traditional" roof are relatively simple and proceed from a set of tradtional assumptions. As a 3d solid, it doesn't have any characteristics that make it superior to an extrude or a set of NURBS surfaces.

There is a set of tools in the "3d power pack" that make maniupulating NURBS surfaces possible. But far be it from me to suggest that they couldn't stand a great deal of improvement! They are not terribly robust or user-friendly (compare to Rhino).

The question of whether the roof object / tool can or should be made more flexible to meet other needs (yours, for example) begs the question of whether those needs can be encapsulated as a set of parametrics.

Link to comment
Petri, I might use more generic 3d objects to model a non-traditional roof.

We don't have even traditional roofs that could easily be modeled with the Roof object. Roof faces - yes, mostly, but not without the calculator.

The question of whether the roof object / tool can or should be made more flexible to meet other needs (yours, for example) begs the question of whether those needs can be encapsulated as a set of parametrics.

Good question. No, preferably not as a set of parametrics, but anything I can do with a calculator and high-school maths (almost 40 years ago) should be doable with a set of roof tools. If I can do the maths, surely the talented Mr Anderson can...

Link to comment
Draw two (overlapping) roof faces and select them. Choose the "Clip Surface" command. The foremost roof plane will clip the rearmost, just as if they were polys. Most path-based objects (objects drawn based on a closed shape) will operate this way in 12.5.1

I do not get the same result. The foremost roof plane keeps its geometry unchanged and merely acts as an outline to cut through the rearmost in top plan resulting in two separate roof planes one of which has a square hole in it with the 2nd one passing through said hole.

Link to comment

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Restore formatting

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.


×
×
  • Create New...