'Wall Styles' to reflect real-world construction

[Amorphous - Julian 171]

01 THE ISSUE WE SEE:

 

• Wall tool works great for larger projects where 'wall types' are standardised to a limited number of variations.
   
• But in smaller, bespoke projects where two sides of walls have a lot of combination of finishes (think 'bathroom on one side, bedroom on other' ... etc), the number of wall types required to reflect these conditions gets out of control. 
   
• See enclosed images: 
  One one side of the wall  (see SIDE 1.png), we have a timber veneer finish bordering an adjacent mirror.  
  Even though the other side in both these finishing conditions are both paint finish (see SIDE 2.png), we have to create two different 'wall types' for this.
   
• You can imagine how the different combinations of finishes can cause the wall types to spiral out of control quickly.
   
• Plus, there are a lot of 'starting and stopping' of walls- Whilst in real-life situation, this wall would be built as one continuous core in the centre. 
   
• The other disadvantage of the current method (refer to the elevations) is, when rendered, a junction line that is supposed to separate the finishes would not appear in elevation.
   
• So, in order to show any kind of construction gap or joint between the two walls, we'll have to make yet another wall type, with a thinner 'component' that sets back from the finished surface, and we'd call this 'gap wall'
   
• In other words, if we want a 5mm shadow-gap between the said mirror and the timber wall, we'll leave a 5mm gap between the two said wall types, and  fill the gap between the two walls with a 5mm long 'gap wall'.
   
• This is not really a feasible way of building a BIM model. 

 

02 THE SOLUTION WE WOULD LIKE TO SEE:

 

• On a construction site, the 'core' (stud, masonry...) of the wall is always constructed first, and then the 'components' (gypsum, stucco, plywood...) are applied over them. 
   
• Importantly, the 'core' is constructed continuously. 
   
• It would be most helpful if 'WALL TOOL' is just the walls- giving us the ability to draw continuous 'core' walls, without 'stopping and starting' the walls to accommodate the finishes components.
   
• If 'WALL TOOL' draws continuous 'Core Walls', there should be a 'FINISHES COMPONENT TOOL', where users can create 'Finishes Build-Up' styles that can be 'snapped' onto the 'Core Wall'.
   
• This way, we can stretch the 'Finishes Component' along the surface of a 'Core Wall' long as it needs to be, and even have the option to add 'skirting, flush skirting, cornices' or the like to finish them off. 
   
• To resolve the issue of displaying junctions between adjacent 'Finishes Components', perhaps we can be given the option to 'show/hide joint between Finishes Components', or even have the option to add trims between them. 
   
• This concept would work great if the Doors and Window objects can intelligently penetrate these 'Finishes Components'.
   
• The final point we would like to say the above method can be advantageous because it will allow each 'Finish Component' to be automatically tagged with a 'Finishes Tag'.
  This can be done by picking up on the properties of each 'Finishes Component' 

 

Edited November 28 by Amorphous - Julian

[CipesDesign 76]

I gave up on Wall styles a long time ago, for this and other reasons. You can do what you describe fairly easily using unstyled walls and then simple 3D modeling techniques.  I recommend using discreet classes for all object types (eg, mirror, tile, etc.)

[Amorphous - Julian 171]

We do use separate classes for each finish (GL01, GL02 classes for different types of glass... TL01, TL02 classes for different types of tiles). 

In fact, the above elevations were generated by the 'walls components' picking-up up on the different 'renderworks texture surface hatches' of the relevant 'class renderworks texture' (a long, round-about way of achieving 'wall surface hatch by class')

 

Yes- we can use unstyled walls with two 'finishes components' on either side of the wall, but it doesn't resolve the issues of the 'start and stop' and having still to define two different 'finishes components' on either side of the 'core wall'. It becomes the same multiple small, short walls as described. 

[Matt Overton 233]

9 hours ago, Amorphous said:

So, in order to show any kind of construction gap or joint between the two walls, we'll have to make yet another wall type, with a thinner 'component' that sets back from the finished surface, and we'd call this 'gap wall'

Shadowlines, movement joints and even precast concrete joints can be done with a symbol that only half breaks the wall. Then skim the wall with another wall for finishes.

 

Oh I don't agree larger project don't have these issues. Once worked with an Architect who wanted to put If-Then-Else statements in a wall schedule to avoid adding more types.

 

Still, as you say, walls could work more like real world and make life easier.

[Jonathan Pickup 212]

other people have suggested that the walls need to have a serious rethink.. One of the options that people talked about was using walls with materials rather than components. When you think about the actual construction of a wall, you might notice that sometimes you have a framed wall that might have different thicknesses and internal finishes but the outside has a single external cladding. Using wall styles can often end up with the exterior having breaks in the external cladding.

 

I have recently been using another CAD program. The way I have been modeling in this program is to create a call wall with an internal finish as a Wall style then cladding the outside of the wall with a Covering. The way the covering works is to create a true 3D wallcoverings. In many cases, I tend to use either vertical metal cladding or horizontal timber weatherboards (siding). The CAD  program, in this case, creates sections that show the true 3D nature of my walls. To replicate this in Vectorworks we could create the core and internal lining with wall styles and then draw another exterior wall for the cladding. We would have a problem with windows and doors that they would not automatically update when we move them in the core wall. The Program that I was using updates the coverings if we move doors or windows.

 

[Amorphous - Julian 171]

@Matt Overton great tip for wall breaks! Thanks for that!

@Jonathan Pickup interesting... but a bit hard to understand without screenshots. Can you share some screenshots? Is this other software package Revit?

[BG 70]

I think having the cladding/wall/roof covering as a separate item - yet is smart enough to recognise windows and doors etc is possibly a good approach to consider going forward. This may also open up the opportunity to have 'proper' claddings eg. weatherboards, roofing trays etc.

[twk 199]

Interesting points by the OP's on adding an option for a finishes components.

 

The other CAD program most definatley, must be Archicad, with its cad image addon:

This package is what I wish VW Architect was.. everything about it was very well thought out for the Architect..

If you have time check out the video's on youtube for their Revision system, and issue management system, its so good. I've developed in-office plugins/scripts to mimic some of that functionality, together with the auto-claddings. But I wish vectorworks developed these natively as they are so essential for architects.

https://www.youtube.com/user/CadimageTools/videos

 

This is not a promo for another CAD software by the way! I enjoy Vectorworks alot, this is just me saying, there's room for improvement.

[Amorphous - Julian 171]

I would like to revive this post, as this is an important topic in the workflow of documentation. 
Creating tens or hundreds of wall types based on 'buildup' or 'finish type' is untenable. 
We need to use BIM as a design tool during the design process, so the team at Vectorworks needs to think about how do we quickly change the interior finish of a room, without a complex process of creating new wall types to accommodate this. 
I hope someone can call Matt Panzer's attention to this issue.