Jump to content
Christian Fekete

Construction details - Drawing and reuse

Recommended Posts

I am dealing with a workflow issue regarding the drawing of construction details. I am trying to solve maybe too many issues at once and I may have to make a choice on which is most important.

I am trying to create a variety of details for residential construction, here are the issues:

Whether to create 3D(part of the building model or separate of) versus 2D details

If using 2D, do I annotate the design layer or the Viewport annotation layer

How do I create a reusable set of construction details, should I include the notes in my symbols?

I usually ending up doing all of the above which is very time consuming, 

How do you do it?

Share this post


Link to post

Christian,

 

I will say first that I feel your pain. Figuring out what to do with details was a large stumbling block for us too, and honestly still is a little...although not as bad. My recommendation is figuring out a workflow that works best for you and accomplishes as much of what you want as possible and then sticking to it...as much as possible.

 

For us, this meant going with a 2D symbol methodology. Each detail is a 2d symbol containing the annotations, notes and linework...drawn at a specific layer scale. Each detail symbol sits within a unique vectorworks file in a standardized library location. When needed, that file is copied into a project file and the 2d symbol is referenced (referenced symbol methodology) into the L-Details file with all the other details, but placed on a specific layer with a layer scale matching that of the detail. This way the standardized location stays constant allows for project specific updates of details, per project without altering the standard detail.

 

We actually use very "dumb" text annotations (no callouts, general notes, or keynote legends) and page-based 2d symbols where necessary. That is because the database management becomes quite cumbersome when combining different details.

 

We have started experimenting with 3D details and feel that this is where the largest strides can be made for us and Vectorworks going forward, but I honestly wouldn't prescribe our current methodology yet.

 

Hope this helps.

  • Like 1

Share this post


Link to post

Thanks Eric for your insight and taking time to respond, it makes sense and seems the easiest to implement. My only question has to do with the reference workflow here. What exactly do you mean by the L-Details file, is it a separate detail file or is it part of one complete file for the project you work on. In other words do you create separate file for every sheet of your CD package?

Share this post


Link to post

No, the L-Details file is our production set or sheet set file. It contains several different design layers at the various different layer scales used to create the details (again, each detail symbol is housed in its own separate file). We then reference the detail symbols into the L-Details file and place them on the appropriate design layer matching the scale of the detail.

 

From there you can create individual viewports to take your details to the correct sheet layer. We often have 10 or more detail sheets, each displaying details at a variety of scales but generally organized by type. Each viewport (between 6 to 9 per sheet) could be at a different scale.

 

Note: There is a difference between a reference viewport (xref for AutoCAD comparison) and a reference symbol. Really any Resource (symbol, hatch, linetype, gradient, record format, etc.) can be referenced from one file to another. In the details case this actually allows our office, since each individual detail has its own file, to work on each detail independently from the file where details are being placed on sheets. Simply updating the reference to that file, updates the symbol to show any updates or changes.

  • Like 1

Share this post


Link to post

Once you make changes to a detail symbol (from either the original/mother location or in the referenced/child location) you should update the file reference exactly the same way you would update a Viewport/Xref style reference...from the referenced file tab. Out of date file references will appear in Red.

Share this post


Link to post

It would be really helpful if VW could set up a template or provide a third place where all these 2D or design layer viewport details (at different scales) could reside. 


We've created a seed file that has a bunch of grids at different scales, in the design layer that is out of view from the model. We try to organize the design layer into zones for 2d work and 3d work.  But this kind of gets all wacky when you have things at different scales, and the unified view options make matters more complicated and confusing. I don't think I could ever get my office to work with symbols and references of details, unless it was really easy to use. 

 

It's kind of like we need a database for some of this stuff that is well integrated into the program. 

 

It would be interesting to see how others organize their workspaces. 

 

 

Edited by cberg
  • Like 1

Share this post


Link to post

Perhaps I'm stuck too much in the past but for details that are common place, paving sections, planting details, things that are repetitive I and my clients usually use 2D files that are standardized in scale and content in an effort to give a uniform appearance and cover the detail sheet uniformly.

 

I have encouraged my clients and I have personally implemented using 3D when creating a unique item for a specific project. In this way I can communicate the idea to the client in 3D models with textures and materials applied. As the 3D model matures and we move forward we continue to refine the 3D model and add parts to some of the detail sections as shown in the attached. This planter started as a simple extrude and evolved to be in the final construction documents. The attached sheet has a combination of what Eric Berg recommended and the 3D approach with one variation. My client wants the drawings customized per project to account for variables like paver thickness or soil composition therefore they do not reference the standard details but import them into there files.

 

Hope this helps.

conceptual image.png

L-503-Roof Terrace Wood Fence & Community Planters.pdf

  • Like 1

Share this post


Link to post

I like the idea of separate 2D files :)

 

As there is no real link between details and 3D Model changes anyway and you

have to keep your details current manually.

Isn't it possible to just copy over a generated Viewport from the 3D Model to

your 2D Detail File from time to time as a drawing underlay to start your Detail

or check for geometry changes ?

Share this post


Link to post
12 minutes ago, zoomer said:

Isn't it possible to just copy over a generated Viewport from the 3D Model to

your 2D Detail File from time to time as a drawing underlay to start your Detail

or check for geometry changes ?

 

I have started doing this process with 3D details. Creating section viewports and exploding them to reintroduce into the 2d detail file. Definitely not idea BIM workflow, but one way of incorporating 3d elements into construction details.

 

I think, in the ideal world, that everything, details included would be object based. This means that, in Robert's example, if I spent the time detailing a specific planter in one project based on the 3d model, that the associated details, worksheets, specs, and any other information would somehow come with it. Imagine if you could standardize your materials and symbols so that everytime you add or update a feature in you model, the associated information also updated somewhere too. Finding a way to link construction details, technical specifications, and any other object specific data has the potential to be the most dramatic time-saver workflow.

  • Like 1

Share this post


Link to post

That would be nice for all Plugin Objects.

 

Something like a Hybrid, but not Top Plan only but in 3D.

(Front, Back, Left, Right View - Section or Elevation Mode)

 

So keeping the 3D Model simple and adding detail and intelligence into Plan and Section creation only.

(like Bricscad ?)

Share this post


Link to post

Thanks for all the input and discussion!  It is very helpful to understand how others use the program. 

 

I feel like there are two different issues we are discussing.  Both are equally important.  VW could benefit from improvements in both areas. 

1. The first deals with best practices for integrating standard construction details into a set.  This information may be independent (but related) to the BIM Model.  But most of the time this might be 2D only.  I think some sort of database management system could be implemented here.  Like the way standard walls, title blocks, plug into the VW Resources Manager.  Imagine if there was a "Detail Manager" function, where you could pull up all your standards into the set.  It would help productivity immensely!

 

2.  The second relates to best practices for translating 3D model information into 2D working drawings.  This is an area in which we (as a firm) are also constantly struggling.  Unless you have way too much time on your hands, there is no way that the model can communicate everything about your design.  Eventually you need other lines, colors, hatches or 2D information to supplement the model.  The question becomes where does this information go?  How do you get good construction details, wall sections, out of the model?

 

We tend to do design layer viewports, which we then draw over, with hatches and linework.  We keep the 3D information underlay "live" so that any changes in the model update through the details. But the process is slow and cumbersome, especially since the viewports constantly need to be updated.

 

Others place everything onto a sheet and use the annotation function as a drawing space.  I've probably argued too loudly that I think this is not very good BIM practice. Especially in a multi-user environment, it becomes really hard for others to figure out where things are located.  How do you know what is modeled and what is added at a later date? How do you export things to consultants in a way that is useful?  How do you save this information to other projects and files?  All of this (while not impossible) becomes very difficult when viewports are "drawn over". 

 

Zoomer seems to be suggesting that the 2.5D feature be added to other views besides Top Plan, which I think might really help us out here!  It might confuse everybody else not working in BIM. :=)  But I think it would allow you to draw in 3D "right over the model", if that makes any sense.  However finding all this 2.5D information scattered throughout 3 Dimensional space might be a bit cumbersome. 

 

But maybe there other options available?  Maybe the program should separate Model Layers from 2D Design Layers which might then get viewported onto sheets.  Is that too cumbersome?  Or does the BIM function need to get a lot more robust so that we don't require anything 2D?

 

Other BIM programs must have already figured this out! Thoughts? Sorry for such a lengthy post!

Edited by cberg

Share this post


Link to post

You can already kind of do your number 1. Just convert you details to symbols and Export them to a set of Library files. Depending on exactly what you are doing you would need to determine what library files your need and how many you need, but once you have them, especially if you save them in your Workgroup so they can be shared, it is relatively easy to add new details into the library. Since these would be symbols, you can access them all through the resource manager. Just have to make sure to name them in a way that makes sense to search them.

  • Like 1

Share this post


Link to post

I find myself designing a lot in 3D and spent quite a large amount of time creating walls, slabs with fills and textures that are usable in 3D and 2D sections and render good in OpenGL for presentation in axon and with cut planes. I realize though that 2D is just really much faster to setup the detail assemblies. The question is whether at this point there is a need to include both in construction docs. I think if one was able to freeze its details so that they are always the same, one could use a symbol system. It seems to me that this is the golden goose. I for myself do mostly residential work in a unique area and this should be possible. 

Share this post


Link to post

Hi all

 

So I think I'm thinking the same way as you all......

 

I've always drawn in 2D and I'm trying to convert to 3D (either completely or partially) but I feel that the section cut throughs of the 3D models aren't as detailed as my 2D drawings would have been, so will I end up having to do my sections/details in 2D anyway and in doing so double my workload? or am I seriously missing a trick here....

 

Share this post


Link to post

Sorry to hijack this thread but I have what I believe is a similar (or maybe the exact) problem.

 

I've created a number of symbols on a 1:1 scale. I want to be able to put those symbols into a sheet layer as details and match the sheet layer scale. We are attempting to have annotations pre-loaded into the symbol so we don't have to redraw those each time.

 

The end goal here is to create a template that users can open and have everything they need located directly in the template file - including symbols (although I'm not sure that's possible).

 

Could someone try and help me figure this out? I'm a bit lost. I've attached a example file with some of the many symbols I've created if that's helpful.

 

Cheers!

Example.vwx

Share this post


Link to post

@CJustinStockton,

 

People often get confused about scale because our Design Layer environment has what is called "Layer Scale". 

 

The "Layer Scale" used by VectorWorks comes primarily out of "WYSIWYG" drawing, pioneered on the Mac (and therefore part of Vectorworks' history). "Layer Scale" exists to allow graphic properties of the drawing or model to be represented properly, as though you were drawing at a particular scale on a piece of paper. It is a scaling value used to allow proper representation for PAGE-SCALED (as opposed to WORLD-SCALED) attributes:

    -Line weight;
    -Line style (e.g. length of dashes);
    -Marker (arrowhead) size;
    -Text size;
    -Hatch scaling;
    -Page symbol scaling;

In "WYSIWYG" drawing, in order to properly display these attributes, there has to be an intended output scale so you can see how the drawing will look at that intended format. The practical upshot of all this is that you should set your "layer scale" to be the same as the predominant output scale of your project. This will necessitate the least amount of attribute-scaling in viewports.

But in all design layers, at all times, you are drawing in world scale. An inch is always an inch, a foot is always a foot, no matter the "Layer Scale". "Layer Scale" serves only to set page-oriented graphics.

 

If you have symbols, they are all stored in a 'hidden' part of the file called the "Library" that contains Symbols as well as other resources (such as hatches and textures). You access all these through the window called the Resource Manager. No need to put them on a layer.

Share this post


Link to post
On 10/13/2017 at 8:57 PM, Robert Anderson said:

@CJustinStockton,

 

People often get confused about scale because our Design Layer environment has what is called "Layer Scale". 

 

 

But in all design layers, at all times, you are drawing in world scale. An inch is always an inch, a foot is always a foot, no matter the "Layer Scale". "Layer Scale" serves only to set page-oriented graphics.

A training and customisation client of mine insists in drawing his huge industrial faciilties (bakeries) in 1:1 scale and then complains that text items etc. do not print as he hopes. I wrote a Conveyor plug-in for him and had serious problems with text items, arrows and such in it. He used to be an AutoCAD user in his previous life. 

Share this post


Link to post
On 10/15/2017 at 3:12 AM, Urbanist said:

A training and customisation client of mine insists in drawing his huge industrial faciilties (bakeries) in 1:1 scale and then complains that text items etc. do not print as he hopes. I wrote a Conveyor plug-in for him and had serious problems with text items, arrows and such in it. He used to be an AutoCAD user in his previous life. 

I've since found a solution for my problem by adjusting the active layer scale, that worked well. 

 

What's funny is that I took a semester long course in Vectorworks in college (taught by a Nemetschek employee) and (from my cloudy recollection) I remember being told to always do your design layers 1:1 and do all of your scaling on the sheet layers. I'm glad I figured this out. 

 

Thanks for your help everyone! 

Share this post


Link to post

Ok, back to the discussion of modeling workflow, my question has to do with at what point do you stop modeling in 3D and you start entering information into your viewports? Do you draw the roof rafters or just the roof cavity and details on top in the VP? Do you model walls with sheathing, insulation, siding  and drywall or just framing and then add details as needed to inform the sections?

Share this post


Link to post

@Christian Fekete This is the question of the day/month/year/decade/millenia of BIM...where do you stop? In my experience, the answer completely depends on the project/client/deliverable/contractor/etc. We generally stop at fasteners/level of detail needed for rendering purposes, but again, if I'm working on a very small project and modeling fasteners can make my life easier down the road, I do it.

 

Sorry I can't offer a more definitive answer, but maybe someone else has this all figured out. I'd love to meet him/her.

  • Like 2

Share this post


Link to post

@ericjhberg  This is the way I tend to do things also.

On another topic, the annotations to inform the plans, sections etc.. and also dimensions (associated to walls in DL but not in VP) Do you add these annotations to the VP or the design layers? 

Lastly, does it make sense to create a model, reference it in single files for each sheet or like I do, keep it all in the same file?

I know it's a long discussion, but the separate model is the setup used by Nick Sonder in Sketchup. Would that save memory, overall improve efficiency ?

Share this post


Link to post

Late in here but here is my pennies worth.

My design layer is 1:100 as most of my output sheets are 1:100 so all is easy, if we do details i would scale the text in the viewport or occasionally set the DL to match the Sheet layer.

Always work up a 3D model but not to final details, these are drawn 2d as I don't need to model the complete building down to to the bolts in 3D.

Text goes on the DL for top/plan items like plans. Text goes in the VP annotation for elevations as the elevations are all generated by the Model so easier to note here.

Model is all in one File, I wil do DLVP of the survey and other files so i don't get all the classes from the survey in my file and i can keep my file size low and more responsive.

HTH

  • Like 1

Share this post


Link to post

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×