Archinect
anchor

Orthographic projection issues in multiview drawings. Help...

Nick_R

Dear archinauts,

This will probably seem trivial to most of you, but I need to sort it out.

Plans, sections and elevations are drawings of an object's orthographic projection on a horizontal picture plane. To represent an object by orthographic projection we usually orient major faces parallel to this picture plane. The corollary of this is that other curved or oblique surfaces appear foreshortened. The parallel planes are measurable and maintain proportions and true size, whereas oblique surfaces cannot be measured correctly. This is pretty straightforward when we have simple volumes or when it is clear which parts of a plane are the most prominent/important. (fig.1)

What happens though when we have an unconventionally shaped building (i.e. bent layout, multi segmented facade with different angles) ? (fig.2)

Elevations :

1. How do we determine which parts of a segmented facade should be parallel to the picture plane? (fig.3)

Sections :

2.How do we select which set of internal walls should be parallel to the picture plane in a longitudinal section? (fig.4)

Elevations/Sections :

3.Do we rotate a building plan or respect its initial orientation? For instance, if we have a southwest orientated building, do we use the "compass" axes (fig.5/A) or the central axis of the building (fig.5/B)?     

4.How do we determine the central axis/axis of symmetry of a building with a skewed plan? (fig.2 - fig 5/B)

5.How do we proceed If we end up with a certain angle that works well for an elevation but not for the corresponding section given the fact that all multiview drawings should represent the building in a single position (so that the information is transferable and complementary).  To further explain my question, let's say that we have an elevation and certain walls are parallel to the picture plane (fig.6) . When we cut through the building to create a section, some walls will appear thicker than they really are due to the fact that they are oblique to the picture plane. Also, most of the internal walls across the facade will appear foreshortened hence not measurable. This in turn leads to confusion when it comes to the thickness of walls and makes most of the section ineffective in terms of communicating information (excluding height). So how do we go about this? Do we draw a working elevation at the expense of the section and vice versa? Do we keep the elevation and create two separate section planes for two different sets of walls? (Fig.6 / Sections A-A, B-B)

Sorry for the long read. Any suggestions would be very appreciated at this point. Thank you.

 
Apr 23, 19 11:05 am
SneakyPete

Decide how it will be built. If it's delegated design, send them your 3d model and move on, checking in to ensure they're getting it done correctly. If it's up to you, then you're (IMO) going about it the wrong way. You're trying to explain how something wants to be built, not trying to capture everything in a series of perfect drawings.

Unless this is your homework, then you're on your own.

Apr 23, 19 12:16 pm  · 
 · 
thatsthat

This looks like a homework assignment to me.

If not, the best advice I ever received was draw what you want built correctly.  If you don't care, don't bother drawing it.

Apr 23, 19 12:27 pm  · 
 · 
Non Sequitur

If you're going to do awkward-shaped plans, you better have a good reason besides "it looks cool in plan" because your other drawings will suck sweaty donkey balls if you don't put much thought at to what you're drawing.  

Apr 23, 19 1:08 pm  · 
 · 

Library time- go look at how Gehry makes elevations for his less than conventional elevations. 

Apr 23, 19 1:11 pm  · 
 · 
eeayeeayo

There aren't so many different faces and angles to that plan that it's so impossible to provide all of the individual facade elevations with each one parallel to the picture frame. You've only got 17 distinct facades.  I can think of buildings with 5 and 10 times that many.  Just keep rotating and draw them all.

For overall building sections you can also do as many as it takes to capture each wing of the building parallel to the picture frame - or sometimes you can do "hinged sections" or "unfolded sections" - in which you just provide break lines indicating where you've rotated your viewpoint (so that what you're providing on paper is always the view parallel to the picture frame, even though if it were stood up and glued to the actual building you'd have to fold the drawing at the building's changes in angles.)

If there aren't a lot of differences in construction in each wing, or different elevational conditions that need to be illustrated, then you might not really need a lot of overall sections, and could just provide one or two "typical conditions" instead. 

Things like wall thicknesses should be addressed in your wall types &
wall section details - it's not usually the purpose of overall building sections to provide that level of detail - so if some walls are shown as cut obliquely in your sections it isn't necessarily going to be at the expense of any critical info.

Apr 23, 19 1:29 pm  · 
 · 
Nick_R

Drawings are just tools to get you to the real thing. True. I usually do @SneakyPete. I'm under no deadline at the moment so thought I'd throw this out there. It's been at the back of my head recently after a discussion I had with a friend. It's more an issue of correct representation. No, not homework, not a job @thatsthat. These are just indicative to better illustrate what I'm asking here. English is not my native tongue so I was trying to make myself clear(-er). Thank you for taking the time to read this and answer. 

@Non Sequitur absolutely. This was not a thread on the design process or developing a concept though so it's fine. 

Apr 23, 19 1:44 pm  · 
 · 

Block this user


Are you sure you want to block this user and hide all related comments throughout the site?

Archinect


This is your first comment on Archinect. Your comment will be visible once approved.

  • ×Search in: