Quantcast
Channel: RevitCat
Viewing all 276 articles
Browse latest View live

How to Indent a Revit Repeater on a Divided Surface

$
0
0
In a recent post by Andy Milburn over on Shades of Grey, he mentioned that it is not possible to indent the edges of a divided surface.  Well, that is absolutely true - Divided surfaces are unlike divided paths, which have an indent setting so that you can push the start or end nodes of the path in from ends of the underlying path.
Divided Path indent settings

The implication of this is that when you use a divided surface to generate a "Repeater" in Revit, the repeated adaptive components will overhang the edge of the divided surface.   In the example below, a one point adaptive component is placed on a node of the divided surface, then repeated - since the adaptive component consists of a cylindrical extrusion centred on the point, it is repeated to all nodes of the divided surface.



Since you cannot indent the edges of a divided surface, how can you prevent the repeater overhanging the divided surface?  If you made the extrusion offset from the adaptive placement point you could make it indent on one or two sides of the divided surface, but then it would overhang even more on the other sides.

However, in Revit there is usually a way to achieve what you want, so I attempted to find a way to indent the repeater, if not the nodes of a divided surface:

How to Indent a Repeater Pattern

You have to use some tricky thinking, and control where the repeater components will land - and prevent it from placing any part of the extrusion in the adaptive component on the edge nodes.
One way to do this would be to create a four point adaptive component:
  • First create a new adaptive component and place four points in a square arrangement
  • Make the points adaptive
  • Join the four points with four reference lines
Now you need to set out a centre point to host the geometry so that it remains centred whatever the arrangement of placement of points
  • Place four more points on the mid-point of each reference line (watch for the triangular snap symbol
  • Join those four points with two more reference lines
  • Place yet another point on one of the central lines
  • When you select the point, it will have an option (on the Option Bar, no less) to host it by intersection
  • Select that option, then pick the other central reference line
  • It will move the point to the central intersection.
  • NB. On reflection, you could skip the last four stages and just have one central linking reference line with a single point on its mid-point.  However, you could this intersection technique for other situations - perhaps if your extruded geometry needs to be more complex than a cylinder (see later on).
  • Select the central point and set its "Show Reference Planes" to "Always"
 

  • Set the active Work Plane to the horizontal reference plane of the central point
  • Place a reference circle centred on the point (assuming you want a cylinder)
  • Give its radius a parameter
  •  Select the circle and "Create Form"
  •  Select the top facet of the cylinder and give its Positive Offset a parameter called "Height"
  • Save and load the family into wherever your divided surface resides
  • Place one of the components onto four adjacent nodes of the divided surface in the same order that they are numbered in the family (its good practice to always use the same convention - say clockwise, unlike the diagrams shown here!!)
  • Select the adaptive component and click on the "Repeater" command (its the Array icons, but with the letter P on it)
  • If you snapped to the nodes correctly it should create a regular pattern
  • The extruded cylinders will be inset from the edge - however, they will only be inset by half the distance between nodes
  • If you want the cylinders to be inset by the full distance between nodes, you'll need to be more tricky:
  • Delete the repeater (you cannot dissolve it or alter its layout)
  • Place a new component on alternate nodes in a square pattern
  • The cylinder should then sit exactly over a node
  • Select the component and "Repeat" it
  • The pattern should spread over alternate nodes on the divided surface
  • The spacing will then most likely be too big
  • Select the divided surface and double the number of divisions (or half the spacing if it is set by distance)
  • Oops - we've now ended up with the same indent as before, so we may need to try another trick
  • The answer here is to coerce Revit into doing what you want by indenting the initial placement of the component 
  • Make sure that the first placement node you select is one node in from both edges
  • Select the component and repeat it
  • This will only work if the number of divisions is equal;  it will probably no longer work if you start playing around with number of divisions after repeating the component - so it pays to set the number of divisions correctly to start with.

If you want to be really clever, you could go back to the original four point adaptive component and give it some reporting parameters and use them to drive the cylinder size depending on the divided node spacing

How to Schedule Panel Locations on a Flat Rectangular Surface in Revit

$
0
0


This post demonstrates how it is possible to schedule repeater component cell numbers on an orthogonal pattern within the Revit Conceptual Massing or Adaptive Component environment:

To set this up, first you need a rectangular surface, which has been divided; then you need to place an adaptive component onto the node(s) of the surface; once arrayed in two directions using the repeater function you can schedule the cell (column and row) numbers on the divided surface.  Of course it isn't that simple - the adaptive component has to be aware of its location relative to a fixed point, by doing some calculations:
  • The example here is perhaps a little more complicated than it needs be to demonstrate the principle, but it does a few other fun things too.
  • The adaptive component to be repeated must be set to “Shared” so it can be scheduled;  it needs to use shared parameters so they can be scheduled too.
  • In this case it is a 5 point adaptive – one for each corner of the base of a rectangle placed on each cell of the divided surface;  the fifth point works as a “Reactor” – it tracks the distance of the component from a base point.
  • This rectangle is made more interesting by having a pyramid on top that has an apex that moves depending where it is in the repeater pattern.
To create the adaptive component:
1.  Start a new generic adaptive family.
  • Place four points in a rectangular shape;  make them adaptive
  • Join the four points with reference lines (make sure 3D snapping is on)
2.  Set up the geometry for the pyramid (optional):
  • Place a point on each line  (if correctly hosted it displays as a small point)
  • For each point assign a (Shared) parameter to its “Normalised Curve Parameter” – for the notional Y axis point, make it “Y_Ratio”;  its point on the opposite side will have its “Measure From” value set to End, rather than beginning, so that they line up.
  • Assign a parameter “X_Ratio” to the notional X axis points (one from beginning, one from end, depending on which direction you drew the reference lines)

  • Join the two opposing points with reference lines
  • Place another point on one of the linking reference lines
  • Select the point and the option bar should show:
 

  • Click on “Host by Intersection”, then select the opposing reference line;  it should move to the intersection of the reference lines
  • Set the “Show Reference Planes” property of the point to Always.
  • Set the work plane to the horizontal plane of the hosted intersection point (it only shows as a single line)
  • Place another point on top of the one on the intersection (ensure 3D snapping is on);  ignore the error message about duplicate points;
  • Select the new point and drag it up in the zed axis – its ”Offset” property should change.  If you are lucky, it should be a positive value (if it is zero, then the workplane or hosting went wrong);
  • Assign a “Height” parameter to it – if it was a negative value you’ll need to assign an interim parameter then convert it to positive with a formula, for the end user to understand.
  • This point becomes the apex of the pyramid;  
  • Join the point to the four corners of the base with four reference lines;
  • Flex the X_Ratio and Y_Ratio and Height parameters
  • Create a Form (surface) on each of the four sides of the pyramid
3.  Setting up the “Reactor” controls (Important):
  • Place a fifth adaptive point to the left of point 1 (bottom left corner of rectangle)
  • Use a reference line to join it to point 1
  • Join points 5 and 4 (Bottom right corner of rectangle) with another reference line – to create a triangle of lines.
  • Place 3 dimensions between the adaptive points 5 & 1, 1 &4, 4 & 5 – for each one make sure to set the relevant reference line as the work plane for the dimension;  it is vital to snap the dimensions to the adaptive points, rather than to line ends, surface corners etc (if not then you can’t use the dimensions later on in formulas)
  • Make each dimension as a reporting instance parameter, for use in the “cosine law” in trigonometry.  We need to calculate the angle in the triangle
  • The Cosine Law for calculating an angle when 3 sides are known, is:
    Angle γ = Arcos( (A² + B² - C²) / 2AB)
  • Revit version of Cosines formula:
    Angle γ = Acos( (A^2 + B^2 - C^2) / 2*A*B)


  • To work out the X and Y components of the distance “A” between points 5 and 1, use trigonometry:

  • This calculation assumes that the angle of the line between P1 and P4 is orthogonal – so it only works with a rectangular repeater pattern.
4.  Calculation of column and row numbers:
  • This requires knowledge of how many rows/columns there will be in the repeater, and the overall size of the repeater


  • Parameters for these need to be built in to the component as shared parameters (for scheduling), as shown below;  these can subsequently be linked to the parent family parameters for these values.
    X Number and Y Number should be integers (count of repeats in each direction)
    Column Number and Row Number should also be integers.
    X Ratio and Y Ratio should be number parameters
  • You may need a couple of checks to take care of when it has zero values for the distances.

5.  Creation of the repeater:
  • The pyramid family needs to be loaded into another family that can support a divided surface – this could be an adaptive component, a mass family or an in-place mass family in a project.  In this example it will be an in-place mass family.
  • Load the pyramid into the project
  • Start an in-place mass family
  • Draw a rectangle of reference lines,
  • Give the rectangle dimension parameters of Length and Width
  • Generate a form (surface) from the lines
  • Select the surface and Divide Surface 
  • Make the nodes visible on the surface (Surface representation)
  • Assign parameters to U Number and V Number on the surface
  • Place a point just to the left of the bottom left corner of the surface – this will become the control point for measuring distance.  It is important for getting the “Reactor” effect working.
  • Place an instance of the pyramid component by snapping the first four placement points onto four adjacent nodes in the same order as you originally created the adaptive points (say clockwise);  place the fifth point on the external point – it is vital that it does not go onto a node of the surface

  • Link 4 parameters to equivalent parent parameters:
    Y_Number to     V Number
    X_Number to     U Number
    TotalX  to  Length
    TotalY  to  Width
  • Hide the nodes on the surface (Using Surface Representation - they cannot be controlled by any view settings)
  • Select the free control point and move it very close to the bottom left corner of the surface.
  • Select the Pyramid and turn it into a repeater 
  • All being well, the pyramid will array itself over the whole surface, but each instance will look slightly different as the apex point is changed depending on its distance from the control point in the bottom left corner.  If not, it could be caused by a problem with the component itself, or else the fifth adaptive point might be hosted on the same point as adaptive point one, in which case it would move with it.
  • Select (tab) any one of the pyramids;  it should display properties including its correct column and row number

  • Finish the in-place mass family
6.  Create a generic Schedule:
  • Add the column and row numbers, and any other parameters you require
  • You will be able to edit the values for Comments and Mark, but no other instance values
  • If you edit the mass family, it allows you to manually select any of the components in the repeater.  Then you can set it to “No Component” or to any other 5 point adaptive component.
When you edit the mass family, the schedule temporarily goes blank.  Sadly this means that it is not possible to drive the geometry from the schedule.  However, it does make it easier to identify and label components in a schedule – useful for adding and editing Mark values to match column and row numbers.

This technique will not work with a curved surface because it all works by calculating the actual distance from the control point to the BL corner of the pyramid and relating that to the overall length of the surface.  It will only work on an orthogonal divided surface, unless you are a mathematical genius and can write formulas to handle more complex geometry!

Inuagural European Revit Technology Conference September 2013

$
0
0
Having recently returned from the inuagural European Revit Technology Conference I thought I'd give a quick report on how it all went.  Having been to all the RTC events this year, and all but one RTC ever, it was interesting to compare this event with others.
RTC2013eu_header_896x194

The Venue

Firstly, the venue was amazing - a mixture of old and new in the historic city of Delft in the Netherlands.  Delft looks like a miniature version of Amsterdam but without the swarms of tourists - very old Dutch gabled buildings lining canals.
  • The welcome function was in the Lambert van Meerten Museum, the setting of one of Vermeer's famous paintings.  We were welcomed by a representative of the City of Delft who demonstrated an extraordinary aerodynamic umbrella that was invented in Delft.
  • The main venue was the Waalse Kerk - a beautiful old church.  There we enjoyed the spectacle of Wesley Benn speaking from the pulpit (sorry he isn't in the pic below).
  • The breakout / exhibition space was adjacent to this in a very elegant high-tech glass enclosed courtyard - they really know how to mix old and new buildings well in Europe (I've seen a lot of this sort of stuff in Belgium and the Netherlands).

  •  The second presentation venue was the Meisjeshuis, just down the canal - well, it was a few minutes walk away beside the canal.  Once a nurses home and then a girls orphanage, it has now been converted to a conference venue.
    The walk back to the main venue was a bit scary on two counts:
      • The alarming lean of the tower on the old chuch - looking as if it would topple over onto the Waalse Kerk any minute;
      • If you were too busy looking at the tower you could be knocked into a canal by one of the multitude of bicycles hurtling along the streets.  Talking of which, how do those cars park so close to the edge without falling into the canals.
      • Mind you, if your car goes into the canal there many bikes to choose from - they even have double-decker parking at the railway station.

      The Conference

      Very sensibly the inaugural conference for Europe was kept small - this made it feel more intimate, like the earlier RTC events.  There were not so many session to choose from, thus making decisions was not as agonising - only 3 instead of 6 or 7 concurrent sessions.  Even so, I still missed one or two things I'd liked to have attended.  The two day length seemed so much shorter than other RTCs - it was over so quickly.  Hopefully it will expand to 3 days next year, now that it has proved to be such a success in Europe.

      Speakers were very international - from Europe, North America and Australasia.  Attendees were from all over Europe, but there were large contingents from Norway, Denmark (one company sent 7 people!), the UK and of course the locals (The Netherlands).

      The Day Families Became Self-Aware
      One of the highlights for me was the session presented by Matt Jezyk  - "The Day Families Became Self-Aware" - an intriguing look at what might happen to Revit families in the future.  In this session Matt started off talking about how useful it would be if families were able to react to their Revit environment - for example, a door that is hosted in a fire-rated wall might need to enable its own fire-rating and change its panel thickness and materials etc.  Currently this is a very manual process for the user having to swap out the door type  (although we do now have reporting parameters that allow us to make the door respond to wall thickness).   Matt then talked about the whole process of API and Dynamo/Python scripting where it is possible to make Revit do all kinds of things, being driven by external code.  However, scripts and API have to be initiated by the user - they cannot react to stimuli or changes in the Revit model itself, so we are still reliant on the user again.  This is all stuff that we know about or are starting to learn of.

      Then the exciting stuff started - Matt proposed a type of scripting that could automatically run itself as a response to when a change is made in Revit.  He showed us a prototype of Revit families that contained Dynamo or Python scripts - these families could be placed into Revit models as normal.  When placed the family would run the script and change the component according to where and how it had been placed, for example a door hosted into a wall could check its fire-rating and adjust itself to suit.  Not only that, but if the wall was to subsequently change, the hosted door would re-check the fire-rating and respond again.  So, no user intervention or checking is required.  Wow!  My reaction to that was the same level of excitement I had when I first saw Revit automatically change a drawing reference when a sheet was renumbered:  I want it, and I want it now!!!

      Needless to say, this was only a proof of concept, so there is no guarantee that it will ever be implemented in Revit.  Of course it takes quite some time for something like this to reach the market, supposing they do go ahead with it - lots of testing and checking needs to happen.  However, the buzz that was generated in the conference after that was quite noticable.  I was only sorry I was not able to attend Kelly Cone or Martijn de Riet's concurrent sessions - but this was unmissable stuff.

      Fractal Fun with Revit Repeaters and Adaptive Components
      Photo by from twitter


      My own presentation went well - incorporating some improvements since last time. 
      It started off showing some fun stuff with Fractal theory and example patterns in Revit:
      Fractal patterns - source Wikipedia
      Koch Snowflake - source Wikipedia
      Fractal Triangles created in Revit


      Fractal Trees created in Revit
      After the fun I demonstrated a few practical examples of how to use nested repeater patterns in Revit, including a picket fence that follows the contours of a site toposurface.
      Adaptive fence component with nested repeaters - follows contours

      Following this was a less practical but still useful example of how to nest repeater patterns inside curtain panel pattern components in order to force Revit to trim the edges of repeater patterns (which it normally will not do):

       

      Then came the demo of creating a Revit model of the roof of Santiago Calatrava's Gare do Oriente in Lisbon.  This consited of multiple repeater patterns nested three levels deep, which allowed the creation of very a flexible parametric model with minimal use of maths
      Repeater one - one-eigth segment of column with variable number of struts
      Repeaters two and three - segment rotated/mirrored around its centre

      Repeater four - column assemly in two way array over station platforms
      Roof assembly at Revit-Dusk
      All of this was done as a live demo in Revit, albeit with some partially prepared sample files in case it didn't work under the glare of an audience.
      Photo by Martin Romby - from Twitter
       Roll on RTC Europe in Dublin in August 2014.

      Circular repeater geometry in Revit

      $
      0
      0
      In previous posts I have described how easy it is in Revit to use a circle as a rig to host regular geometric shapes like squares and hexagons.  It occurred to me the other day that circles would also be ideal hosts for parametric stars;  put that together with Repeaters and you get some interesting possibilities . . .

      Here is how to create a parametric star:
      • Create a new adaptive component;  
      • Place a point at the origin; make its reference planes visible
      • Set the horizontal ref plane as the work plane
      • Place a reference line circle onto the point
      • Make its radius dimension into a parameter
      • Select the circle
      • Divide Path
      •  It will create a divided path on top of the circle with 6 nodes
      • Create another Adaptive component
      • Place two points, and make them adaptive
      • Place a line between the points, remembering to enable "3D Snapping" first
      • Save the second family (eg. as "AC 2pt Line")
      • Load it into the first family
      • Place one of the components onto two non-adjacent nodes
       
      • Select the two-point component and click on the Repeat command
      • You should end up with a triangle
      • NB.  if you placed the two-point adaptive line component onto adjacent nodes you'd get a hexagon
      • Actually, a triangle is only half of what we want, but we need to backtrack to get the correct result - so, UNDO the repeat command
      • Place a second two-point adaptive on alternating non-adjacent nodes
      •  Select both lines and click on Repeat
      •   
      •  You should end up with a six pointed star, which is now a "Repeater"
      • Select the divided path, and change the number of nodes to 5, giving you a five pointed star
      •  Change it to 4, which gives a cross (not so useful)
      •  Change it to 3 for a triangle
      •  Change it back up to 7
      •  And then to 8 - this gives an eight pointed star that could be useful in tiling patterns when repeated itself on a divided surface
      •  As the number of nodes increase, the shape becomes closer to a circle
      • Now all you need to do is turn the number of nodes into a parameter to get yourself a one-point adaptive parametric star component
      • Oh, one more thing - you need to make the nodes of the divided path not visible otherwise they will show up when this component is placed in a project or another family





      How useful is this going to be?  We'll investigate that in a future post . . .
      Here is one possible use - geometric patterns


      Geometric Patterns from Parametric Revit Stars

      $
      0
      0
      In my previous post I showed how to make a parametric star in Revit - one that has a variable number of points.  Here are some examples of how that might be used in a parametric pattern.  I have shown something a little like this previously and also in my RTC 2012 presentation, but that was done using a fixed number of points in the star, although it changed shaped parametrically.

      Repeater Star Patterns in Revit

      Since the parametric stars are made as adaptive components, they can be placed on a node of a divided surface (in a conceptual mass or another adaptive component)
      Once the star has been placed it can be arrayed using the "Repeater" command to form a geometric pattern
       
      By changing the radius of the circular rig in the adaptive star, it changes the size of the star and hence the pattern
        Once the stars overlap it gives a completely different effect

       
      Going back to the original size, the number of points on the star can be changed.  however, a five pointed star does not result in an interesting pattern once they overlap each other.  Likewise, 4 or 3 pointed stars don't look good (crosses and triangles)
      Increasing the number of points to seven, does work reasonably well, although it looks a bit messy as the overlap increases!
        


       An eight point star works well on a square grid pattern:




      You can also change the background grid on the divided surface.  In this example one of the grids is rotated by 30 degrees:

      The resulting patterns can be triangular or hexagonal in nature:
       Just by changing the size of the star (radius of underlying crcle) you can get radically different patterns.








      I hope that someone finds a use for this flexible way to create parametric patterns in Revit (2013 or later).

      Rival Revit Environments - Traditional vs CME

      $
      0
      0
      For the first decade of its life, Revit really only had one method of creating 3 dimensional forms - in the Family Editor (aside from those generated by system families).  When Revit 2010 was released we suddenly had a whole new way of doing things with the new conceptual massing tools.  Since then, many people have been grappling with the seeming divergence of these two very different Revit environments.  Now we have the prospect of such tools as Dynamo introducing yet more ways to work in Revit - however exciting it seems, Dynamo has the potential to cause even more disparity and fragmentation of not only the workflow but the development effort by Autodesk.  This has been worrying me for some time, and I've been thinking about how best to convey our concerns to Autodesk in a positive constructive way.

      Recently I was reading Paul Aubin's new book:
      Renaissance Revit:  Creating Classical Architecture with Modern Software
      In chapter 1 he discusses some of the differences between the Traditional Family Editor and the Massing Environment (Conceptual Massing Environment or CME as Autodesk refers to it).  I realised that there are a whole lot more differences, that are not clearly documented anywhere, so I thought I would try to address the issue here.

      Before I do that, a little more about Paul's book: 
      Greek Doric
      Although the overall intent of the book appearsto be about using Revit to create classical architectural elements, it is in fact all about using Revit to create some tricky geometry - and the techniques that he describes could be used for all sorts of purposes.  For example, in chapter 4 he describes how to constrain arcs, circles, elipses and splines that might be used to parametrically control families.  I wish I had read that before embarking on my own exploration of the subject, which had nothing to do with classical architecture - it would have saved me a lot of time and effort.   The early chapters of the book also set out a lot of good practise that anyone should apply when creating Revit families, for whatever purpose.

      Skipping on to Chapter 11 gives us a good general introduction to creating families in the Conceptual Massing Environment before launching into ways to apply that specifically to classical architecture.  Again, these principles are relevant to many other situations when building Revit models.  Thanks Paul for a fascinating new book.

      Traditional Family Editor vs Conceptual Massing Environment


      Below is an analysis of the differences between the traditional Revit Family Editor and the Conceptual Massing Environment, follwed by some specific differences that apply to Adaptive Components - these generally follow the massing methodolgy but have a few more quirks of their own.  The table below is compiled from some of Paul's observation and some from my own experiences of working with adaptive components in particular.  This is not a definitive list, and I may have made errors or omissions.  Please feel free to comment, and I may update the list in the future, based on feedback.

      I hope that Autodesk can make use of this analysis to help bring the two rival Revit environments back together in the future– and I use the word rival quite deliberately because they seem to be fighting each other much of the time.

      I have traffic-light colour coded the comparisons:



      Red = Bad
      Orange = Getting there
      Green = Good

      Conceptual Massing Environment
      Traditional Revit
      3D work environment
      ·      Offers greater flexibility
      ·      Can be very confusing
      ·      Need to continually set work plane;
      ·      Can pick work plane on the fly just by picking level or reference plane
      ·      But it is easy to make mistake with work plane
      ·      Levels visible in 3D (but not for in-place massing)
      ·      Reference planes visible in 3D (but not for in-place massing)
      ·      Reference lines used extensively to generate forms or host elements
      2D work environment
      ·      Limitations  when working with unusual shapes in 3D
      ·      Simple, familiar workflow (for Revit users)
      ·      Work plane often set automatically by the view; 
      ·      Can set reference plane as work plane only if named first
      ·      Must use set work plane command (unless prompted)
      ·      Hard to make a mistake with work plane
      ·      Levels not visible in 3D
      ·      Reference planes not visible in 3D
      ·      Reference Lines used almost exclusively for controlling angles
      Levels
      ·      Can create multiple levels within family (but not In-Place CME) 
      ·      Created levels do not relate to levels in project
      Levels 
      ·      Cannot create multiple levels within family(unless already in template) 

      ·       Upper level in columns & 2 level Generic model do relate to project levels
      3D forms created from selected elements
      ·      Greater flexibility
      ·      Harder to understand 
      ·      Unpredictable – eg. Cannot nominate which element is for path (result is a guessing game) 
      ·      Easy to make mistakes
      ·      Often fails to create form
      3D forms are sketch-based, on a 2d work plane
      ·      Limited range of 3D forms
      ·      Simple to understand 
      ·      Predictable – eg. Sweep path is nominated   
      ·      Hard to go wrong once learnt
      ·      Seldom fails to generate form (usually understandable when it does – eg lines zero length)
      3D forms “unlimited” 
      ·      Limited by what you can draw, and by certain software limitations (unknown so can’t be listed here!)
      ·      Can be solid or void
      ·      Boolean combinations –  does allow solid to cut solid
      ·      Booleans can be confusing (must control picking of form vs surface)
      ·      Can loft multiple selected shapes, which can be at varied angles
      Five basic 3D forms
      ·      Limited to: Extrusion, Blend, Revolve, Sweep, Swept Blend
      ·      Can be solid or void
      ·      Also limited boolean combinations of above (join & cut)
      ·      Can only create lofting between 2 shapes - with “blend” (ends must be parallel) or “swept blend” (ends perpendicular to ends of sweep path).
      Profile families
      ·      Cannot use traditional 2d profile families to generate forms in CME – but can use flat generic adaptive families (easy enough but different methodology)
      Profile families
      ·      Can use traditional 2d profile families to generate forms
      Editing Forms
      ·      Forms can be changed from one type to another (extrusion to blend to loft etc)
      ·      Lock/Unlock Profile command to enforce extrusion
      ·       Add Profile command available
      ·      Boolean commands to cut or combine forms
      Editing Forms
      ·      Forms cannot be changed from one type to another – each of the five primitives are not interchangeable
      ·      Profiles, sweep paths, and dimensions can be changed easily
      ·      Boolean commands to cut or combine forms
      Editing shapes of profiles
      ·      Complex rules;  easier to make mistake
      ·      Depends on how the lines/form were created:
      o  Base  reference lines can be moved without editing profile
      o  From lines and profile is “locked” – can edit
      ·      Edit Profile commandsometimes available (like trad. edit sketch)
      ·      Add/remove segments in Edit Profile mode
      ·      Add Edge commandsometimes available
      Editing sketch profile
      ·      Very simple;  almost foolproof
      ·      Select form, “Edit Sketch”
      ·      Adding/removing segments very simple in sketch mode
      Changes to Work Plane of profiles in forms   
      ·      Somewhat complex and not intuitive:  need to select original elements that were used to create form – only works if reference lines;  model lines get consumed by form, so cannot select them.
      ·      Can only change work plane of selected elements to a parallel plane
      ·      Extrusion with “locked profile” will display Positive and Negative Offsets if top or bottom face is selected;  (but work plane of original lines is not displayed at the same time) – so it is possible to change location of a top face relative to bottom face (not possible with blend or loft)
      ·      User can place a dimension for the height of the form, but this involves extra steps; it stops temporary dimensions working for that face;
      ·      If the height dimension is turned into a parameter it may remove spot elevations and shape handles of forms back in the project environment
      Changes to sketch profile Work Plane   
      ·      Very simple – “Edit Work Plane” or “Pick New” commands available to relocate original sketch and thus modify or move form
      ·      Using “Pick New” work plane, can totally change orientation of an extrusion or blend.
      ·      Extrusion or blend have properties:  Workplane; First End; Second End so it is easy to change the opposite end of the form by amending End property.
      Identify or Change absolute location of a face in CME (eg. If you need to know RL of top of an extrusion)
       
      ·      This is a nightmare!  It is really difficult:
      ·      Extrusions display a positive & negative offset for a face only if profile is locked – therefore it won’t work for a blend
      ·      Work plane property is only displayed in a separate dialog box; and only if the original reference planes are selected; not possible with model lines.
      ·      It is not possible to place a Spot Elevation annotation in the CME
      ·      If you place a Spot Elevation annotation onto an in-place mass before editing, the annotation disappears as soon as you edit – so you cannot see the RL as you drag the face of an extrusion!
      ·      In the project environment, if you drag the face of an extrusion (mass family) using the shape handle, an attached Spot Elevation will not update dynamically– only updates after you let go the handle
      Identify or Change absolute location of a face in Model In-Place family (eg. Need to know RL of top of an extrusion)
      ·      Very simple: Edit Family; select form to see properties
      ·      Extrusion or blend have properties visible in same dialog box:  Work Plane; First End; Second End (dimensions) so it is easy to know its absolute location.
      ·      Easy to change Second End property of extrusion or blend to an absolute location (assuming base work plane is a level, grid or named reference plane)
      ·      It is not possible to place a Spot Elevation annotation within In-Place family
      ·      If you place a Spot Elevation annotation onto an in-place family before editing, the annotation disappears as soon as you edit– but this is not so bad because it easy to identify RL from work plane + second end property
      ·      In the project environment, if you drag the face of an extrusion (in-place family) using the shape handle, an attached Spot Elevation will update dynamically
      Panelling
      ·      Divided surfaces can only be done within CME, not in project environment
      ·      Divided surfaces are fairly flexible (grids or intersects),
      ·      Divided surfaces are not always predictable or easy to control
      ·      Curtain Panel Pattern families are special types of components that can only be applied or modified as patterns within the CME.  This is very “unRevit-like”.
      ·      Only Curtain Panel Pattern or Adaptive families can be applied to a divided surface as a pattern of panels or repeater – but this must be within CME
      ·      Adaptive and Curtain Panel Pattern components can be individually placed on divided surface nodes (only if visible) in CME – they will move with changes to nodes
      Panelling
      ·      Curtain walls and systems can be placed in the project environment; 
      ·      Curtain wall/system divisions are limited
      ·      Curtain wall/system divisions are fairly predictable
      ·      Panels in curtain walls and systems are regular components that follow normal Revit rules
      ·      It is not possible to apply a traditional Revit family to a divided surface as a pattern of panels (or suchlike) – it may be technically possible by nesting, but that is not a realistic option
      ·      Adaptive components can be placed on divided surface nodes (only if visible) in Project Environment – but they will not move with changes to nodes.  NB. Node visibility can only be changed by editing the mass family.
      Categories - CME
      ·      Mass family cannot be changed to other categories
      ·      Adaptive families can be changed to some limited other categories (but not to structural)
      ·      Adaptive families cannot be changed to mass category
      Categories - Regular family
      ·      Regular family can be changed to many other categories but not to mass, adaptiveor system  (except by obscure workarounds)
      ·      Changing category will inherit some characteristics of new category but depends on what is normally in pre-set templates
      File size & performance - CME
      ·      much larger files
      ·      multiple nesting can increase size significantly
      ·      can run very slowly, especially with multiple nesting
      File size & performance
      ·      smaller  files
      ·      multiple nesting only increases size a small amount
      ·      run moderate to fast depending on formulas and nesting
      Line selection in CME
      ·      Chain of lines is automatically highlighted for selection (this is mighty confusing and wastes much more time than it could ever save)
      Line selection in traditional family editor / project environment
      ·      Chain of lines is only highlighted for selection using the tab key
      Copy and Paste
      ·      Cannot copy and paste into CME from traditional family
      ·      Cannot copy and paste from project environment into CME
      Copy and Paste
      ·      Cannot copy and paste into traditional family from CME
      ·      Can copy and paste lines from project environment into traditional family editor (limited)
      Nesting
      ·      Cannot nest CME or Adaptive families into traditional families
      Nesting
      ·      Can nest traditional families into CME or Adaptive
      Group command not available
      Group command is available
      Create Similar command not available
      Create Similar command is available for in-place families
      Does not have “Annotation” tab in the ribbon
      ·      Cannot place text
      ·      Cannot place symbols
      ·      Cannot place Symbolic Lines
      ·      Cannot place Detail Components
      ·      Cannot place Detail Groups
      ·      Cannot place Masking Region

      ·     Does not have model text.
      ·     only way to get any kind of text/label/tag into a mass or adaptive component is to nest a generic component that has model text in it
      Does have “Annotation” tab in the ribbon
      ·      Can place most detail/annotations to represent 2D information in plans sections and elevations
      ·      Cannot place filled region in model families (have to be nested)

      Divide line & Divide Surface available
      ·      Mismatch between numbers on divided path vs surface
      ·      Divided path/surface node visibility setting is obscure
      ·      Divided path/surface nodes visible in project
      Divide line & Divide Surface not available
      Arrays – in CME / Adaptive
      ·      Repeater command available
      ·      Repeaters relatively flexible but hard to predict
      ·      Repeaters can be 2 or 3 dimensional (inc radial)
      ·      Array command not available
      Arrays
      ·      Repeater command not available
      ·      Array command available
      ·      Arrays very limited but predictable
      ·      Arrays only 2 dimensional or radial
      Volume properties
      ·      “Gross Volume” mass property cannot be scheduled or tagged
      Volume properties
      ·      “Volume” property of forms in generic category (only) can be scheduledbut not tagged
      Control command not available
      Control command available (for flipping components in project)
      Visibility Settings not available
      Visibility Settings available (for controlling display levels and 2d view visibility)
      Parametrics & Formulas
      ·      Allows all available Revit formulas (with many limitations)
      ·      Reporting parameters can be used in formulas only if derived from Adaptive points (? No other “host” elements)
      Parametrics & Formulas
      ·      Allows all available Revit formulas (with many limitations)
      ·      Reporting parameters can be used in formulas only if derived from “host elements” – ie. Baked into templates (eg. reference planes, levels, walls)


      A few more comparisons related specifically to Adaptive Components:

      Adaptive Component Environment
      Traditional Revit Family Environment
      Can snap in 3d - this is a huge advantage, allowing easy creation of geometry in all kinds of directions in 3d.
      ·      3D snapping is off by default, leading to frequent errors
      Cannot snap in 3d in family editor (big limitation in flexibility)
      ·      Limited line selection for 3d sweep paths
      Placement of adaptive components in the project model:
      ·      Does not lock to orthogonal (bad news)
      ·      Does snap to vertices in 3d; 
      ·      adaptive points remain locked to vertices even when they move
      ·      Does not snap to toposurface even when category is “Site”
      Placement of regular components in the project model:
      ·      Does lock to orthogonal, and take advantage of all the Revit automatic snapping options
      ·      Only snap to vertices in 3d on a preselected work plane/surface; limited 3D line selection for beam placement
      ·      Does snap to toposurface when category is “Site or Entourage”
      Voids in adaptive components:
      ·      Adaptive components cannot have a host element baked into the template – so It is not possible to include a void element in the adaptive component that will automatically cut the host face it is placed onto
      ·      “Cut with voids when loaded” capability is very limited in categories it can be applied to – it cannot cut curtain panels, curtain systems,  mass categories (big limitation). Only works in projects.
      ·      void repeaters have very limited usage– they can only be placed on flat surfaces of walls, floors, furniture, casework etc.  This is a huge wasted opportunity as it means that we cannot place void repeater patterns onto massing (to quickly represent windows on a mass model), curtain panels (perforated metal panels), or any curved surface of any category
      Voids in regular components:
      ·      Some family templates have a host element baked into the template (face-based generic, doors, windows etc)– so It is possible to include a void element in the component that will automatically cut the host face it is placed onto
      ·      Regular components cannot be used in repeaters, so the “face-based” families with voids in them cannot be used

      ·      “Cut with voids when loaded”  only works in project, not in nested families

      Point Elements
      ·      Points are available in CME and Adaptive components
      ·      Points can host  geometry and components
      ·      Points can be hosted on lines/arcs
      ·      Points can be used to control arc midpoints using “Start-End-Radius” arc (from v2014 only)
      ·      Placement of (Adaptive) points on Level work plane do not snap to reference planes in 3d views (but do in plan)
      ·      Points not available
      Nesting & Hosting
      ·      Traditional families can be nested into adaptive components
      ·      Nested family behaviour is unpredictable – particularly location and 2d rotation– behaviour changes radically depending on whether hosted on points or just placed on a work plane
      ·      Line based traditional families can be nested in and two ends can be hosted on adaptive points
      Nesting & Hosting
      ·      Adaptive components cannot be nested into traditional families
      ·      Nested traditional family behaviour is fairly predictable – particularly location and 2d rotation
      3D Orientation
      ·      Adaptive families can be very unpredictable when placed in the project environment, especially when families are nested
      ·      Elements can be hosted on points, which have their own orientation controls – so that different parts of the model can be differently oriented
      ·      Options for point orientation are extremely confusing
      ·      Nested families can be oriented and located as part of the whole family or else controlled by points they are hosted on – can be very confusing
      ·      Nested components can orient themselves differently in the project environment to how they are oriented in their host family – can be very hard to manage (somewhat better from v2014 onwards)
      3D Orientation
      ·      Most traditional families will be inserted into the project in the same orientation that they are modelled – the exception being face-based, or work plane based (where “Always vertical” is unticked).
      ·      The whole family is inserted in the same orientation
      ·      Nested families are oriented as part of the whole family, even if they have “Always Vertical” ticked (overridden by parent orientation in the project). Simple and predictable

      Winder Stairs in Revit 2013

      $
      0
      0

      Revit Stair By Component


      Revit 2013 introduced the concept of stairs by component.  Two of the new stair run component types were for creating "Winder Stairs".  

      Winder Stairs

       
      When I first tested these new tools I really struggled to get them to work at all, let alone well. I suspect that many people have since tried and failed to use them successfully. Here are a few tips that might help:
      • It is recommended that you only attempt to work with winder stair components if you have plenty of time and patience!
      • It may be more efficient to start a stair as a winder component but convert it to sketch as soon as it starts wasting your time.
      Depending on which country you are working in, two different winder styles may work for you:
      • in Continental Europe it is common to use "balanced winders";
      • In Australia they are not suitable, so you need to use "single-point winders".
      In Revit it is relatively easy to create a stair with 5 winders; but it is extremely difficult to adjust it down to only 3 to create a legal winder stair for use in Australia ( the maximum in the BCA) - it takes a lot of fiddling around with dimensions.

      L-Shaped Winders


      This requires just one mouse-click to place the start of the run, in the orientation shown in the tool icon.   
      • You can mirror the run before placing it, by ticking the “Mirror Preview” box on the options toolbar ;   
      • You can also rotate the stair by pressing the spacebar before placing it.  
      • It places the run with the number of risers you have set, depending on the full remaining stair height required; alternatively you can override the number of risers in the overall stair instance properties before you place it, then set the number back to the full number afterwards.   
      • There are two styles of winder stair available – “Balanced” or “Single-Point”
      Balanced Winder Stair
      Single-Point Winder Stair
      • Balanced winder stairs generally have angled risers spread over most of the stair – the change in angle of each riser is incremental.  These are commonly used only in continental Europe.
      • Single Point winders generally have parallel risers for most of the stair, while the angled risers are concentrated around the change in direction, with only a few winders.  These are more commonly used in the rest of the world, outside Europe.  This implementation of single-point winders has many limitations, and is very difficult to work with – it is impossible to create a stair with only 2 winder treads;  it is difficult to achieve only 3 or 4 winders.  The main reason for this is that the number of angled risers is controlled by setting the number of parallel treads.
       

      • This system works ok for balanced stairs, but for single-point stairs you actually want to define how many winders there are, but you are forced to do the reverse and calculate how many parallel ones you need in order to achieve it.

      To place a single-point winder:

      • First set the desired height and run width properties;
      • Place a winder run, accepting the default layout;
      • Change the overall run lengths using the arrow shape handles;
      • Change its properties to a single-point winder;
      • Increase the number of parallel treads for each leg, but do it incrementally so that you can tell at which number it fails;
      • Fine tune the overall run lengths – place reference planes so that the shape handle arrows can snap to them (the only way to control it - important step).
       


      • It is almost impossible to control the location of the setout point for the winders – it seems that it has to be outside the run, not on the edge of the run as is normal in Australia.  The closest it will go is 26.35mm from the inside corner (in each direction), or 0.8mm in one direction, and 51.7mm the other. 
      • NB. when placing a winder stair, if you have just ticked the Mirror Preview  box, you may need to get the focus back on the main canvas before pressing the spacebar (use the middle mouse button to do it).

       U-Shaped Winder


      These work in a similar fashion to L-shape winders.   If you use a balanced winder, Revit will balance the entire stair layout around the three legs of the run - this may work for you?


      Single Point U-shaped winder stairs do not have any way to control the number of parallel treads in the middle section of the stair – thus rendering it completely useless unless you only require one parallel tread!
       

      Modifying L-Shaped Winder Stairs



      Single-point winder stairs are very difficult to work with and modify in Revit - so you may need help!  Balanced Winders are not discussed here as modification is somewhat simpler.


      Once you have created a Single-Point winder stair, most likely you will have set the number of parallel treads on each leg to as high as Revit will let you (to keep the number of winder treads down to just 3).
       

      Almost any subsequent change that you try to make to the winder run will result in an error message saying that it cannot be done. 
      For example, if you try to adjust the number of treads on each leg of the winder by dragging an arrow shape handle at the top of the run, you will get an error message:
       


      Most likely this is because it cannot deal with the adjusted number of parallel treads – it is trying to add a riser to the top, and remove one from the bottom of the run;  but the number of parallel treads at the start is set in the properties so it just fits into the run before the winders start.  When you try to change the run, it can no longer fit that number of parallel treads before the winders start. 
      Solution:
      The solution to this is to reset the number of parallel treads back to 1 or 2; 
       
      Then make the change to the run (by dragging the top shape-handlle arrow for example - this will shorten the lower part of the run, and reduce the number of parallel treads again)
       
      After that you need to reset the number of parallel treads to what will fit – it may not be the same as before, so you may need to recalculate how many are required.
       

      This technique will probably be required for almost any change that you need to make to a single-point winder stair, including changes to stair width, height or dragging any shape-handles.

      Shape-handles
      The arrow shape-handles at the top and bottom of a winder stair behave like other stairs.  However,  it seems different because a winder stair does not usually have landings.  On a stair with a landing, dragging the arrow handles will move the stair ends by increments of one tread;  on a stair with no landings the shape handles allow free movement (no increment snapping) - hence the need for reference planes to snap to.

      The square shape-handle in the middle of a winder will adjust the location of the selected leg of the winder - it moves the leg sideways, and as a result will adjust the length and end point of that leg while the other leg does not move (its end point stays fixed).  If you have a single-point winder, you will almost certainly get an error message if you try to drag it (refer to solution above).

      Temporary Dimensions
      Winder run lengths cannot be changed using the temporary dimensions - therefore it is wise to create reference planes so that you can snap to them when dragging shape handles.





      Spiral and Curved Revit Stairs

      $
      0
      0

       

      Creating Curved and Spiral Stairs

      New Stair Tools were introduced in Revit v2013 – these included two methods to create a curved stair run by component:

      Full-Step Spiral


      This tool requires two mouse-clicks to place a curved or spiral stair – the first click is for the centre, the second for the first riser location & stair radius (inside, centre or outside depending on your Location Line setting). It does not require a third click for direction – this will normally be counter-clockwise, unless you press the space-bar first. It will place a run with the full number of risers required to meet the stair base and top level properties that you have set.

      It will allow the top of a spiral to go over the base (ie. More than 360 degrees). The pop-up help note says “Use to create a spiral run that is greater than 360 degrees” – this is a little misleading as it can also be less than 360.



      Some issues to watch out for when placing spiral stairs:
      • Spiral stairs creation does not snap to orthogonal, so make sure you place a reference plane first so it can snap to that - to get centre and first riser aligned orthogonally.
      • Press the space bar after the first mouse-click to flip the orientation (it is too late to flip it after the second mouse-click).
      • If you press the “Flip” command after placement, it will only reverse the direction (swap first/last riser), but will not mirror its location.
      • Make sure that you get the radius correct first time, because it is almost impossible to change it accurately afterwards in v2013 (its better in v2014).

      Center-Ends Spiral

       

      This tool requires three mouse-clicks to place a curved or spiral stair – the first click is for the centre; the second for the first riser location & stair radius (inside, centre or outside depending on your Location Line setting); the third click controls the direction and the end of the run.




      It will not allow the top of a spiral to go over the base (ie. It must be less than 360 degrees) – if you try to go past the start, it just reverses the direction of the stair and reduces the number of risers in the run. However, it can be extended past 360 later on. Alternatively you can add a second Center-End spiral and Revit will add a landing between them - by default the options bar setting will be ticked so that this happens automatically.
       



      Modifying Curved Stair Components   

      Controls for changing arc and spiral stairs are somewhat limited:
      The arc /spiral radius cannot be changed accurately after placing it (see below) - using the circular shape handle in the centre of the run will allow you to drag the radius value but it will not snap to anything, so it is impossible to get a precise value. 

      Flip    

      The Flip command reverses the direction of the stair.  This is available from the ribbon during edit mode or else when the whole stair is selected not in edit mode - displays as a small arrow when the stair is selected.

      Align Tool

      You must be in Edit mode to use this command on individual stair components (otherwise the whole stair will move).    It will behave differently depending on what you select to align and what the relationship is between linked components.

      • If you try to align a curved run edge or centreline, it will not select the curved edge at all - ie. it can't be done.
      • If you align the end point of a riser line to a curved line, it will move the run , but in the process will change the radius of the stair and none of the other riser ends will align - so don't try this!
      • if you try to align the ends of a curved run, it should work providing the reference line is radial from the centre of the arc; but the landing length (between runs) will be changed
      • It will not align the sides of curved landings.
      • if you align a landing end (where it meets a run), it will most likely move the whole landing, which in turn will move the associated runs, in effect moving the whole stair. Sometimes with a landing it may It gives a cryptic error message.


      Shape Handles

      There are four different types of shape handles that you get on curved stairs, and they each behave differently, and their methodology also depends on whether other components are joined on:


      Circle handle in middle of a curved stair run.
      • Dragging this circle will adjust the radius of the arc defining the centreline of the stair run.  It does not change position with the location line – it is always the centreline of the stair. 
      • This will not snap to any increments or elements, giving random dimensions, which renders it effectively useless – hence you can never accurately change a curved stair radius after placing it in v2013. 
      • This is the only way to change the radius of a curved stair in v2013
      • Refer to Temporary Dimensions for an alternative method in v2014 .
      • If you have a landing on a curved stair, attempting to use the centre-circle shape handle will also move the other run (not selected one) thus distorting the landing.




      • Side-Shape handle arrows behave like the centre circle – they won’t snap to elements or increments so they are next to useless.
      The dot and arrow handles at the base and top of stairs behave like straight stair runs: 
      • Dot shape-handle will add/remove risers from that run.  Warning: height of that end of the run is adjusted so the base could be up in the air;
      • Arrows will add/remove risers from that run but remove/add the same number to the other end of the stair.  Height of ends will not change but location in plan will change.


      Temporary Dimensions

      Sometimes the temporary dimensions that you get when selecting a stair component will not work, when you expect they would:
      • Spiral stair run width cannot be changed by its temporary dimension – you must use the properties.
      • Spiral stair radius cannot be changed by temporary dimension in v2013 but can be in v2014.
      • If a curved stair has a landing on it, it is not possible to change the radius of both runs together, so in v2013 it will not be possible to adjust at all – just delete and start again (and lose all the annotation etc)!   
      • In v2014 you can use the following procedure:
       
      • If the stair has a landing, you need to adjust run by run, which does some interesting things to the landings in the process
      • It also seems to change the run width, but do not be alarmed - just reselct the run to see it has the correct width
      •  Once the radius is adjusted on the second run, it corrects the landing shape
      •  However, the landing length is now wrong
      • This needs to be adjusted by rotating one of the runs or else by dragging the shape handle arrows on the landing.  Most likely you nee some construction lines first, in order to get the landing length right, in which case you could use the align tool to get the runs in the right place (do not select the landing).
       


      Move Command         

      This is not particularly useful for changing curved stairs:
      • Moving a landing will move the whole stair as one;
      • Moving a run will distort the landing.

      Rotate Command

      This is useful for adjusting curved/spiral stairs:
      • Rotating a curved run about the arc centre-point will reduce or expand any attached landings.   Most likely you will need to set up some construction lines to snap to so that the landing length is correct.
      • Rotating a landing about the arc centre-point will bring both runs with it, effectively rotating the whole stair.  
       

      Revit Stair Landings - Part 1: Creation

      $
      0
      0

      Creating Component Based Stair Landings in Revit




      As part of the new (in v2013) component based stairs,  landings between runs are one of the component types.  They are pretty straightforward to create, but there are a couple of little tricks described below.  Modifying them is more complicated, and that will be covered in another post.

      Landings – can be created by several methods:

      1.  Automatic Landing


      Draw two stair runs with a gap between them and let Revit place a landing between them for you. By default the options bar setting will be ticked so that this happens automatically.





      On a dog-leg stair, Revit will automatically make the landing width the same as the run width

      2.  Pick Two Runs



      Pick two runs that stop and start at the same height, and Revit will do its best to join them with an automatic landing. 

      3.  Create sketch

      Used for creating non-standard landings where Revit does not or cannot create the landing you require. 

      4.  Convert component

      This tool can be used to convert an automatically generated landing component into a sketched component, so that it can be modified to fit more complicated geometry - appropriate to use when Revit creates something close to the desired result;  thus saving you time in creating a sketch from scratch. 
      Once converted, it will no longer behave like an automatic landing - it will not adjust itself automatically when adjoining components are modified;  it will not have any shape handles.  Be warned, once you convert to a sketch you cannot convert back to Automatic - so you have to modify manually or else delete and replace with an automatic landing, and lose any hosted elements or annotaion in the process.
      Once you have converted a landing, it sits there like a lump  of concrete - it won't allow you to just start modifying it immediately, but requires another step (pardon the pun):  first you have to select the landing and click on "Edit Sketch"

      5.  Landing at the top of a stair

      It is not possible to create an automatic landing at the top of a flight of stairs – as that requires two runs to join to. Two workaround alternatives are:
      • Create a sketched landing; or 
      • Place an extra run at the top so that Revit creates an automatic landing; then delete that run – the landing will remain, but will be converted to a sketch based component

       
       

      6.  Landings on complex stairs

      Revit will normally only create landings between two runs, but you can sometimes force it to work with more  - For example, if you have a stair that splits into two parts at the landing:
      • First you need to draw the lower run and one of the upper runs
       
      • Then mirror the upper run
      •  Delete the landing
      • Place a new landing by picking the two upper runs.  If the landing edge happens to be coincident with the lower run then you may be lucky enough for Revit to join them automatically.

      • If the lower run is not coincident, then you would need to convert the landing to a sketch and modify it.
      • If the landing and runs were all joined, but you move one away, Revit will warn you that it needs to unjoin the stairs;  but what it does not warn you is that it automatically converts it to a sketch based landing.
       

      7.  Landings on multi-level stairs

      If you create a multi-level stair, Revit will not put landings on the intermediate levels.  This may or may not be your desired result.
       
        If you do want intermediate landings, there are a few tricks you can use on Revit.
      Firstly you could try adding the intermediate landing yourself by the previously described method for creating a top landing (add another run, then delete it).
      Once you set it back to multi-storey the landings show up, although they do not join properly to the upper storey of the stair
      There is another big problem with the landings:  The railings will fully enclose the mid-landing and top landing.
      Most likely this is not what you want!   I will cover how to deal with this in a subsequent post on multi-storey stairs.

      Next post will be on modifying landing components.

      Revit Stair Landings - Part 2: Modification

      $
      0
      0

      In the previous post I talked about creating stair landings in Revit.  Now its time to look at modifying landings, which is very different to the old sketch method (pre-v2013).

      Stair landing components can be modified individually but they interact with run components in unexpected ways, so you need to learn the rules as it is not always intuitive. If stair landings have been created by the new sketch method or have been “converted”, then the interaction with runs is more limited.

      To modify a landing component you generally need to be in edit stair mode. The exception to this is changing landing properties, which can be accessed by tab-selecting the landing without editing the stair.  There are various methods to modify a landing, each with their own benefits.  There are also some subtle differences between landings in v2013 and v2014 (and later), to be described further on:

       Align Tool

      The Align tool will behave differently depending on what you select to align and what the relationship is between linked components.

      • if you align a landing edge, it will most likely move the landing, which in turn will move the associated runs, in effect moving the whole stair
       
      • If you align a run edge, it will move that run. If there is a linked landing, it will also move the edge of the landing, changing its size; it will not move the next run
       

      • NB. Align does not work on the centreline or edges of a curved stair

      Move Tool

      If you use the Move Tool on a landing (in plan), it will also move the connected run components; on a simple stair with one landing connected to two runs it will effectively move the whole stair;  on a more complex stair the end result depends on the other connections (see later)

      Simple Stair



      If you want to move just the landing but keep first/last riser locations, you have to use some reverse logic by moving the whole stair then adjust the risers back. 
      • You can use the Move command on the landing component, or select the two runs and move those – either way the effect is the same. 
      • You then have to drag the arrow shape handle at top or bottom of the stair to return the risers to their original location, and Revit will automatically add/remove risers from the connected run


      Complex Stair (multiple landings)

      If the stair has two or more landings, then the intermediate run will be connected to both landings.  The knock on effect of moving one landing would be to move the free run but not the intermediate run (which is attached to the other landing);  the end result is to change the landing shape or size:
      In this example the landing is selected and moved to the right and it automatically changes from rectangular to L-shaped.  The free run at the base of the stair moves, but the intermediate run cannot move - so the landing has to change shape to remain attached.
      The rule to understand here is that when a landing is moved in plan its height cannot change thus something else has to give (its shape and/or attached run locations).

      Move Run to Adjust Landing Shape

      Another "not so intuitive" workflow is to use the Move Tool on an attached run component to adjust the landing shape.  In this example the run is moved in order to add a stagger into the landing so that both runs no longer align.

       
      Revit v2014 has additional shape handles on the landing, which allow alternative more intuitive workflows to achieve this (see below).

      Move Landing in Section to change stair layout

      Stair layouts can be changed in section or elevation too.  The landing height can be changed by moving it up or down to resolve headroom issues - it will snap to the nearest multiple of riser heights, so you don't need to calculate the exact amount to move it by in order to maintain equal risers.  Revit will move the run start/end points to suit.



      Shape Handles

      For general information on stair shape handles refer to my previous blog post on the topic.  However, that does not cover landing shape handles, which are a little more confusing. In addition, the original v2013 landing shape handle behaviour was changed for v2014.

      The landing height can be adjusted by using the arrow shape handles on adjacent runs.  The number of risers on one run will be reduced, while the other is increased, thus changing the start and end locations of the stair without moving the landing in plan:

      NB. Do not use the filled dot shape handle at the base of the stair, unless you really want to change the height of the start of the stair.

      You can use a side arrow shape handle to make a landing width greater than that set by adjacent runs, but it will not let you make it less.
      Once a landing is wider, that offset is sometimes maintained if an adjacent run width is changed - but not always.  It seems unpredictable.

      Revit v2013

      In v2013, landing shape handles are fairly limited in what can be achieved.  On a rectangular half landing (U-shaped stair) there are 4 shape-handles.  
      If you want to stagger the runs where they meet the landing, it seems natural to select the landing and try to adjust it using the shape handle where the landing joins the runs. It won’t let you drag the handle towards the runs, only away from them. But this actually gives a different result - it cuts a notch and it moves the outside of the landing and to maintain its  depth (although it may be a desirable outcome in some situations).  To achieve a stagger in the landing we must move one of the stair runs towards the left (described previously).


      Landing Width/Depth

      Modifying adjacent run widths will affect the aligned landing width.  Landing sides will continue to align with runs unless you over-ride that using the landing side shape handles.
      In v2013 it may also affect the depth too, but the rules are tricky:  changing a run width will only affect the landing below the run, not above it.

      The rules seem random until you test it systematically as shown below:

      Revit v2014 Shape-Handles

      In v2014, landing shape handles have been improved, but at the same time the automatic control of landing depth has been broken - this is something you need to watch out for.  Additional shape-handles are available between each run and adjoining landing.  This allows you to put a stagger into a rectangular landing more intuitively by dragging the landing shape handle (rather than relying on moving a run).

       
      The shape handle in the stair well works similar to v2013 in allowing you to put a recess cutout into the landing.  However, in v2014 it will not maintain the landing depth - but it does give a warning message that the "depth is less than run width".  This warning is triggered by Revit comparing the landing depth with the smallest actual width of either attached run, not the minimum run width set in the stair type properties (actual width can be made smaller than the minimum required).
      Fortunately the landing depth can be corrected with temporary dimensions or the outer shape handle.  NB. the inside shape handle will not snap to any increments or give a temporary dimension, so you need to put in a reference plane to snap to before correcting the landing depth.

      Landing Width/Depth

      Modifying adjacent run widths in v2014 will only affect the landing width but not the depth.  This may be logical if you make one run wider, but if you make both/all runs wider you would want the landing depth to match - but it won't, so you have to manually alter all landings, meaning it is only too easy to get a landing depth that is smaller than the narrowest run. The outside line of the landing will only change if the outside shape-handle is explicitly moved.

      Landing Properties

      Landing heights can be changed in the properties dialog box - you need only type in an approximate value and Revit will correct it to the nearest multiple of riser heights.  This will not move the landing in plan, so the end result will be that adjacent runs will alter to accommodate the new landing height.
      Landings do not have Width or Depth instance properties, so you can only adjust those dimensions by:
      • Change the adjacent run widths;
      • Move adjacent runs;
      • Use the shape handles; 
      • Temporary landing depth dimension (v2014 only)
      • Convert to sketch (not advisable).

      Temporary Dimensions

      • Landings do not have any temporary dimensions when selected in v2013.
      • Landings do have landing depth temporary dimensions when selected in v2014.
      • Spot dimensions can be placed on landings in plan and section/elevation, but they cannot be used to adjust landing heights

      Convert to Sketch

      Automatic landings can be converted to sketch landings so that you can modify the plan shape to something that is not possible to create using automatic landing features.  Automatic landings can form a wide variety of shapes, but cannot:
      • Have rounded corners
      • Have notches cut out of outside corners (only in stairwell)
      Convert to Sketch should be the last resort, and should be avoided where possible.  This is irreversible - the only way to make a landing revert to automatic behaviour is to delete and place a new one, which would remove any attached elements, modifications or annotation.
      Once a landing is converted to a sketch:
      • It will never have shape handles;
      • It will never adjust in plan when adjacent runs are moved;  
      • it will cause error messages requiring landings and runs to unjoin if an attached run is forced to move away from the landing;
      • It will allow height adjustments

      Tips and Tricks

      • If you want to change the landing width it is better to change the adjacent run widths to control it automatically (unless you don’t want them to align).
      • In Revit v2014 it allows the landing extent/depth to be smaller than the narrowest run width. To fix this you can use the shape handles on its outer edge or the temporary dimension.
      • in v2013, landings do not have handles where it meets a run, so you can’t use shape handles there - to stagger the risers for example; you have to MOVE a whole run to achieve that. 
      • Be careful with the Move command on landings because it will most likely affect adjacent runs, and may end up moving the whole stair.
      • If you have created a top landing, any changes to the adjoining run below it may not be able to flow through to the top landing (as it cannot change in plan) and it may give you a message about "stairs and landing cannot be joined".
      • The quickest way to change a stair layout might be to change a landing height property - this can be done without even editing the stair

      Separating Model and Detail Lines in Revit

      $
      0
      0
      Have you ever wanted to select all the detail lines in a view in Revit, without selecting any model lines?  Or to select all (and only) the annotation on a view?

      Easy, you might think - until you realise that Revit makes it quite hard to distinguish between some model and detail elements.  Lets try it:

      1.  Filter

      Select a whole bunch of elements and go to the Selection Filter.  Note that it is inconsistent in how it displays the categories/subcategories:
      • It does not distinguish between model or detail lines, but it does separate out line types
      • It lists "Detail Items" as one category, when in reality they could be filled regions, detail components, repeating details etc.
      • Stairs are listed, and stair components are itemised separately
      • Text Notes are listed but not separated by type (as lines are)
      • etc - the list goes on
      It would be so helpful if it distinguished all those things properly and consistently.

      2.  View Visibility

      Since you cannot separate model/detail lines using the selection filter, you could try the Visibility/Graphics dialog box.  Then you discover some more Revit madness:
      • Some categories are wrongly assigned as model - eg. Detail Items, which are clearly annotation categories as they are view specific
      • Model and detail lines are again all grouped together under Lines in the model tab - wrong!  
      • Even worse, Lines include sub-categories for Area Boundaries and Room Separation lines, which should really be subsets of Areas and Rooms; and these line subcategories don't even behave the same way although they perform similar functions - Area lines are never visible in 3D, while Room separation lines are visible in 3D.
      • Just to add to the confusion, if you go to the Annotation Tab, you'll find a whole bunch of stuff that is certainly not view specific: 
      • Grids, Levels and Reference Planes should most assuredly have their own tab called "Datums"
      • Scope Boxes, Section Boxes, Callouts, Sections, Elevations are all to do with view definitions - and they can all potentially be seen in multiple views (some in 3D too).  What on earth are they all doing under the Annotation tab?  They need a separate tab - or maybe grouped with datums under "Setouts"?
      • Adaptive Points under the Annotation tab - just plain crazy!  And there are some other point types that you can't even control visibility of from this dialog box (eg. divide path/surface), which is endlessly frustrating.
      • For a new user, this stuff is confusing as hell.  Its a mess that is long overdue for a fix-up.
      • And it doesn't even remotely help us with this task.

      3.  Select All Instances in View

      Well, this just plain doesn't work for detail or model lines, so forget that idea.

      4.  Isolate

      If you select a detail line and try "Isolate Category" it does at least work - but again it does not distinguish between model and detail lines.  However, it does actually isolate by sub-category for lines (like the selection filter) - possibly useful, but it is yet another inconsistency.

      5.  Properties

      When you select a line and look at the properties, it is not immediately apparent whether it is a model or detail line.  But if you look carefully there is a discrete little checkbox under the Graphics Heading.  It would be very laborious checking each line by this method.
      If you select a mixture of detail and model lines the checkbox is even more discrete - you need to look carefully to see the tick is gray not black

      6.  Tooltip

      Lets suppose you had a staircase on plan and you wanted to verify if it was a model stair or whether some naughty cad drafter had taken a shortcut by drawing linework.  You could just hover the cursor over the stair lines - it will pop up with a little tooltip.  In this example it tells you the viewname and that it is a detail line;  in the case of a model line or stair it would also give you the workset and category.  You might also be lucky enough to have your eye drawn to this stair by the dodgy looking door opening onto it!!  That should be enough to warn you that it is probably not modeled as a real stair.

      7.  Worksets

      If your project has worksets enabled then you may find this useful for separation of model and annotation categories.  Because model lines are assigned to a model workset you can hide or isolate them by workset;  Detail lines and all other true annotation always belong to the view workset.  View references (sections, callouts etc) have their own worksets.

      You could try the nifty new(ish) Worksharing Display command on the View Control Bar - choose to display by Worksets:
      All the model worksets will be displayed by colour;  annotation categories will remain unchanged.  However, the fifth colour (purple*) is too dark to easily distinguish between that and black or gray lines.
      Go to the Worksharing Display settings, and change the purple colour to something slightly brighter
       Then it should be easier to visually check where the detail lines are
      It might be easier to read if you hide the surface shading by making the view wireframe.

      This still does not help you to actually select all those detail lines quickly, so here's one last thing to try:
      First, reset the worksharing display to normal.  Then . . . 

      8.  Workset Visibility

      If you go back to the Visibility/Graphics dialog box, we can get it to do one last trick.  We could turn off all the model elements to leave the annotation visible.   However, you can't just use the neat looking "Show model categories" checkbox at the top left - because the software designers have left us with a chaotic jumble of categories under each tab.
      But you can control it another way.  Go to the Worksets tab. 
      Select all the worksets listed (they are only model worksets, as you cannot select view worksets here)
       Then set their Visibility Settings to "Hide"
      Click Ok and you will be left with only detail lines, detail items and text (ie. the dumb annotation).  Warning:  all hosted annotation will not be visible either as they will only display when their host model elements are visible (ie. clever annotation - tags, dimensions).
      Now at last you can easily see and select all those dodgy detail lines

      9.  Convert

      There is a neat little utility that the software designers put into Revit a few years back - it lets you convert selected lines from detail to model and vice versa.
      Warning:  If you don't know which type you selected, it just goes ahead and converts them without letting you know which direction. 
      If you selected a mixture of model and detail lines, it asks you which you want to convert
      So now you can easily convert those dodgy stair detail lines into slightly less dodgy stair model lines - at least they will then show up on multiple views.  
      The next task will be to persuade the person who drew them the error of their ways.  I offer no advice here on that subject, except that there is one tool that could help you - if you have Worksharing Display ON, then the tooltips offer up a little more information.  Use it as you will!
      It should be noted that if you need to recreate a new central file by "save as", then all the elements in the model up to that point will have your name listed under "Created by".

      * Discrete film reference for those paying attention

      Revit Multistorey Stairs - Method 1

      $
      0
      0


      Loretto Chapel, Santa Fe
      With the new component based stair tools in Revit v2013 (and onwards) there are now three distinct ways to create multistorey stairs. In the USA it would be spelt "Multistory", but that is another story.  In the rest of the English-speaking world we put an "e" into storey to distinguish it from a tale!

      Each method has its pros and cons. Here I will deal with method 1, which uses the old “Multistory Top Level” setting.  Refer to a following post for method 2 (additional stair runs).  In Method 3, Revit now allows spiral stairs to run around more than 360 degrees to create multistorey spirals.
      Taihang Mountains, China

      Creating Multistorey Stairs – Method 1

      This method is similar to the old (pre-2013) Revit method, but there are some subtle differences described in the process below:
      • Create a typical one-storey stair; go to the stair properties
      • Change the “Multistory Top Level” property from None to an upper level on the list. Assuming floor to floor heights are all equal, it will replicate the lower stair on the upper storey(s). With this method, the upper storeys must be the same heights as the lower ones;  alternatively, if you have changed the stair Base or Top Offsets to something other than zero, then the Multistory Top Level needs to be a multiple of the "Desired Stair Height" shown in the properties.
      • Note that it will not add the intermediate landings at each storey level – you need to do that yourself (see below)
      • If the upper storey heights are different, you will probably get an error message. If you proceed it will store the warning with the stair and create intermediate storeys, but will not create the flights for the top level 
      • If the odd storey height is somewhere in the middle of the building, Revit does not build a reduced height stair like “Floor 7 ½ ” in the film Being John Malkovich 
       

      • Instead Revit gives a warning about that floor and keeps building the stair with flights at standard heights to match the original but stops short of the top.
      • If the levels are changed so that all are matching, the stair will not adjust itself automatically, nor will the warning disappear – you need to manually reset the top storey level to get it to add the missing storey. 
      • Also, if the levels all start off equal and one is subsequently changed, Revit will not adjust the stair nor will it give an error message – it just won’t align the landings with levels. You can use this to your advantage if you want a multistorey stair to go up through a non-standard storey height but with standard run heights.



      • (Hot Tip) In fact you can take this a step further (pun intended) by creating a special level just to control the top of your multistorey stair, and give that level an explicit name. The height of that level must be a multiple of the “Desired Stair Height” (typically the height between the lower two stair levels). It does not matter what other levels exist between the Top Offset and Multistorey Top Level
      • Then change your stair properties to reach that Multistorey Top Level
      • Once that is done you can move the special level out of harms way – it can even go below the lowest stair level underground. You must not delete the level otherwise the multistorey part will disappear – so it is a good idea to rename it to warn people not to remove it 
       
      •  This method will eliminate the warning message for the stair, for those who have a zero tolerance policy for warnings.

      Tread Numbers

      The stair Tread Number tool will not work on the upper flights – it will only allow you to select the “real” stair runs, not the virtual ones created by the Multistory setting.
       

      Intermediate Landings 

      Revit will not add the intermediate storey landings for you
      • You can go back into Edit mode to add a landing at the top of the defined stair by creating a sketch landing or by adding an extra run at the top, then deleting the run - the remaining landing will be converted to a sketch based component 
      • Once you complete the stair, you will notice a few very odd things happen
      • The upper stairs will not join properly with the new intermediate landing
      • This can be partially rectified by changing the run property “Extend Below Base” to a negative figure matching the landing thickness (it will also extend into the floor at the base); it won’t actually join the landing but at least it will be modelled correctly at the junction.

      Railings 

      • Revit will also create railings all around the intermediate landings effectively blocking off the upper flights; it will also join the inner and outer railings into one element
      In the next post I will show how to stop the railings from doing this

      Revit Multistorey Stair Railings - Story 1

      $
      0
      0
      Here is a curious story to tell:  This is a description of how to manage railings on a multistorey stair that has been created using the "Multistory Top Level" stair property in Revit (Method 1 in my previous post), and where you have incorporated intermediate landings on each building level of the stair.  The resulting railing will be a single closed loop - both inside and outside joined together. 

      The description below assumes that you followed the steps in the previous blog post  to get to this stage.  If your multistorey stair does not include landings at the storey levels (maybe you used floor slabs instead), none of this applies.

      Edit Railing
      If you try to edit the railing to get rid of the section running across the landing/run junction or the outside edge of the landing, you will run into problems in Railing edit mode

      After you delete the unwanted lines and then finish the sketch it will give an error message – it requires a single continuous chain of sketch lines, not a loop within a loop (even if you remove all overlapping lines)


      To solve this the railing needs to be first separated into two parts. This can be done by two different methods:
      1.  You could duplicate the single railing and then edit each one separately, stripping them back to an inner and an outer railing (Process not described here).
      2.  Alternatively, take a few steps back in the stair creation process, to edit the railings before you delete the temporary top flight:

      • Edit the outside railing (because of the top run there are two separate railings). 
      • Delete the segment associated with the extra run

      • Make sure there is a break in the loop by dragging the end of the landing rail line away from the first run line; the gap must be at least 32mm otherwise the lines will automatically rejoin! (That is about 1 ¼ inches in that quaint imperial system used in the USA).

      • NB. If you want the sloping section of rail to extend one tread beyond the lowest riser (common practise), then you need to extend that line by one tread depth before making the 32mm gap.  It may give you a slope warning, but it is not clear why because the railing will be parallel to the flight
      • One method to stop that warning is to make the sketch line overrun dimension just short of a tread depth – by 1mm say (for a 250mm tread, the railing sketch line should extend 249mm past the lowest riser)
      • Then you need to do something similar for the inside loop railing – delete unwanted segments and make a gap. You might again want to extend the sloping rail line by one tread (minus 1mm), and then make the gap in the short landing line

      • Another alternative is to extend the short landing line by at least 32mm, and leave a gap of 1mm to avoid the sloping railing warning. The best option depends on how your rail is defined relative to the sketch line location and how you want the join to work
      • Finish the railing sketch

      • Edit the stair to delete the top flight – the railings should now be separate. NB. This process does break the automatic nature of the railing sketch – it will no longer update when the stair changes

      Once you make the stair multistorey again, the railings should behave better. The small gaps may not be perfect but they are a whole lot better than the junctions that Revit makes normally

      If you don’t want the railings to run across the back of the stair landings (eg. doors onto landing), you can edit the railing to remove segments but you'll lose the side railing of the landing (because you need a continuous chain of sketch lines in the railing)

      Then you need to add a new railing to the landing sides: host it on the stair if you want it to be the same on both levels (it will become multistorey if the stair is); or host it by level if you want different railings on each landing
       

      Watch out for the “Tread/Stringer Offset property for the railing so that it will align properly with the other railings (defaults are different for automatic railings and manually created railings)

      Non-multistorey railings must be hosted by level, not on the stair – that way they can be different on each level

      However, if you find all this is too fiddly, or else you need the top landing to be different then you may need to try Method 2 for multistorey stairs. 
      You could also lobby Autodesk to fix the railing tool so that it works better with the new stair by component tool!  Maybe one day we won't have to invent these horrible clunky workarounds.

      Revit Multistorey Stairs - Method 2

      $
      0
      0
      Following on from the last couple of posts, here is an alternative method for creating multistorey stairs in Revit - typically where you need the runs and landings to be different to each other on each building level.  However, this method could also be used even when upper levels of the stair are the same as lower levels
      Staircase up the Rock of Guatape, Columbia
      Method 1 uses the "Multistory Top Level" setting to create virtual stairs above the components that you have modelled;  it also creates virtual railings on the upper levels of multistorey stairs - these have all kinds of limitations.  Method 2 does not use the Multistory setting, but relies on you modelling every component of the stair - since Revit 2103 lets you place run above run and landing above landing, you can do this.  It has its own set of limitations as we will see below, starting with a simple example:

      Three flight One-Storey Stair

      This example assumes an overall stair height greater than a typical floor to floor (4 metres in this case); or else you may need the stair to end up on the opposite directions at top and bottom.
      • Start with placing two runs, and it will create the first landing.  Note that Revit numbers the first and last riser on each run
      • Then place another run starting on top of the lowest run (this was not possible prior to Revit v2013);  it starts getting confusing already - it tells you how many risers are remaining to finish but its hard to distinguish visually where you are at

      • The top run will display riser numbers when selected, but only after it is placed - this helps you distinguish which run you have selected
      • It is much easier to see what is happening in section, elevation or 3D, although riser numbers are not displayed; you do get the shape handles in section/elevation
      • When you finish the stair, you will get the inevitable warning message
      • Revit will go ahead and create the railings, with only the usual glitches as it tries vainly to go around the inside corners
       
      • The reason for the warning is that the automatically created railings contain sketch lines that overlap each other.  Revit would not normally allow this to happen - certainly if you tried to sketch it that way yourself it would not let you.
      This warning message is a bugbear of the new Revit stair by component tools.  The software developers apparently knew that this would happen because Revit allows the stair and its railings to break the normal rules.  As users, we have assumed that the new stair tool was part one of an update process and that new railing tools would follow, and resolve the problem - but we have not seen any new railing tools yet.  We live in hope!

      So, Revit will create these "railings with warnings" but woe betide if you need to modify the railing sketch lines - it is almost impossible to untangle the overlapping lines.

      If you do need to modify the railing, refer to the next post  Multistorey Railings - Story 2.


      Multi-Level Multistorey Stairs

      Once you start going past three flights and over multiple building storeys it gets more interesting.  There are two approaches to defining the overall stair height and riser height - the options are somewhat akin in principle to the Array command in Revit:
      Both techniques will not allow you to vary riser heights for different storeys, so that aligning landings to storey levels may be a problem.

      Define Second Level Technique
      For the Top Level of the stair, you can define the upper level of the first building storey that the stair spans, even though you intend for the stair to keep going above that level.  Revit will calculate the riser height and number of risers for that building storey only.
      • After placing the first couple of stair runs, when you start placing more overlapping runs, the automatic riser numbers in plan will display the Desired Number of Risers + the additional number
      •  Just for the record, the size of these automatic riser numbers is controlled by the view scale - if you want to be able to read them, just bump up the scale temporarily
      • The riser height will be constant for every run - this means that on the runs above the defined Top Level, you may not be able to align the stair exactly with levels where upper storey heights vary
      • Once the stair is complete, you will be able to label the step numbers using the "Stair Tread Number" annotation tool.  The numbers will display the correct riser number within the whole stair rather than the "Desired +" that you get with the automatic riser numbers.  This is a different result to method 1 (virtual upper level runs created using the Multistory Top Level setting) where the upper levels cannot be numbered at all

      Define Last Level Technique
      For the Top Level of the stair, you could define the highest level of the stair.  Revit will calculate the riser height and number of risers for the whole stair
      • The riser height will again be constant for every run - this means that intermediate landings may not be able to align exactly with middle levels where any storey heights vary
      •  The automatic riser numbers in plan will be the actual number within the whole stair

      Once you complete the stair, you will get the inevitable warning message
      You can ignore this message if you never intend to edit the railing.  If you do want to edit it then refer to the next post about Multistorey Railings - Story 2

      Revit Multistorey Railings - Story 2

      $
      0
      0
      Apologies to Kevin Costner (hidden behind the railings)
      In the previous post, I described creating a multistorey stair using "Method 2" in Revit 2013 (or later) - ie. one that has runs or landings modelled above of each other.  Using that technique, you would certainly encounter this warning message when you finish the stair:
      • The reason for the warning is that the automatically created railings contain sketch lines that overlap each other.  Revit will break its own rules and allow this to happen in this situation only.
      Revit will happily create these "railings with warnings" but will seldom let you change the railings.  If you need to modify the railing sketch lines there are some tricks that you can employ to change the right sketch line, and untangle the overlaps.  We'll start with the process for a simple railing, then go on to the tricky stuff later.

      Three Flight Railing
      Let us look at the railings on the three flight example described in the previous post, where the top run sits above the lowest run
      • When you complete the stair, Revit will go ahead and create the railings, along with the warning - it may be a warning for the internal railing only, or else both railings, depending on the extent of run overlap
       
      • In this example the the base of the top run sits above the base of the lowest run;  the top run is shorter so the tops of the runs do not align - so there is only one warning
      • Let us suppose that you need to extend the lowest railing by one tread (as required by the building codes in Australia)
      •  First edit the inside railing.   Note the overlapping sketch lines for the first and third runs
       
      • You need to select the line related to the lower run - in this case we are lucky because the runs are different lengths, so we can select the longer one relatively easily
      • Extend the line to the left by one tread depth (250mm in this case) - it will now line up with the return on the landing.  When you try to finish the sketch it will give an error - this time it will not complete the railing because you have joined a sketch line belonging to the lowest run to a landing line at a different height (much too confusing for Revit to handle)
      • The only way around that is to undo;  try again but stop the line short of the landing return - the closest it will let you get is 26mm (just over an inch) before it automatically snaps to the corner. 
      • Well, that isn't quite good enough because the railing is still an inch short.  I guess you could fudge it by adding a railing termination, but that is not going to work in every situation - so that is not acceptable either.  Looks like we'll have to resort to another workaround.

      • NB. If runs one and three were the same length exactly above each other, you probably wouldn't even get to this stage because both ends of the sketch line would align with the end point of other lines - you'd get the error message whatever you do

      Warning: here follows another bizarre Revit workaround to get you out of trouble with multistorey railings (not for the faint-hearted).

      Dances With Wolves (or Railings With Warnings)

      We need to untangle the sketch lines so that you can distinguish each one and make sure that they remain linked only to the correct line in the chain when you modify them.  The solution comes from nature - a spiral.
      The way to achieve the untangling is to make each overlapping run in the stair a bit smaller than the one below it, so that you end up with a spiralling railing sketch.  NB. it is better to make the runs smaller as you go up (for reasons explained later).


      In our three run stair example it is very simple:
      • You may need to start with the railings as they were automatically created - either undo any changes or else delete the railings and replace them using the Place On Host command

      • Select the top run of the stair;  change its width property to 1mm less than it was (or try 1/32nd inch for you imperials) - in stair edit mode or just tab-select the run without editing the stair
      • You may get a warning that the Actual run Width is less than the Minimum, but you can ignore this for the moment
      •  Edit the inside railing - notice that you have two distinct lines only 1mm apart
       
      •  You can drag the end of the lower run all the way until it meets the landing line (no need to leave a 26mm gap this time)
      • When you try to finish the sketch it will probably give an error;  so you will need to drag the line back so it stops 1mm short of the landing line - hopefully acceptable;  if not, you can make the gap even smaller: anything less than 0.4mm will show dimensions rounded down so they appear correct.
      • This time it should work when you complete the sketch
      • Once the railing is fixed, you probably need to reset the run width back to its original dimension.
      Rules of engagement
      • Note the Location Line of the stair run before you change its width - it needs to be set to the opposite side of the run from the railing that you are trying to change (or center if you want to change both railings
      • Normally, if a stair is modified, then its automatically generated railings will update to match - except:
      • Once you edit an automatically generated railing sketch in any way it breaks the connection between stair and railing - so that modifications to the stair layout will no longer be applied to railings automatically
      • Because of this behaviour, you need to apply the run width trick before editing the railing sketch, so that the railing sketch becomes a spiral pattern
      • If you have already edited the sketch, the trick would only work if you delete the railing and place a new automatic one
      • Once the railing is fixed, set the run width property back to its original dimension.   The stair will correct itself, and the railing will maintain its 1mm offsets - but only if you did edit the sketch
      • This is the reason for making upper runs smaller - when they are corrected, the railing sketch lines are not dropped off the edge of the stair runs (which would happen if you initially made upper stair runs larger).  In other words, your sketch spiral needs to remain on the stair runs not in the stair well.
       
      • You are much better off working with the sketch lines that Revit creates - modify those rather than creating new lines, so that you have a better chance of Revit understanding which lines relate to which runs or landings
      • It is ok to let railing sketch lines cross over each other.  The important thing is to keep the correct lines joined to each other at corners
       

      Troubleshooting (Horizontal Railing Segments)
      Most likely you will have runs/landings on exactly on top of each other, so it won't be so simple: either it will not complete the sketch or else you might get the "horizontal railing problem".
      It quite often happens that one segment of the railing becomes inadvertantly hosted on something other than the run - it may be on one of the two adjoining landings or else the base level.   There are several things to watch out for to avoid this:
      • Make sure that the sketch line segment is on the run component in plan (or on its edge);  if it moves off the run it will lose its hosting; if the run moves the railing segment may not move with it.
      •  The railing sketch line has "Slope" properties when selected (on the Options Toolbar) - make sure it is set to By Host or Sloped
       
      • If the sketch line is set back from the edge of the run, then the corresponding landing railing line should also be set back from the landing edge (even with only one landing).  Sometimes offsetting the landing sketch line by 1mm will correct the horizontal run railing line problem

      Multi-Level Multistorey Railings

      Once you have mastered this technique for three run stairs, you can apply the same technique to stairs with four or more runs.  All you need to do is make each pair of runs smaller as you go up the stair. 

      Landings
      Don't forget that the landing sketch lines also need to be offset. 
      • You could do this by adjusting the landings themselves - but it gets fiddly as landings don't have depth properties so you need to drag the shape handles, which will not snap to increments.  Refer to modifying landings
      • Alternatvely you can just offset the landing sketch lines yourself - it is easy to distinguish which is which because their lengths are different due to changed run widths
       
       
      I hope this technique helps someone to get their railing sketches to work correctly - it is not as bad as it sounds once you get the hang of it.  It won't help resolve the railing junctions, but that is another story!

      Revit's most hidden commands

      $
      0
      0
      Recently I did a Revit demo of divided paths and repeaters.  Afterwards someone said they had spent a day trying to figure out one simple thing I showed - they searched everywhere, including Youtube for videos but could not find out how to control divided path node visibility in a project.  That got me thinking about some really hidden away commands and features in Revit that are not at all

      Revit contour label orientation

      $
      0
      0
      A few times I have tried to search for information on how to control the orientation of contour labels in Revit - always without success.  It seems that Revit likes to orient the text of contour labels to be readable only if you are walking uphill on a Revit toposurface.  That works fine until "uphill" happens to be in an Easterly direction or even worse in a Southerly Direction, then the text

      Swept Blend using Profiles in Adaptive Components

      $
      0
      0
      This process might seem obvious to anyone who regularly works in the Revit Conceptual Massing Environment (CME), but for those who are starting out with Adaptive Components, it might help the transition from the old Revit family environment. The question is: How do you create a solid extrusion or sweep using a profile in an adaptive component ? A lot of people advise against nesting components

      RTC and Revit Egress/Escape Path Update

      $
      0
      0
      RTC Australasia 2013 At the Revit Technology Conference (RTC) in Auckland a couple of weeks ago I learned a whole lot of new tricks in Revit.   I usually think that if I learn one or two things that might save me hours or even days of work in Revit then it is worth attending the conference.  This time I learned a whole swathe of new techniques, including (but not limited to): How to scale

      Revit Step Pattern Repeaters

      $
      0
      0
      I saw a question recently about how to create a step pattern repeater, so of course I had to figure out how to achieve it. You could just use a 1/2 step or 1/3 step curtain panel pattern on a divided surface, but that may not be exactly what you need, and the predefined surface patterns are not always easy to modify if they don't suit your purpose. Here are a couple of ways to achieve it with
      Viewing all 276 articles
      Browse latest View live