## Monday, February 22, 2010

Long time no blog... or whatever. :)

Anyway.

After a comment by Rick on this post, I was wondering whether everybody feels this way. I mean, we've all been (forced to) getting used to the Ribbon. And I have to say, it's been a fairly easy transition.

The Factory have made a good effort and I find the setup (mostly) logical. And I want to know how you guys feel. Do you find everything is in its proper place, are you searching for some functions or would you rather go back to the GUI of R2009?

Go forth and vote! And feel free to leave a comment on what you were looking for or why you hate the Ribbon.

Labels: GUI, Vote of the Week, VoW

## Monday, February 15, 2010

The results are for the Vote of the Week, weeks 6 and 6 of 2010.

**Not unexpectedly the majority of you uses dwg/dxf to exchange information between the parties concerned.**

What I found interesting is that there is nobody using the ifc-format and that there are still a few people using paper.

At the office we are using a combination of pdf and paper, but are trying to start using other means, like native rvt or at least dwf.

Labels: File Types, general, Vote of the Week, VoW

## Thursday, February 11, 2010

In my last post I described the math for creating a catenary. Now it's time to get down and dirty and create a family that does this.

The plan is:

- Create a family with a line with a dot at the end that I can hang my line onto - we'll call it "catenary line".
- This family will move the dot horizontally and vertically depending on user input.
- The height of the line will be computed in this family.
- We then create a family "catenary", into which we nest the "catenary line" and which does computing for
*c*.

I start by creating a new

*Conceptual Mass*(

*R > New > Conceputual Mass*). And since I want the catenary to be vertical, I set the work plane to the

*Center(Front/Back) - Reference Plane*. On this plane I draw a vertical

*Line*and place a

*Reference Point*at the end.

I then add

*Dimensions*to it and give these dimensions

*Parameters:*

**x**and

**y**.

Now for the "fun" part: the math. I open the Family Types dialogue and enter the following parameters and formulas:

A quick runthrough:

Parameter | Explanation |
---|---|

nr | number of this line (segment) |

nr_of_segs | number of segments the catenary will have |

x | The distance of the line from (0,0) calculated by multiplying half the length of the arch with the position of the line (segment number divided by total number of segments) |

xn | conversion of the distance from (0,0) to a pure number (as opposed to the length x) |

y | height of the line at this point (this is where the catenary-formula comes in) |

tlen | half the length of the arch |

u | number calculated with Newton's Method to get to a |

a | constant to put into the catenary formula |

e | Euler's Number |

Armed with this I can now load this family into the catenary family and start creating the catenary. (Which I'll discuss in the next post).

Labels: Catenary, Conceptual Mass, Formulas, general

## Tuesday, February 9, 2010

As promised in this post, I'll go into making a *parametric family* for a catenary arch. I'll do this in two parts, since I can imagine that the actual math is not that interesing to everyone.

So, if you've always hated math and don't want anything to do with it, go here instead and stop reading. This is gonna get boring.

For everyone else, here we go.

The **goal **is: Make a *parametric family*, where I can enter a *width *and a *height *for the arch and get a catenary.

First the general formula for a catenary is this:

and it looks like this:

As you can see, the graph moves upwards with every increase in

*a*(I have used

*c*in place of the

*a*, just because I had found the formula with a

*c*first). But what we need is a graph that goes through (0,0). So we'll modify the formula:

and we get:

Great. Now we only need to find the right

*c*to fit our purposes and off we go. The way to do this is to substitute the height of the arch for

*y*and half the length of the arch for

*x*and solve for

*c*. Easy.

In this example I want an arch 30m wide and 5m high:

Now we run into 2 problems:

- Revit does not know what a cosh (Hyperbolic Cosin) is.
- You can not solve this function for
*c*.

Revit does know how to calculate

*e*to the power of

*x*(Well, sort of. You do have to tell Revit what

*e*is, but oh well...). So that's sorted.

To tackle problem 2, we use the Newton Iteration to find a value for

*c*:

The two functions are:

and

We put this together and within 5 or 6 steps we have a value for

*c*that works for our purposes.

...and all this to create a simple arch!

OK, so much for the basic math. In the next part I'll walk you through how I actually set this whole thing up in Revit so it works.

My thanks go to everyone at wikipedia for their articles on the math and to John Forkosh Associates for their online LaTeX machine.

Labels: Catenary, Conceptual Mass, Formulas, general

## Thursday, February 4, 2010

I am currently working on redesigning an existing building. I have the drawings so I went ahead and set the whole thing up in Revit.

In order to start the designing process, I wanted to make an

*in-place mass*and export this to the

*Conceptual Mass*editor. So I started my mass, gave it a name and started modelling.

But when I tried to export it, I found no way of doing so. I tried via grouping it and exporting the group, exporting as a family, and a few other "official" ways.

So I figured, what the heck. I'll do it quick and dirty: copy & paste!

Surprisingly this worked, but only with a little trick. When I tried to copy the entire mass, it wouldn't work. But when I just grabbed the volumes I had created

**within the mass**, I could paste them in the conceptual mass editor.

Or, to be more clear, this is what I did:

- create an in-place mass in a project file (
*Ribbon > Massing & Site > In-place mass*), give it a name and start editing - once done with the
*forms*within the*mass*, copy the*forms*by selecting and pressing [ctrl] + [c] *R > New > Conceptual Mass*, select the conceptual-mass template and open- [ctrl]+[v]

*in-place mass*in the

*conceptual-mass editor*and can play around as much as you like.

Labels: Conceptual Mass, Families, R2010, Sketch-Mode

## Wednesday, February 3, 2010

For a project I want to use the form of the catenary, which is the form a free and flexible chain adopts when hanging from 2 points. If you invert this you get a very stable arch form.

One example for this is the arch in St. Louis:

This form is based on a mathematical formula involving e (approx. 2,71828182...)

Revit is supposed to know this number and by using the command exp(x) you can calculate e to the power of x. At least that's what the manual says.

Unfortunately, exp(x) will calculate 10 to the power of x. Took me half an hour of troubleshooting my formula to find that out.

Once I have the catenary done I'll post a tutorial on how to make one.

## Monday, February 1, 2010

Here are the results for this weeks **Vote of the Week**:

Looks like a lot of you use the render function in Revit for final presentations.

We have decieded against that, as we are afraid that the models will get too big and cumbersome to be effective. Please comment on whether that's the case and what kind of experiences you've had.

Labels: Renders, Results, Vote of the Week

Today we were trying to print a drawing with the rooms filled in in gray. This is what we got:

Not really satisfactory.

After some googleing and try-and-error we came up with this:

Much better!

The solution: in

*R > Print > Settings Setup*switch the

*Hidden Line Views*setting from

*Vector Processing*to

*Raster Processing*.