Revit Content That Works

April 20th, 2013

Too often Autodesk® Revit® content is viewed as not much more than a 3D CAD block. Much time is wasted scavenging the Internet in hopes of finding a family that never quite suits its intended purpose. It doesn’t have to be this way. By investing a little time and thought there will never be a need to download again. Custom Revit content can be developed that meets specific design criteria, reduces repetitive tasks, and fully leverages Revit’s automated tagging and scheduling features.

Let’s use a standard floor poke-thru as an example. What is the advantage to building a custom family in-house versus using one that has been obtained from a third-party site? Downloaded versions likely are nothing more than the housing of the poke-thru. This is adequate for reserving real estate early in the design process. It does nothing to address the additional views in which it may appear, or scheduling.

The thing with an item such as a poke-thru is that it is an assembly of various components, possibly from multiple manufacturers, that need to be specified. These items that aren’t being accounted for are equally important. Quite often there can be several different configurations of the same model poke-thru in a job. Capturing this information correctly is critical.

For a new Revit user, the approach would most likely be to place the device on the floor plan and tag it as usual, then import an existing 2D detail into the Revit model and edit it manually. This approach not only creates extra work, it also ignores the scheduling and tagging features at which Revit excels. Editing tags manually always leaves an open window to error that Revit can eliminate. By creating a family of nested sub-components, all the information is placed directly into the model with the family and then extracted into different views as needed—just as Revit is intended to work. Doing this creates a truly integrated workflow.

In the standard 1/8” plan view there are three poke-thrus (see Figure 1)—all of the same make and model. Graphically they are all represented as a double circle. 

Figure 1

From cutting a section view, it’s pretty apparent on the outside that they are all the same (see Figure 2). The only difference is that the one on the left (PT1) is rotated 180 degrees. However, the Revit tags are reading them as unique objects (PT1, PT2, and PT3) in plan and section views. 

Figure 2

Making a detail view to be placed on the details sheet will show what qualifies them to be tagged differently. This is done by creating a callout and dropping the view range several inches below the surface to see inside.

Figure 3

The detail view in Figure 3 shows that all three devices contain HDMI, VGA, and duplex inserts, in three distinct configurations. All of these pieces need to be accounted for and appear in a schedule or BOM. For the demo here, there is a very limited scope on the devices and types of inserts being used, so it isn’t too hard to figure out specifics.

Dealing with a live project where there can be a wide variety of inserts used, possibly in different models of poke-thrus and in larger numbers, it can be difficult to keep track of. By using this nested component family, Revit is now keeping track of everything (see Figure 4). 

Figure 4

Placing one instance of each unique poke-thru family in the model is all that is needed to cut consistent, integrated plan, section, detail, and schedule views. Nothing is ever drawn, saving time. The only manual editing needed is for populating the initial parameter information (Type Mark, Description) for the family. This can be done easily in the schedule. It is just as easy to make edits in the schedules as it is in a spreadsheet. The need to draw an elevation, draw a detail, insert a spreadsheet/schedule, or manually edit the tags has now been eliminated. Any edits made in one view will automatically update in the other corresponding views.

So how can families that make our lives easier and recoup the investment on Revit be designed?

Diagram: Before modeling anything, lay out a basic assembly diagram (Figure 5, PT1). This will help with thinking through how the pieces will fit together in the finished family, and how to build the individual components. While the individual poke-thru pieces are simple, it took some trial and error to determine the correct sequence. Most items likely won’t require quite this level of coordination to construct. 

Figure 5

Name: How the individual components and finished family are named is very important. When the family is inserted into the project it will show up in the Project Browser under the Families menu. All of the nested sub-components will also show up there as objects that could inadvertently be inserted into the model individually. The naming scheme should identify these items in a way that will tell the user they are part of a greater assembly, and not to be placed on their own in the model. 

Figure 6

For the demonstration, a “C” at the beginning of a name is used to identify a sub-component (see Figure 6). Notice that the poke-thru housing is considered a sub-component. Find a variation on this theme that is compatible with the established naming convention.

Model: There are always multiple ways one can extrude and create Revit content. The individual components need to be created based on how they will be assembled in the finished family. Give some thought to what direction the extrusions will be created  (bottom up, left to right, and so on) and determine the correct origin point for easy insertion.

The duplex receptacle in the sample was created from one side out so it will line up correctly to the pieces next to it. Had it been built with the insertion point dead center, positioning it correctly would be difficult. To ensure that the sub-components can be placed quickly and accurately, reference lines are built in as snap points. Be sure all components are face-based families as the face they will be hosting to is each other. All component families should be set to the same Family Category. If this isn’t done, items won’t schedule properly and it will not be possible to automatically swap out components.

Detail: One of the most difficult discussions pertaining to Revit content is how much detail to model. There is no one right answer. Different stakeholders require different sets of information from the same items in order to complete their work. The philosophy applied here is that an object should be modeled to a level of detail that allows it to be identified visually by a non-technical person (the person writing the check).

Figure 7

In the case of the poke-thru, the client should be able to view a model of the exciting new space, and when they see the device on the floor say, “That’s the thing where I plug in my laptop to do a presentation. I understand.” (Figure 7) Accomplishing this took only a few basic extrusions and revolves. There were three critical dimensions to capture per manufacturers spec—the overall diameter, the core diameter, and the overall depth. Even with that covered, it is still basic in appearance.

To capture the view placed on the detail sheets requires a different approach. Those HDMI and VGA inserts may be tiny, but they do have a lot of detail. All that detail could be modeled. It would take a considerable amount of time and likely lead to a crashed model. Instead, a very basic model of the insert is created. 

Figure 8

An iso view (Figure 8) shows that the HDMI insert consists of two basic extrusions and one void extrusion. Revit will hardly even notice it’s there. To present a heightened level of detail, create and insert a 2D detail component. Detail components only show up in the view in which they are placed (see Figure 9). For all the inserts, they have been put into the plan view. Notice that it doesn’t show up in the iso view even though it is part of the family.

Figure 9

Because detail components are only lines, the memory they take is nominal. As an added bonus they served as a template here for creating the extrusions, which saved a few minutes.

Schedule and Parameter: The sample schedule is pretty basic. It shows the various components in the model, the nomenclature used to identify each component, overall count totals, manufacturer, and model number. This may be all that is required. By tagging items in a different manner and adjusting the scheduling filters, a schedule could be created that tells us what exact items are in each poke-thru family, or each poke-thru by space.

This could be very useful on a large project where there are many unique configurations to be located and assembled. If it doesn’t exist already, a shared parameter can be created for any piece of information that needs to be scheduled or tracked. Don’t think of this just in terms of what is placed on the issued sheets. It is very useful to have “work” schedules in a project that can monitor what is going on in the model, or provide a reality check. All schedules can be made part of a project template so at the beginning of a new job they are already in place. All that is required is to look at them.

Miscellaneous: Other features that could be added to this family might be a void extrusion to cut the core. Do this first, before inserting any sub-components into the new family. It doesn’t appear to work on linked models, though. When the family is completely assembled, add conduit connectors. If this is being used in the same model in which electrical is working, add a connector to the duplex so it can be circuited. If desired a lid can be added and given a yes/no parameter to control visibility. Beyond that the sky is the limit. What does the job require? (Figure 10.)

Figure 10

Like all other design software, Revit is nothing more than a blank slate. Its custom content editor opens a new world of design possibilities and the flexibility to create. Making use of these features is key to getting the maximum return from the software. By learning to work with Revit, Revit will work for you.

Join AUGI Today

Become part of the largest Autodesk community

About the Authors

Andrew Baranowski

Andrew Baranowski

Andrew Baranowski has the best of both worlds. Working at ESD, the largest MEP consultancy in Chicago, he is heavily focused on developing Revit processes and creating custom content for the low-voltage department, while anchored with the AV team as a CTS. Since 2009 he has championed Revit’s benefits, and worked through its frustrations to find the greatest ROI for the team. As a member of the NIBS-BSA, Andrew serves on the Product Development sub-committee for V3.0 of the US National BIM Standard. In a former life, he spent seven years in the Chicago Carpenters Union as a Journeyman Cabinetmaker, designed plastic components for a global manufacturer of office supplies, and received a Bachelors degree in Arts Management from Columbia College Chicago. Contact him at


Appears in these Categories