After flirting with BIM for years, our office is now committed to doing most new projects in Autodesk Revit.
I’m glad about this because BIM is clearly the future of design and construction. It’s a smarter way to approach construction documents and building design.
So I’m now getting the opportunity to become familiar with Revit as I use the Revit viewer to survey in-progress models for information needed to edit specs for projects. Our office has not switched to one of the specification writing add-ons that assist in coordinating Revit models with specs. I want to look at that again soon, but my impression is that our use of nomenclature and annotations, as well as Revit families and variations thereof in the models, will have to get way more sophisticated, rigorous and consistent before it makes sense to make the switch.
So far, there’s not much difference between how I approach information-gathering for spec writing on Revit projects versus Autocad projects. Or for that matter, just using hard copy in-progress drawings.
I look at the model and/or the drawings on screen, and edit as much as I can with that information. Then I talk to the project teams to iron out the issues.
I still find differences between the model and written documentation such as memos and my notes.
- A project memo may indicate that the client wants ceiling-hung toilet partitions; the model shows floor-supported partitions.
I still find nomenclature which doesn’t conform to industry standard lingo.
- A Room Finish Schedule may indicate “acoustical tile” when the product shown on reflected ceiling plans and the model property box is “acoustical panel’.
I still occasionally find inapplicable details, mainly “drafted” details dropped into the model, or maybe just left there because they’re part of a project template of some sort. Or awaiting editing and refinement?
- Ceramic floor tile edge details on a project that has no ceramic tile flooring scheduled in the finish schedule.
I occasionally find things in the models that are either erroneous or just suspicious. The main problem seems to be that objects are chosen because their geometry comes close to the object that’s required, but they’re just not quite right. A few examples:.
- An opening assembly in an exterior wall may say “interior” in its property box.
- A wall assembly may have a stated R-value in its property box that’s markedly different from what an experienced designer can ballpark using known insulation r-values for that type of insulation.
- Some things still show up as generic, such as “filled region - Sealant” or “rigid insulation”. More information than that is needed in order to edit specifications.
- An opening schedule may indicate that a particular opening has a flush door with no glazing, while an interior elevation in which the door appears shows it as having a vertical glazed lite.
- The property box for a roofing assembly may refer to EPDM roofing, even though annotations in the model, and the specs, specify TPO roofing.
- The property box for a coiling steel door may describe a rated assembly, while the Opening Schedule has the door as unrated.
The cure for all of these situations, it seems to me, is twofold:
- Patiently navigating the Revit learning curve, steadily building the skills to make the families and objects more accurate and reliable.
- Conversation and coordination with the rest of the project design teams. Lots of questions to clarify intent and careful editing, both model and spec, are necessary to create a unified, consistent, reliable package of information.
I’m not complaining about these things, mind you. I’m just pointing out that BIM is not a panacea for either architects or spec writers. I’m grateful to finally have the chance to become familiar with BIM.
But just in case someone out there thinks that using BIM means that spec writing is now completely automated thanks to Revit, and that the model writes the specs: Not quite yet.