Pan Joists

[Scott Wilson]

What is everyone doing for pan joists? For me, it would be preferable to model the floor as one big thick slab and insert the pans as voids.

I tried creating a void sweep "beam" floor host family to see if this would work. The family comes in but could not be used to "cut" a standard concrete floor slab.

I tried the same thing with an in-place mass instead of the floor slab, and the void family wouldn't cut it, either - even with the cut tool. I could, of course, cut the mass with an in-place void.

Do voids lose functionality when brought in as a family? Am I missing something?

SW

[david_peterson]

We've done the opposite, create the stem as a beam element, the slab, as a slab, and then in section you need to use the edit line work tool to change it to an invisible line. But we haven't had to use this method in a building yet, and I know this presents issues. But I'm not sure we are even going to use the Schedules in RS for our joist, so I'd call this more of a Hack than a solution.

[Scott Wilson]

Seems you could get a much more accurate model that's easier to generate by modeling the pans instead of the joists/beams. With all the crazy conditions pan joist systems sometimes generate, it would be much easier to locate pans and control where they start and stop than to try to model the concrete. "Pan voids" (with parametrics for pan depth, pan width, draft, etc...) laid out similar to beam elements in a floor would be a really powerful tool.

SW

[david_peterson]

I'll agree that voids would be a better idea, but did you get the you void beam family to work? That's why we started with the other route.

[Scott Wilson]

I never got a family to work. I wound up creating a pan joist profile with an accompanying catalog file. For the floor, I created an in-place family with a concrete mass the thickness of the pan joist system. Then I used a void sweep with the pan profile to create the voids. This works great for creating the 3D geometry, but in plan Revit won't generate hidden lines for the geometry below the slab. Anyone know why?

SW

[Paul Andersen]

Hey Scott, nice model and creative approach.

Probably your best route for the hidden lines in plan view would be to:

1.) Set your plan view to wireframe and Edit your in-place slab family.
2.) Choose Symbolic Lines from the design bar.
3.) Select Hidden Lines [projection] for your line type.
4.) I would recommend selecting the pick arrow from the options bar and check the box next to lock. (this is the fastest way to select the lines you want to show as hidden and keep them locked to the underlying geometry in case you have spacing or pan size changes down the road) It would be nice if there was a 3rd pick option of a crossing window for tasks such as this.
5.) Finish your family and set display back to hidden line.

You can also use the linework tool but it is view specific resulting in duplicated efforts on enlarged plans and sometimes they can go a little crazy.

[chad_lueptow]

I'm surprised that RS3 hasn't addressed this. Regardless I've created a pan joist family, which can be scheduled. Which is probably the right approach since they will be scheduled similar to concrete beams. Once you place your slab "over" the joists the correct hidden line representation is displayed. In section you will need to trim the line between the slab and the joist. If you'd like to take a look at this family please email me directly. clueptow@flad.com

[Paul Andersen]

Hey Chad, we've used a similar approach to yours and David's (creating an external beam family for the stem) and I agree if you need to schedule them it's probably the best way to go. You can also join geometry between your family and the slab which should clean up your intersecting line automatically in section. I think RST3 has at least two valid approaches to modeling this type of system depending on what your needs/goals are (external stem family or in place voids for the pans). Some other benefits of the stem as an external beam family are the ability to use the beam system tool to control your spacing as well as being able to construct an analytical model if necessary. My only gripe with the current tools are that if you wish to model the beam or stem to have it's top be the same as the top of slab it does not correctly display as hidden line in plan regardless of whether you join geometry between the slab and beam or not. We've also experienced some sluggish behaviour if you join to much to the slab.

[Scott Wilson]

Ideally, I would much prefer to model the joists as beam elements for the same reasons. We typically schedule our joists and beams, and a beam system would be much easier to lay out.

The problem is we see many pan joist buildings with irregular geometries. Pan joist systems in these buildings result in masses in the concrete that you can't model with beam families (unless I'm missing something). You are basically having to sculpt the concrete to get it in the right form. It seems that modeling the pans, which are always rectangular, is the path of least resistance.

Chad and Paul, with a beam system approach, how are you able to model things like angled joist framing, haunched girders, slab drops, etc... and still have your plans and sections look correct?

SW

[Paul Andersen]

My last two projects were relatively uniform and didn't require any in-place families but as you have pointed out this is normally not the case. Most concrete projects are going to require a fair amount of in-place or custom external family work. For most situations I think I'd lean towards leaving the slab a slab. Beams and joists as an external beam family when possible and combinations of in-place solid and void extrusions, sweeps and blends to model the rest.

[david_peterson]

My last two projects were relatively uniform and didn't require any in-place families but as you have pointed out this is normally not the case. Most concrete projects are going to require a fair amount of in-place or custom external family work. For most situations I think I'd lean towards leaving the slab a slab. Beams and joists as an external beam family when possible and combinations of in-place solid and void extrusions, sweeps and blends to model the rest.Seeing that this looks to be a big pain, Has anyone ever done a large scale complex pan joist, or waffle slab CIP structure in RS yet? And on another Note, Has anyone heard from the factory on what they plan to do about these types of free form CIP structures/foundations? There has to be a better way.

[Scott Wilson]

Bumping this back up to the top...

Anyone have any anwers to David's question?

[jrichardson]

We are currently working on a 4 story wide pan and joist system. We have developed in house families for the beams, joist, and bridging. Parameters are built in to show taper left or right.We have a couple of different beam families to accommodate shallower joist framing into deeper beams.

With the current limitations of RS3 we had to build in a workaround extrusion at the bottom of the beam to allow the hidden lines to show properly on plan. The extrusion is on a subcategory and the 3d geometry is on another subcategory. 3d geometry gets turned off in plan to display properly. We used slabs for the infill areas.

It is working OK for us on this project. I am hoping it can be fine tuned in the future. The model has also become a little bit slower than usual. I am guessing it is because of all of the extra joining of geometry that Revit is doing with our modified families. We have not attempted any analysis with this method. It has allowed us to model it as it is built and flip up the sections vary quickly and accurate.

jamie