Glass Balustrade panel railing

[damon.sidel]

I'm trying to create a railing that looks like this:
railing.JPG
I've always found railings that use panels quite cumbersome. I want the handrail to be continuous as well as the L-channel at the base. The balusters are held in from the end and corners and the last panel is smaller than the rest. I'd like the smaller end panel to be constant, but the central panels have to change width depending on the condition so that they are equal.

Since I want the railing to turn the corner, I'd like it to be one railing. However, the size of the central panels from the front to the side will not be the same.

Any thoughts? (The image you see is a generic model family where I just built up the geometry for the whole balcony. I'd prefer to make the balcony a link so I can have real floors, walls, and railings.)

[Dimitri Harvalias]

As much as I hate hacks like I'm going to suggest it's often easier than dealing with the limitations of the railing tools.
Option 1 is curtain wall.
Option 2 would be a wall type. The handrail and bottom channel would be integral sweeps and the posts and clips could be a face based family that would 'cut' the wall to create the gaps. Simple and quick, quantifiable.
In both cases I'd make groups for the common layouts/configurations and copy them as required.
Just as an aside, your engineer or building authority might have an issue with the 'secured one side' configuration at the corners and ends of the guards.

[damon.sidel]

Dimitri, Thank you for your response. I hadn't thought about either option. However...

Option 1 is curtain wall.

...this seems like it would be just a tricky. Only benefit would be that I could adjust the spacing of the panels with the grid lines more easily than the railing. Still there are a lot of pieces to worry about.

Option 2 would be a wall type. The handrail and bottom channel would be integral sweeps and the posts and clips could be a face based family that would 'cut' the wall to create the gaps. Simple and quick, quantifiable.

Again, quite a few pieces to worry about. Interesting, though... I'll look into this one.

Just as an aside, your engineer or building authority might have an issue with the 'secured one side' configuration at the corners and ends of the guards.

Yeah, you've got a point there. I'm not entirely happy with the configuration in general, but it is only conceptual at this point. It is for a look right now... just getting into the details for the balconies, so this is really just to get a reaction.

Thanks again.

[nancy.mcclure]

Railing_Formulas.pngI'd make this a formula-driven component family, with 2 nested components (glass panel & vertical support/clip) and 2 sweeps (bottom channel and handrail). Insertion point would be the exposed corner of the balcony. Set up parameters that define the 1st/last panel width (narrow sections) and takes overall side length - 1st/last / max panel size. Set mid-field panels to adjust to that derived length, then array them in plan based on a resulting panel#. Set sweep paths to adjust to side lengths.

[Dimitri Harvalias]

Did you just whip that up Nancy or was it just sitting in the closet somewhere
I would strongly second your suggestion. Ultimately families do offer the most flexibility and are probably most efficient for something like this. The user always needs to balance needs (how fast do I need it, how many conditions does it need to satisfy, does it need to schedule and how, is this a one off or worth 'having in the closet') against all the other stuff (am I able to create this, do I want to create this, does it involve elephants or cows (shout out to Marcelio)).

[damon.sidel]

Nancy,

Thank you for the thoughts. That would indeed be a great way to do it. I'll have to think about the use as Dimitri says. I've been able to model the 4 balcony types I need as generic models. They each have the same style railing, though, so when I have to change/edit the railings, I'll have to do it in 4 places. It may well be worth creating a family, although I don't like the prospect of creating such a robust family that duplicates what railings (don't) do. Oh well, just have to wait for 2014 and hope they update the railing tool!

[nancy.mcclure]

Did you just whip that up Nancy or was it just sitting in the closet somewhere

It was created for a specific project, but has become a great training tool on embedding formulas in arrays, and just general analytical thinking of conditional parametrics, organized to be 'driven' by only a few key parameters by the end user. Also, once the functionality was proven, the 'pieces' swap out very nicely, so this framework can swap out glass panels, wire cables, different support brackets. Well worth the investment of time to build (about 3 hrs, including the nested components)

No cool meshy elephant skin, though.... I admit!

Damon, that all said, I do advocate for using curtainwall families to generically represent the design intent during schematics. Quick, easy, good for quick length/cost validation.