Adaptive Component Family Overhangs Border with Tile set to Partial

[rauburn]

Hi,

I am working on a custom ceiling tile grid and have an adaptive family nested into a "Generic Model Pattern Based.rfa" that I have applied to a Divided Surface on an In-Place mass in my project. Long story short - the adaptive component overhangs the divided surface instead of cutting off at the border. I have set the Divided Surface Border Tile to "Partial," but it still runs past the border. I have attached a screenshot showing the desired outline of the ceiling with a grayed out pattern showing what Revit is giving me when I finish editing the mass. Any suggestions how to force Revit to cut the pattern (and the nested generic model) off at the edge of the mass?

[ToanDN]

It is documented here.  

https://knowledge.autodesk.com/support/revit-products/troubleshooting/caas/sfdcarticles/sfdcarticles...

However  @RDAOU  has shown he could overcome the bug in 2022.  

[RDAOU]

Hi @ToanDN 

That post: I didn't have an issue to overcome. I simply modeled a 1/2 step the way I recalled we did. I never said it was 2022

This post: The image in the OP is of poor quality. As bad as it is, the image doesn't look like a 1/2 Step.  I do not mind looking deeper into it if a family and a Revit context is provided (but wont be today). How is New Orleans 

5 post 3 ideas + a reply and a query on pattern based... lots of people using pattern based lately or wht?! no one fancied them much in the old days...The generic adaptive was THE thing! What happened?!

[rauburn]

The obvious workaround is to manually place all of the adaptive components and tediously create one-off variations for each of the unique edge conditions. Hoping that there is an easier way before I go too far down that path...

[rauburn]

Thanks both for the quick reply! @ToanDN my problem is slightly different as the overlap is by a whole-panel, not  just a uniform 2" beyond the edge. I will keep that post in mind however, in case I encounter that issue in the future.

@RDAOUNew Orleans is getting back on her feet, and I will be happy to buy you a beer next time you visit! I am attaching a simplified project file with an In-Place Mass family and the nested components (all files in Revit 2020). And I apologize the Super Panel family is a bet of a mess, but as far as I can tell it is not a problem with the Super Panel family, rather something I am missing when it is hosted to the divided surface in the project. The strange thing is that the panel works exactly as expected inside a conceptual mass family (nice, neat, trim edges), but of course when I try and bring the conceptual mass into my project, the hosted panels drop-out. Why would this panel work on a conceptual mass but NOT on an in-place mass? Much appreciate your help and guidance! 

I am digging into Pattern Based because sadly, the Repeater tool has been missing since Project Vasari if I recall correctly? And of course Pattern Based beats the pants off manually placing families onto nodes in the divided surface any day.

[ToanDN]

What?  Repeater tool is always available.

Anyhow, my only suggestion is model the mass with 2" offset inward to compensate the overhang.

[rauburn]

Thanks again, @ToanDN . I have attached another image that I hope comes in at a higher resolution so you can see that it is not a uniform 2" offset, rather the border panels are simply missing completely instead of coming in "partial" as I specified in the properties of the divided surface.

I do however  appreciate your kind reminder that the repeater tool DOES in fact exist, it's been so long since I used it I forgot it is that little icon with a teeny-tiny "R" on it, time to check my eyeglass prescription!

[barthbradley]

All I see in your two screenshots is a bunch of funny-looking "Rectangle" panels.  How'd you get there?  Post your file.  That would be the easiest route to a solution.    

[rauburn]

Thanks for taking a look @barthbradley. I posted my files up above in Message 5 but just in case they didn't go through I have attached them again to this message. Basically what I am trying to achieve is the trimmed partial panel effect as shown in the conceptual mass family inside of the project file with the in-place mass. Alternately, if there was a way to bring the conceptual mass into a project file with the panels attached to it and visible, that would work for me too!

[barthbradley]

...yep; I just noticed that 5 seconds ago.  Sorry.  My bad.   

[barthbradley]

...yah, your panel is badly constructed. Have you flexed it?  

[barthbradley]

...there's a lot of work to be done here. I may bail.  Betcha @RDAOU or @ToanDN beats me to the finish line. 😉

[rauburn]

Thanks, @barthbradley. The panel is over complicated and I am working on cleaning it up a bit. But even with a simple adaptive component I still run into the issue of panels failing to trim at the border of any divided surface.

I guess my simple question is - am I missing something? Or expecting a Generic Adaptive Component family hosted to a Divided Surface/In-Place Mass family to do something it's not designed to do (i.e. neatly and automatically trim around the edge of the surface?) It seems like it's possible - in my example of the Conceptual Mass Family everything is working perfectly. I hope this doesn't sound like a total NOOB question, but is there some functional difference with how Revit treats divided surfaces on Conceptual Mass families vs. In-Place Mass families?

[barthbradley · Accepted solution]

Yeah, I don't understand why the nesting of an Adaptive Point Family to another Adaptive Point Family, which is essentially what the Pattern-Based Panel Family is.  I have routinely nested Work Plane/Level Based families (some times three deep nested families) into Pattern-Based Families, but I don't think  I have ever nested an Adaptive Point Family.  But, beyond that, your Adaptive Point Family is constructed poorly.   So who knows;  maybe  if it were properly built it could work.  

Here's a tile roof done with a Divided  Surface and a Pattern-Based Panel that is one Tile. The Tile is made up of 3 nested families (tiered nesting; A into B, B into C, C into the Pattern-Based Panel Family).  

Here is your Mass Divided Surface using a simple  Custom Rectangle Pattern-Based Family.  Works well. However, it blows up when I try to use your Custom Rectangle.  

Lastly, and I think it was mentioned above already, you could not do a Pattern-Based Panel at all. Instead, use the Repeater Tool with your Adaptive Family.  There are a lot really good videos on-line about this. Just Google them.

Repeat Adaptive Components | Revit Products 2016 | Autodesk Knowledge Network

[rauburn]

Thanks again, @barthbradley! Once again, simple is best and the lesson learned here is to model everything directly inside a Pattern Based family straight from the get-go, no need to nest an adaptive component inside.

On a side note, @ToanDN I am now seeing the 2" overhang problem in Revit 2020, so I am following your advice to model with a 2" indent inside the family. Has it been confirmed this bug is killed in Revit 2022? I am not ready to upgrade my project yet, as I usually wait for a few hotfixes to catch any new bugs....

[barthbradley]

Yes, the "2-inch overlap" is still with us in 2022 - unfortunately. 

[RDAOU]

Hi @barthbradley Barth only tags me when he is in a good mood

@rauburn sorry for the late reply (got delayed with a few things)

Sorry for the delay...I wasn’t really on this topic but since Barth has tagged me again and all the muscle stretching is done and a solution has been accepted, here is my 2 cents  - just to clarify why the overhangs

You started the errand with a generic adaptive. Normally when you model a generic adaptive you the it straight to where it will be placed and hosted (the project) – no detours.

• Nested generic adaptive rarely (to never) function properly. they tend to misbehave and get disoriented if double nested - better if you never do that
• Taking it straight to the model and using repeaters, such form and dimension 16'x13' with such zigzag shaped surface would mean PATCHING. If double nesting a GAC works as a workaround and would cut and trim off the adaptive and fill in the gaps around the edges people working with adaptive component wouldn’t be choosing the hard way, would they?
• Why doesn't that work? BECAUSE if you cut of the adaptive component it means you are trimming its limbs (The adaptive placement points driving the panel geometry – in your SuperPanel.rfa) that's why every time you are finishing the sketch mode the adaptive component is restoring its original shape. (ie: it is adapting - what it basically is supposed to do - and it cannot do that with the 3rd and 4th point being trimmed)
• A similar thing happens when you try to place a 4pt adaptive but you escape after the 2nd point. The adaptive component still force place itself to have all 4 points in the model
• To test the above, I simply created a square shaped 4pts panel (very basic and simple) and you can see in the GIFF that the trimmed 4th and 3rd Limb always gets restored giving you the overhang effect. (So yes that is neither related to the 2” overhang outdated incident referred to above NOR your adaptive component SuperPanel.rfa regardless of how poor or good it is)

What is the solution? Simple

1. Use Generic Adaptive model and Patch the corners
2. Do exactly what you did in the Adaptive (lines pints and shapes) but straight in the pattern base

[JasonLLINDNER]

this is what we here call grinding. I wish I'm able to understand how this software works like that. By the way I learned about adaptive families from your videos on contribution site here. You had a link in your signature below but unfortunately it is not there anymore.

[RDAOU]

@JasonLLINDNER 

Thanks! I'm flattered... yes I took the links down a couple of years ago, downloaded what was worth keeping, deleted some and no longer use screencast.