Development and Updates for the Medeek Wall Plugin

[Medeek]

Until someone sends me a model of a wall with steel studs, track and a couple of openings with headers etc... I will probably put this on the back burner for now. At this point I don't have enough information in order to properly add steel framing to the plugin. More study will be required.

[Medeek]

Version 0.9.9q - 12.03.2018
- Enabled "Over-the-Post" termination (single rise gooseneck and starting ease) for all handrail profiles.



The utilization of this termination option won't really become important until I setup open sided stairs with the accompanying newel posts and balusters.

Also note that the delta Z for the starting ease and also the gooseneck is plus 4" for now, at some point I will probably want to allow the user to customize this height differential(s), but in the US market 4" seems pretty much standard.

View model here:

https://3dwarehouse.sketchup.com/model/17a5e7b1-37fe-495e-b25c-e7c5006fa9a4/Handrail-Test-3

[Medeek]

I'm not even sure what to call this next tool that I am considering.

Perhaps the Blocking Tool would be the appropriate name.  Basically I need to have a tool that will allow the designer to insert various/miscellaneous elements throughout the model.  They may want these elements within a wall panel assembly (group) or outside of any group.  The use of this tool would be primarily for additional studs or blocking. 

The parameters would be:

Edit menu only:

Length:  models units
Rotation:  Degrees

Draw and Edit menus:

Size:  CUSTOM, 2x4, 2x6, 2x8 etc...

Depth:  in. or mm (grayed out when standard size selected above, available when custom size selected)
Width:  in. or mm (grayed out when standard size selected above, available when custom size selected)

Category:  BLOCK, STUD, JOIST, POST

Level:  1, 2, 3, 4, 5, 6  (this parameter will be implemented in the future for wall panels, stairs and beams as well)

Material:  LUMBER, LUMBERPT (custom materials from global settings as well)

Framing:  2D, 3D

If the elements are inserted within the wall panels, they must be retained during a regen, not a problem since I already have that issue worked out.

These elements will be fully parametric.

Are there any additional options or parameters that anyone else would like to see made available with this feature?

[Medeek]

This tool will create simple rectangular members only. Which means I will probably remove the POST category since stand alone posts will require a whole host of other features such as wrapping and associated base and cap hardware.

I've had some feedback that this feature is probably not needed at this time, but I've also have some comments specifically requesting it. Before I progress any further feel free to voice your opinion on the matter. This is really quite a small module and most of the code will be recycled from the beam and stair modules so I don't anticipate taking more than 48 hours to complete it however I don't want to invest time into a feature that will rarely be utilized.

Ultimately the plugin is for you, not for me.  I don't design houses anymore, I just design the software that designs the houses.  As such your opinion of what tools are made available carries more weight than mine does. 

[brilliantbydesign]

I think that it vital that the user can add in their own custom items to a wall assembly.  It is also vital that these items stay there on a regeneration of the wall.  Other plugins I have used reset on the Regeneration which is incredibly frustrating!!

There are definitely standard approaches to construction however it is often for the non-standard that 3D modelling is most useful.  I think the key is to think carefully about the balance between something being parametric whilst still flexible and fast.  I personally find it pretty frustrating when I am locked into a completely parametric system which makes modelling natively in SketchUp difficult or slow.  Someone who has good sketch-up skills will always want to be able to use those skills when something falls outside the normal parameters. I hope this is helpful. 

[Medeek]

I think the key issue with which I am trying to address with this new tool is best summed up when you compare SketchUp (SU) with Chief Architect (CA).  Both can be used to model a structure, one is fully parametric but is quite locked down and restrictive while the other is very free form, allowing the user to do as they please.  The downside to this freedom is that the program has no way of keeping track or making sense of all these custom changes and hence the parametric ability cannot natively exist.

CA does a nice job of keeping everything well contained but its 3D environment locks the user down too much in my opinion, and for the designer (who is not too different from an artist) who wants to express their creativity, I think this can be too restrictive.

Being able to insert "custom" geometry into the wall, roof and foundation assemblies, whilst categorizing and tracking it maintains the parametrics (and estimating) but also allows flexibility.  Being able to retain this custom geometry after a regen is critical to the success of this paradigm.

In a nutshell the plugin is trying to maintain the flexibility of SU while giving the user the parametrics of a program like CA.

[brilliantbydesign]

You said it far better than I could.....exactly my thoughts!!

[brilliantbydesign]

I think one thing that is perhaps forgotten in all this is that to model something in SketchUp properly using non-parametric tools the user must actually understand construction well.  With fully parametric programs the user is able to create a heap of geometry with very little knowledge of the actual real use and purpose of that geometry.  Although many find this frustrating when using SketchUp I think that ultimately it means that accuracy is increased and mistakes are avoided.  I am a builder by trade and when I draw in SketchUp I am literally imagining myself on-site building with my hands.  This means that the end product is a true digital representation of what will happen in reality.  I think this is the beauty of the free-form modelling - it requires the user to actually build step by step producing better results in reality!!

I also think another thing to consider is not just the ease of creating geometry but how this is then going to be communicated to third parties.  Drawing a model in SketchUp is one thing but presenting and communication that digitally and on paper is another.  Ultimately the model must be organised in such a way to allow for ease of documentation, customisation (as variations occur) and also ultimately quantification.  Builders always need to know what to build, how to build it and how much it will cost.

[Medeek]

Running the stair module through a few tests this evening, found one minor bug and was fairly pleased with the stair envelope for checking headroom height:



I've also added one additional termination (OTP with a 2 riser gooseneck) which is typically used where you go from one flight of stairs to the next as show below:



As long as the riser heights match (like they should) for each run of stairs then the 2 riser goosenecks matches up perfectly with the starting ease of the next run of stairs.  Of course the specified hand rail height for each run must match as well.

Note that the white color (handrail fittings) components are not being automatically generated by the plugin those were manually inserted, however the brown sections of handrail are automatically generated and they matched up perfectly as expected (Z height).  I left the fittings white so you can see what elements were required to be brought into the model.

These fittings will be included with the plugin in the library/handrail_fittings subfolder.  If I get ambitious I may have to actually model up some volutes for the bottom of the handrail but for now the list of supplied fittings (for the LJ6010 profile) is:

- S7011 (right handed)
- S7019
- S7020
- S7021 (right handed)

You'll also notice that in the top image I've created a landing with a 2x4 pony wall supporting it (sorry barely visible).  When you go to create walls like this it would be nice to have the plugin ignore any surrounding walls and basically treat these walls as completely isolated from the rest of the structure.

With that in mind I reworked the auto-corner configuration algorithm ever so slightly, so that it is now possible to place any number of wall panels within an over arching group.  What this does is effectively isolate these walls from any other groups within the model.

I will need to make some updates to the estimating module so that it is smart enough to look for groups in the root of the document with embedded walls.  I will also need to make a video demonstrating this technique, and when and where it would be useful.

Granted, I have not extensively tested out this new feature so I would say proceed with caution but my preliminary testing shows that it is quite effective and convenient when modeling sub-assemblies within a larger context.

[Medeek]

Version 0.9.9r - 12.05.2018
- Added a "2 riser gooseneck" Over-the-Post termination option for all handrail profiles.
- Adjusted the auto-corner configuration algorithm so that wall panels can be placed within larger groups within the root of the model (wall panel isolation).
- Fixed a bug in the Over-the-Post section of the handrail/stair module.

[Medeek]

Here is a quote from a review of Nick Sonder's book on Amazon that parallels some of my own thinking.  The reviewer first touched on his use of the book but his final remarks regarding SketchUp and design work really hits home:

This books has a lot of different tips when working with SketchUP. But I have to say, after going through many weeks thinking I was able to use this for production drawings I was sadly mistaken. By time I was able to get everything exported out to Layout, the drawings just did not look good at all. The resolution was way too grainy for my liking. The vector overlay was too bulky and raster was too pixelated. I am sure there is a way to fix this, however the next part made me re-think the whole process. Layout was so slow to regenerate the image. Each time I would pan it would pulse the screen and my workstation grade computer was just not able to do what Layout was requesting of it. My computer and workstation graphics card is not by any stretch old or limited. The thing runs all other software great. This was a huge disappointment...

All around if you are thinking about transition from your existing software to Sketchup Layout I would recommend some hesitation. However, if you want to learn a decent way of putting a Sketchup model together, I do recommend the author's techniques. I still use them for normal Sketchup use, I just cannot see investing in the time and patience with regards to Layout work. If someone is thinking of transitioning it might be better to look at an actual BIM software and if you are like me Chief Architect seems much more appropriate. I want to love this, because I think Sketchup is by far the most flexible software when it comes to heads up design. You are not stuck in dialog boxes, which for design flow and immersing yourself in the architecture SketchUP is great. I wish Trimble would take a few notes from a software like Chief Architect and simply some of their rules and plop them into Sketchup. I also think if Sketchup spent more time thinking how Architects might use Layout as full production drawings would be great, you really should not have to go through some of these crazy steps to get great drawings from SKUP. It's just not quite there, and this process does not make it that much smoother.

I'm not wanting to disparage Nick Sonder's work or his workflow, I think out of anyone his is some of the best.  My concern is with Layout and SketchUp itself as an architectural tool.  I feel like the plugins are starting to bridge the gap or parametric edge that Chief Architect has had over SketchUp (with a fair distance to go yet).  However, the other area we are falling down in is the creation of construction drawings and documents.

I am excited to start work on the automation of this piece of the puzzle and really dig into the Layout API, however at the same time I am genuinely concerned with Layout itself, and how well it functions as the 2D engine underneath the hood.  SketchUp's recent updates to Layout in the last two years/iterations have been tepid at best.  We really need a solid 2D drawing environment, something that can go toe-to-toe with AutoCAD.  I realize that this is a big ask, but it needs to happen.

I've already invested a considerable amount of my time into developing these plugins and I will continue to invest more.  I've been full time at it since April of this year and rather than work a salary man's job I chose to pursue this much more interesting path (we will see whether this was foolish on my part in the months to come). 

As I continue to develop these plugins I think it will not only benefit myself and the designers who use the software but also SketchUp itself as more architects, engineers, designers, contractors and draftsmen are able to utilize SketchUp as their primary design tool.  The work I do, as well as others like PlusSpec and John Brock to name a few, is helping put SketchUp on the map in the architectural design world.  We are helping pull users of other design software, such as Revit and Chief Architect, and converting them to SU.  We are trying to do our part.

It is now up to SketchUp to boost the Layout development and do their part.  We need a world class 2D engine.  Without it, we can't compete, it doesn't matter how good my 3D models are, they have to translate into construction documents, and it has to be seamless and effortless and a joy to use.  Unless Layout is brought up to par I will be forced to go alternative routes such as exporting floor plans to DXF etc...   This is really not the route I want to go but right now I am seriously considering it.

[brilliantbydesign]

I think this is very helpful Nathaniel.  The reality is Sketchup never set out to be a program for Architects or Designers.  It has just moved that way because it was such an intuitive excellent 3D drawing program to start with.  It is also much cheaper than alot of the software out there in the marketplace which gives it a certain appeal.  There is no doubt that a high-level user has to develop a good workflow to get around some of the inadequacies of the software and make it efficient.  I think in the long run, as developers add in functionality, SketchUp will become more and more competitive.  I think Trimble would be wise to invest some time into Layout as this is where most people come unstuck.

I think in Nick Sonders Case the nature of his work (high end, custom homes) means that SketchUp is the best option because alot of his building practices are not easily automated.  For someone just drawing up standard houses much more automated and specific software probably is better.  Personally, I love the open source, evolving nature of SketchUp and the fact that it is the underdog biting at the heals of the big boys.   

The other huge advantage is that the customer can look at the model using free software which will work on nearly any computer.....this has been huge for me in my business! 

I sometimes get pretty frustrated with Layout but I have found that over the years my techniques have developed to streamline things.  Michael Brightman has an excellent system which really helps at the layout end. The main issue I have layout is with the limitation of the dimensioning tools.  It will be awesome when this becomes automated and removes the risk of "snapping" to the wrong place.

One potential issue I see is that there are all these developers producing plugins which do specific things but don't necessarily work together.  The user usually still has to do some legwork in getting everything flowing.....this is certainly an advantage of the big players.   Excuse my musings and feel free to delete them if not applicable to this forumn.........I just love sketchup!!

[Medeek]

I'm adding two extra parameters which will allow an extension of the bottom of the stringer as shown in the detail below:



This detail was taken from literature for LVL stringers. Notice the use of framing anchors to help bear some of the vertical load.

I prefer to extend the landing back to catch the full bearing of the stringer but I guess there are some situations where additional headroom is needed or other configurations where this method of stringer support is optimal.

[Medeek]

First look at the stringer extend option with its associated notch:



You can also enable the thrust block with the extension/notch but I don't really see the point to doing that, it only further weakens the stringer at its point of bearing.

I'm thinking I should probably extend the side spacer down all the way until it meets the landing/notch, unless there is a good reason not to.

Also I've been reviewing all of the html menus and there are quite a few places where I am displaying
or requiring input in inches and it would be nice to also display the same dimension in ft-in. (fractional), I am looking at this now.

[Medeek]

Version 0.9.9s - 12.06.2018
- Added two parameters in the stair module to allow for extended stair stringers.
- Draw and Edit Wall menus now display wall height in feet and inches (fractional) when using an imperial units template.
- Added the action: "Regen Wall Assembly" in the context menu for all wall assemblies.

I've also gone ahead and enabled the feet-inches dimensions for other applicable dimensions within the global settings tabs (Walls, Door, Windows, Stairs).