First tutorial video:
Engineering Video:
The SketchUp model used in the video can be found here:
Valley Truss Set:
Hip Roof Framing:
SketchUp model used in the video can be found here:
Testing the limits of the Medeek Truss Plugin with complex hip roofs. Note that the roof primitives have not been trimmed back. This is primarily a study to determine what additional programming would be required to generate this type of roof automatically:
This for me would be the holy grail of hip roof framing…
If you’re looking for another Grail after you’ve found this one;
- differing eaves heights
- dormer windows
- Half-gables
marvelous work so far
As you can see there is still a tremendous amount of work that can be done here. To that end I’ve started a KickStarter project which if successful would allow me to devote at least 4-5 months of my undivided attention on the programming of this plugin:
https://www.kickstarter.com/projects/128644708/medeek-truss-plugin
2 Likes
I have been giving some thought to dormers and how to specify them. Below is a dormer design I framed in Solidworks a few years back for a garage design:
Of course the numerous ways in which one could frame a dormer is probably beyond the scope of this discussion but I would like to consider what geometry needs to be determined in order to orient and size the dormer.
After some thought I think the following points, planes and lengths would define the geometry of a typical dormer:
1.) Main Roof Plane
2.) Attic Floor Plane
3.) Offset from exterior wall below (how far the dormer is out of plane from the ext. wall below).
4.) Pitch of Dormer roof
5.) Dormer width
6.) Dormer height (distance from attic floor plane to top plate of dormer wall)
7.) Window width, height, and header depth
8.) Dormer wall thickness (2x4 or 2x6)
9.) Dormer rafter depth
10.) Position of Dormer along length of building
Items 2,3 and 10 can be combined into a single point selection.
Pondering complex hip roofs this afternoon and considering the graphic below:
A few rules seem to emerge:
1.) An outside corner will always create a hip that is 45 deg. from each leg of the corner.
2.) An inside corner will likewise always create a valley 135 deg. from each leg of the corner.
3.) Where two hips or flying hips meet a ridge will extend 135 deg. from each hip.
4.) When two valleys meet at 90 deg. they will terminate and a ridge will extend at 45 deg. from each valley.
5.) With a building with walls only running north-south or east-west all ridge lines will always be north-south or east-west.
6.) Likewise all hips, valleys and flying hips will be oriented northwest, northeast, southwest, or southeast.
7.) When a valley meets a ridge, (they will always meet at 45 deg) a flying hip is generated that is 90 deg. from the valley and 135 deg from the ridge.
8.) When two valleys meet at 180 deg. from each other, the result will be either to flying hips perp. to the valleys or the degenerate case of 4 valleys and 4 ridges.
9.) When a valley and a hip meet each other at 180 deg. then two ridges that are 45 deg from the valley will be the result (typical L shaped roof).
10.) When 4 hips meet the result is a pyramid.
There may be a few other degenerate cases I’m missing but I think that covers it.
Variation in plate heights, pitches, overhangs and even mixing hip and gable (half hip, dutch gable) further complicate the matter. To begin with I need a algorithm to generate the roof planes, then the framing just falls out from there. To create the roof planes I need a fairly robust straight skeleton implementation.
Not sure if SketchUp will allow it but why not try and advertise your crowdfunding project on their facebook page to build up some momentum? 322,000 like might count towards something 
. I would really love to see this become a worthwhile venture for you.
SketchUp was already gracious enough to feature me in their last newsletter so I don’t know if I want to press my luck by requesting more feature time on their Facebook page. Unfortunately, I did not have the Kickstarter campaign advertised when the feature hit so I missed the momentum I could have had, my bad.
Regardless if I get funded or not this venture has already proven wildly worthwhile and rewarding for me on a very personal level. Its opened my eyes to SketchUp and what can be done with it. Prior to this I never really gave SketchUp serious notice or thought of it as a design platform that could do “real” work.
I’ve enjoyed the time spent plugging away the code and my understanding of all truss and roof types has increased significantly. Its taken me a while to wrap my head around the API, but the more I use it the more its grown on me. I applaud the SketchUp programmers and powers that be that had the foresight to make this interface available to the “average Joe” like myself, and the ability to create something that never existed before.
I will continue to make progress on the plugin but a funded Kickstarter campaign would give me the ability to devote my undivided attention to the project for a good 5 months at least. Considering what I’ve been able to do in three months while juggling other jobs I think I could do some great things and really make this plugin shine.
Realistically though, I think this plugin does have a limited audience at this stage in its development. Not every architect or designer is interested in the nuts and bolts of roof framing, trusses or other engineering related items. There is a certain group of people that find the plugin very useful and are as excited about it as I am and then there is a large majority who don’t see the need. Which brings me to the next question, what should I add to make it more useful and applicable to “your design needs”.
One thing to keep in mind: your ponderings are based on boxes where every wall is at 90º to the next wall: Hard-coding the numbers might be a mistake for future developments where you might have (eg) a 45º wall or a shaped bay.
I understand what you’re saying. I do wonder if there is more demand, or could be more demand, than what’s apparent. As an example, I know of two companies working on similar projects for Revit, although quite a bit more developed than yours currently is. They basically concentrate on timber and steel construction.
http://www.aga-cad.com/products/packages/wood-framing
In my personal opinion, I think the problem is that SketchUp isn’t/wasn’t all the well regarded for it’s ability to satisfy the needs of professional architects and construction engineers. However, I feel the wind is changing direction. SketchUp just lacks the professional add-ons to create the demand. Our company looked into Revit, but it just seemed overly hard and complicated for what we wanted to achieve. Sketchup was quick and easy but lacks “off-the-shelf” add-ons to make it viable, for us at least.
Version 1.2.6 - 01.18.2016
- Sill plate option (advanced) enabled for top and bottom bearing floor trusses.
Work on the manual is progressing however for those interested the red colored boards in the image are called “ribbon boards”. This is fairly typical for floor trusses. The notch purposely left in the truss to accommodate the ribbon board is called the ribbon cut or ribbon notch. Continuous ribbons provide stability for installed trusses, and also provide a solid nailing surface for the edge nailing of floor sheathing. This eliminates the need for larger and more expensive “rimboard” solutions required by dimensional lumber and other engineered wood products. 2x4 lumber is common, but any dimension of 2x lumber can be used for the ribbon board.
Similar to the complex hip roofs I need to program the floor truss/TJI module so that it can automatically frame out any non-rectangular floor plan. It seems like my goals keep getting more lofty and ambitious as I dig deeper into this, at some point I should reach the end of the rabbit hole.
Version 1.2.6 - 01.19.2016
- Added Solid Sawn Floor Joists (metric and imperial)
- Sill plate and Sheathing options (advanced) enabled for TJI and Solid Sawn floor joists.
In a related note but slightly unrelated I’ve decided to take a small break from directly programming of the Truss Plugin and apply some effort to the integrated Truss Designer. I’ve had a number of requests by various parties and individuals to add in plywood gusset plates to the engineering of the Truss Designer, in lieu of standard metal plates (Mitek, Simpson, ITW etc…). This will allow for the design of site fabricated wood gusset connected trusses, something still found in remote agricultural locations. I’m thinking a couple weeks and I will have this programming task complete, at least for the common fink truss, and then put it out there to gauge the response. This new feature will be available from the Truss Plugin when fink trusses are analyzed.
Version 1.2.7 - 01.26.2016
- Added advanced roof options for hip rafter roof (sheathing, fascia).
Couldn’t resist adding in these options for hip roof since it was mostly there already from the gable roof module.
I’ve uploaded a test L-Shaped structure with a hip roof:
I’ve created both hip roof primitives with the plugin, that was the easy part.
I then placed a valley rafter with its centerline (top) inline with both roof planes. I think I’ve got it right. What I am unsure of however is the best way to terminate the framing at the intersection of the valley, lower ridge and flying hip. Once I have a handle on how a carpenter would actually construct that junction I think I can proceed to start work on a secondary roof module for both hip and gable rafter roofs.
upload of DC hip roof
you can switch the hidden joints to see a correct view of the joint, but I use to put a 38x38 solder in lieu of a fancy joint, thus all hip and valley top cuts were the same for convenience
in perspective view, you wil note the center line follows the rafter top edge, rafters then meet the hip edge and set up higher than the valley so center lines meet
roof - broken hip.skp (1.8 MB)