Model the impossible hip

[GeneDavis]

I'd like to help a client built some difficult roof returns, by building full sized models that can be used at job time to cut and install framing parts.  It is a real 3D challenge.

 

A center gable is pitched at 11.25 and hipped wings at L and R are recessed back a foot or so.  Specified on the drawings is the radius roof overhang at the hipped roof edges, but nothing is called out for the returns that are shown in elevation.  I suspect the plans were done in Revit.

 

The section on the architect's section detail call out the curve for the flare as 108" radius with point of tangency where the 4.5 pitch roof meets the building line.  Nothing further is said about what to do for the gable returns, which in my opinion are the key signature architectural feature of the building.

 

Pics attached here show the Revit-drawn elevation of the return, where I am in Chief with it, and a micro-plan to focus just on the geometry.  The software goes nuts when I try to join these at hips and valleys.  Equal-equal is OK, but the program produces weird results for the irregulars.

 

As with returns of this type, fascia height is a constant.

 

I know that if the "hip" could be generated, it would not be linear in plan view.  It is a curve, and not a circular segment.  How these things get built is a mystery to me.  Maybe shipbuilders are better equipped mentally to handle it, because curve fairing is necessary, I think.

 

I know nothing about Chief solids, and all the boolean stuff that goes with it.  Maybe Chief solids can generate the curve and I can plot it, then get busy with the band saw.

 

 

Curved roof edge study.plan

[javatom]

I did not have time to look closely but I did notice something on the plan.  You have the roof planes set to degrees of pitch instead of the "in:12" setting"  Open the roof plans and change those settings.  The 11.25 pitch they are talking about will make sense then.  Your plan shows it in the other setting which would be the same as a 2 3/8:12.

[GeneDavis]

The degree setting came from me doing a workout in 2D CAD to examine points of tangency.   And you really need pitch in degrees when working with curved roofs if you want them tangent to the roofs they join.

[javatom]

Yea,  That makes sense.  My system must have imported your plan wrong.  It came in with those roof planes almost flat.

[javatom]

Gene,

I got something that looks close but has many strange things going on to get it to work.  Chief will not seam this together.  I did it by manually pulling edges together until the seam looked right.  The strange thing is that this had to be done while in plan view.  Moving the curved planes in 3d made it all explode.  I know this is not the eave edges you are looking for but maybe you can use some parts of it to get what you want.

javatom1119935922_Curvedroofedgestudy.plan

[Chopsaw]

Hi Gene,  If I am understanding what you are after.  The radius for the hip rafter of the two 108" R. curved roof planes is 207 11/16" R.

 

Not sure if there is a mathematical way to confirm that one but I just used the tools I know in Chief.

 

 

 

 

[Alaskan_Son]

21 hours ago, GeneDavis said:

I know that if the "hip" could be generated, it would not be linear in plan view.

 

This is not true.  If each of the 3 roof "planes" has the same arc shape, then the plan view would look completely normal (linear).