Chief developers fiddled with the roof spec, and it's broken (and I reported it to tech support, who sent it up to development)

[GeneDavis]

Maybe it's been this way since X zero, but I cannot say because all I have right now is 13 and 14.

 

First of all, it's not broken.  It just a little hinky when you mix pitches in a plan and want a common fascia height.  If you've the time to look and explore open the attached file and follow along.

 

A simple L shaped plan with hipped roof has six walls (by definition as it's an L shaped footprint) and six roof planes.  All roof pitches are 12/12 except for one of the inside-the-corner planes, which is specified at the wall as 5/12.  Overhang is 18" by default except that because I have "same roof height at exterior walls" checked in roof height section, the irregular roof of 5/12 pitch will run out further in auto roof build.

 

Build the roof, and check roof spec for all six planes.  For whatever reason, Chief "reports" in the roof spec dialog that the fascia heights of the two in-the-corner roofs, one at 5/12, the other at 12/12, have a fascia height lower than the other four planes, the distance is 3/16".

 

Here is the strange behavior.  One of those two roofs, the 5/12, when you look in section view or 3D, does have its fascia top lower, but the other one, the 12/12, has its fascia RAISED from the others.

 

Fine, you say.  I can manually edit to get every single fascia top to match height.  But you won't succeed.  Fiddle with it and see.  I put an image here of the plan view, with the 5/12 plane selected.  The others are 12/12.

Fascia heights with multi pitch roof.plan

[robdyck]

So Chief builds the roof "wrong"...that doesn't mean we need to. We aren't building rockets here! The "problem" is simply understanding that Chief is placing the finished fascia directly under the finished roof surface. Obviously to a computer, this is simply a rise/run calculation. Chief's sheathing and shingles don't 'flex' up or down like they might in the real world.

Because we don't model a "flex" in materials, we can simply adjust the dimension of the finished fascia board. A carpenter might do that in the real world, or they might just nail them on outside of the sub-fascia and sheathing and let the shingles and drip edge flex a bit to deal with the overwhelming discrepancy of 3/16". For those noticing such issues from street view, May I suggest using drip edge and gutters.
 

Come to think of it... Chief models rafter tails and truss tails incorrectly as well. Chief should be modelling them all out of alignment by +/- 1/2" in every axis. After all we want it to reflect real world conditions. Now try modelling your fascia!

 

It may be time to let this topic rest.

 

[GeneDavis]

Thanks, Rob, but my view is from the accuracy-in-3D side, which is what this program should be all about.  I can build it to precision in Sketchup.  It's simple 3D with intersecting planes.  Chief ought to be able to do it right.

 

This model took me 16 minutes to do, and I'm slow with SU.  I don't think it's too much to ask of Chief that a perimeter roof line with multiple pitches springing from it be built accurately.  The fascia is, in Chieftalk, a molding, a 1x8, dead level all the way around.  The planes represent the top surface of sheathing, which is the way Chief builds, and that surface line extends down to intersect the outboard top edge of fascia.  It's all just simple trig, if you are writing the code for the 3D build.

[mthd97]

Just a quick troubleshooting question, has the birds mouth height changed on the lower pitched roof segment or is it the same ?

 

I am guessing that you have already taken a cross section camera of each pitch type and zoomed into the connection point to see if anything might be going wrong ?

 

 

Edited November 10, 2022 by mthd97

[robdyck]

10 hours ago, GeneDavis said:

I can build it to precision in Sketchup

Do you have all the various layers accounted for in that SU model? Keep in mind that roof layers vary in vertical height when the pitch changes.

If you build the same roof in Chief and eliminate all the various layers, you will get the same result as in SU. All elevations will nicely match.

Remove ridge caps and remove shingles. Change osb to 1/16" thick. Uncheck Eave Sub Fascia. All fascia heights will now match.

[robdyck]

Before chasing one's tail too vigorously, one should have a plan for what to do if they catch it!

 

[GeneDavis]

Heh.  My corgi doesn't have that problem.

 

In my example of an irregular roof, all 12 pitch with one plane at 5, roof default calling for "same roof height at exterior walls," the math for the determination of fascia height comes from these variables, all set in the roof defaults:  pitch (12), overhang (18 for this plan), structure (9.25 here), surface (here at .625 but the .500 sheathing layer is used in the calcs), and plate height ("ceiling" in floor default = 109.125.)

 

And yes, thicknesses of roof assembly parts come into play in the various math solutions that come into play when Chief parametrically builds.  In this instance, we are using the sheathing.  Chief does not call it that in its spec dialog.  It allows a multi layer buildup.  You could have shingles, a membrane, rigid insulation, furring, a radiant barrier, whatever, but down at the bottom the layer is what Chief uses for thickness.  I think of it as the sheathing layer.  That layer thickness is also used in the determination of where Chief builds the subfascia.  

 

But where to place the fascia.  It's the extra credit problem in your second semester 9th grade math final.  You can solve it graphically with your CAD program, or your construction calculator.  Or you can write some code.  Whatever.  The fascia height is calculated (show us your work), and then . . . 

 

. . . it's time to calculate the position of the "wild" overhang of that bastard roof, the one with the 5-pitch.  Again, you have all the variables.  This is the one Chief gets wrong.  Don't know why, but Chief places the fascia a little too far outboard (too much overhang), and this results in a fascia height for this roof plane that is lower than that of adjoining planes.

 

Two plan files attached.  One called EDITED, the other called CHIEF.  In the CHIEF file, wrong fascia placement, you can see in the section how a little CAD work determines where the fascia should be.  A precise measurement is taken using a CAD line.  In EDITED, the bastard roof (sorry, that's what I call irregular planes) has its overhang edited using the measurement taken in the other file.  This brings its fascia height to the correct level.  The hips and valleys are manually moved to join (Chief won't join them for some reason.)

 

So there's a fix, but it should not be required.  There's a bug and I await Tech Support's reply.

 

 

EDITED.plan CHIEF.plan

[mthd97]

My device had trouble downloading those attachments GD ? a PDF is usually good for most.

 

Yeah even down under we understand the “B” word for that type of roof or valley. The carpenters usually say “That’s because it’s a B to build” lol.

 

I your case it’s a “B” to change the fascia alignment etc.

 

Hope they fix it up for you.

[Alaskan_Son]

@GeneDavis, there's actually more going on than meets the eye in that plan (and possibly other similar plans).  It goes beyond improper calculation--in fact, it may not be a calculation issue at all but rather some other bug.  Check this out... 

If you adjust the Fascia Top Height of RP-4 to match RP-5 then all your fascia will align.  If however you open up RP-3 you'll see that it now has a different fascia height from what it was.  Changes to RP-4 are affecting RP-3.  Super weird for sure.

[GeneDavis]

@Alaskan_Son

 

Thanks, Michael!  Glad a pro like you stepped in here and took a look at it.  Your post lends credibility to this, which is a discussion of a bug report.  I mistakenly hit the solution button, so you are credited with a SOLVE.

 

If you are still interested, take a back cut section through so you can view RP-3 on L and RP-1 on R.  See the attached snip.

 

What is a real mystery, is how while when opened for spec, these planes report differing fascia heights, RP-3 being lower, a CAD line drawn through the RP-1 fascia top intersects the fascia for RP-3 at precisely the top!  No diff!

 

Since having reported a ticket with Chief, I've no response yet, but will report when I get one.

[VHampton]

Thank you for starting the thread Gene.  

 

When I've attempted to make varying roof planes merge at a common roof edge height, the fascia will often behave as you are describing.

 

Meaning that roof edges won't always line up, even when the top of fascia input is a perfect match. The gutters don't always line up as well. 

 

A 3D molding (for fascias and gutters) has been my work-around for many years.

 

That said, Robert appears to have described the optimal method. Thank you. ...It's great to learn more about making Chief work as it should.

 

 

 

[mthd97]

On 11/13/2022 at 12:49 AM, GeneDavis said:

@Alaskan_Son

 

Thanks, Michael!  Glad a pro like you stepped in here and took a look at it.  Your post lends credibility to this, which is a discussion of a bug report.  I mistakenly hit the solution button, so you are credited with a SOLVE.

 

If you are still interested, take a back cut section through so you can view RP-3 on L and RP-1 on R.  See the attached snip.

 

What is a real mystery, is how while when opened for spec, these planes report differing fascia heights, RP-3 being lower, a CAD line drawn through the RP-1 fascia top intersects the fascia for RP-3 at precisely the top!  No diff!

 

Since having reported a ticket with Chief, I've no response yet, but will report when I get one.

Thanks, I can clearly see the discrepancies.

[StephenGreene]

On 11/9/2022 at 3:47 PM, GeneDavis said:

Maybe it's been this way since X zero, but I cannot say because all I have right now is 13 and 14.

 

First of all, it's not broken.  It just a little hinky when you mix pitches in a plan and want a common fascia height.  If you've the time to look and explore open the attached file and follow along.

 

A simple L shaped plan with hipped roof has six walls (by definition as it's an L shaped footprint) and six roof planes.  All roof pitches are 12/12 except for one of the inside-the-corner planes, which is specified at the wall as 5/12.  Overhang is 18" by default except that because I have "same roof height at exterior walls" checked in roof height section, the irregular roof of 5/12 pitch will run out further in auto roof build.

 

Build the roof, and check roof spec for all six planes.  For whatever reason, Chief "reports" in the roof spec dialog that the fascia heights of the two in-the-corner roofs, one at 5/12, the other at 12/12, have a fascia height lower than the other four planes, the distance is 3/16".

 

Here is the strange behavior.  One of those two roofs, the 5/12, when you look in section view or 3D, does have its fascia top lower, but the other one, the 12/12, has its fascia RAISED from the others.

 

Fine, you say.  I can manually edit to get every single fascia top to match height.  But you won't succeed.  Fiddle with it and see.  I put an image here of the plan view, with the 5/12 plane selected.  The others are 12/12.

Fascia heights with multi pitch roof.plan 3.93 MB · 13 downloads

I couldn't get them to match with the longer overhang on the 5/12. However, I did get them all to match up perfectly by moving the 5/12 overhang back to 18", like the rest of the roof. Very strange. 

Gene's Roof.plan