jcaffee

Elliptical/Bell gable end roof

Recommended Posts

I'm trying to model an elliptical gable end roof (see picture) with little success.  Please help.

 

The attached plan has some of my efforts (plan and elevation) and a blank scene if you want to try.  I've tried single, double, and triple plane iterations, but it's beyond me.

 

jon

post-52-0-75080400-1440404356_thumb.png

bell roof 02.plan

Share this post


Link to post
Share on other sites

Jon,  I took your 3 lane arch model and made the middle roof an arch,  (you had it flat).  I think this approach is a step closer to  what you want.  

 

I do not have time to truly figure it out and I am wondering how easy it would be to replicate the profile you want.  I would think you could do a cad line following the curve of your roof,  and then break that ellipse/oval into 3 segment and from there maybe you can model the three roof planed it requires to build what you want.

 

Nice challenge,  if I have time later I will take a stab at it.'

 

The attached model is only me messing around with your 3 plane arch where I made the middle roof arched.

 

 

 

 

bell roof 02 scott 1.plan

Share this post


Link to post
Share on other sites

Scott, I took your idea and extrapolated a 5 plane--no go.  I just can't get the ellipse shape, just arches.

 

jon

Share this post


Link to post
Share on other sites

Scott, I took your idea and extrapolated a 5 plane--no go.  I just can't get the ellipse shape, just arches.

 

jon

If you understand an ellipse,   it is simple 4 arches with different read....  I will see if I have time later

Share this post


Link to post
Share on other sites

Scott is right, 3 arched roof planes can do it. The middle one flat but smoothly connected to the left an right wing which have relatively less radius of curvature compared to the middle one.

Share this post


Link to post
Share on other sites

Thought this was interesting but remember I don't do rooves so can't finish it.

This is a little closer, I just have no idea how to clean it up.

Also a pdf with your picture and where I got the arcs to start- ended up having to change those but that may be pilot error:)

bell roof 3 arc.zip

Roof pic.pdf

post-85-0-50372100-1440427377_thumb.jpg

Share this post


Link to post
Share on other sites

Here is my frustration.  There are 4 variables,  

 

1.   ANGLE AT EAVE

2.   ANGLE AT RIDGE

3.  RADIUS AT ROOF SURFACE

4.   FACET ANGLE  (I IGNORE THIS,  I DO NOT KNOW HOW TO USE IT)

 

It seems like I should be able to define two of the first 3 variables and the  other variable will auto correct.  It does not work.  Why can't I define the ANGLE AT EAVE and the ANGEL AT RIDGE and the RADIUS AT ROOF SURFACE auto corrects?   If I could do this I would make Jon Caffee very happy.

 

Okay  CA,  how would you do what Jon Caffee wants.  It should be relatively simple.

 

Sorry Jon,  I can't do it.  But I would think that 3 curved roof planes will get you  what you want if we could control the first 3 variables.

Share this post


Link to post
Share on other sites

Thought this was interesting but remember I don't do rooves so can't finish it.

This is a little closer, I just have no idea how to clean it up.

Also a pdf with your picture and where I got the arcs to start- ended up having to change those but that may be pilot error:)

Marc,  I cannot read your file.  Did you close the file before you attached?

Share this post


Link to post
Share on other sites

OOPS. I got nowhere with the 4 variables Scott notes- made some progress changing the pitch to a 45 degree angle, can't figure out how to lose that little corner sticking down. (or much else :)

bell roof 3 arc.zip

Share this post


Link to post
Share on other sites

this is a little better-the problem I can't figure out is-to make an approximate ellipse out of arc of circle-

first the arcs need to be tangent to each other

BUT

the arcs are not quadrants of a circle.

The top arc is not a quadrant but  I can't find a way to make the side arcs less than a quadrant. Up to you roof guys.

bell roof 4.zip

Share this post


Link to post
Share on other sites

I went back to the pdf and measured the angle for the side arcs to get an approximate pitch.

This isn't bad.

 

On another note I see some of you uploading plan files without zipping them- I can't get that to work (using firefox) is there a setting somethwere?

 

Edited, did one more adjustment- close, depends on how accurate you want it but this one is smooth (#6)

post-85-0-98232900-1440429868_thumb.jpg

bell roof 6.plan

bell roof 6.zip

Roof with overlay.pdf

Share this post


Link to post
Share on other sites

I went back to the pdf and measured the angle for the side arcs to get an approximate pitch.

This isn't bad.

 

On another note I see some of you uploading plan files without zipping them- I can't get that to work (using firefox) is there a setting somethwere?

 

Edited, did one more adjustment- close, depends on how accurate you want it but this one is smooth (#6)

Hey Mark,  darn good effort on your part.  I wish you would do a vid on your process. 

 

Did you get my complaint that I could not set the eave and fascia angles?  Nice job.

Share this post


Link to post
Share on other sites

Yes Scott, I got nowhere with the eaves, only made progress when I changed the pitch to an angle on the side arcs.

Don't have any vid capture so can't-

 

Put photo into Bluebeam PDF-drew in circles and sized until they looked about right.

Added lines so I could measure distances and convert arc sizes to full scale.

Calibrated the measure tool in Bluebeam to match the width I pulled off the Chief elevation then measured as needed, worked out the radius of the arcs based on scale.

 

In Chief

changed the arcs to match radius I pulled from the pdf.

Changed the width of the center arc to match the pdf (selected line-lock center,change length for both front and back.

Same for the side arcs.

Fooled with the pitch. At first I hit upon 45 degrees and stopped. Later I realized I could change the pitch so measured the angle of the chord of the arc in the PDF and used that as a starting point.

Moved things up and down with transform replicate till things matched. (they jumped around every time I changed something -remember I don't know how to do rooves:)

Noodged top arc radius and width, noodge up and down...like that.

 

Finally shot a cad detail from an elevation in CA- export as picture- placed that in the pdf file. Increased transparency and sized till it looked close just to check it all.

Share this post


Link to post
Share on other sites

Scott is right, 3 arched roof planes can do it. The middle one flat but smoothly connected to the left an right wing which have relatively less radius of curvature compared to the middle one.

OK nice mark, actually I wasn't around PC but exactly that was was what I meant. And one important thing to know here is, the ridge angle of the wings must be equal to the eave angle of the middle one. Also included the same roof done by 4 roof planes. The only difference is the pitched of the two roof planes in the middle will have a pitch equal to 1/2 x thier eave angle( the ridge angle of the wings). The cross check is also included in the plan for the formula.

post-2517-0-66918300-1440434168_thumb.jpg

post-2517-0-88600200-1440434185_thumb.jpg

ell roof 1.plan

Share this post


Link to post
Share on other sites

Here it is done as an "old school" geometric construction.

the roofs are set to the angle of the construction, and then curved to the constructed radius.

This is done with auto roofs, so it's real easy, precise, and predictable.

post-62-0-97085400-1440435502_thumb.jpg

post-62-0-06445900-1440435530_thumb.jpg

post-62-0-18307800-1440435547_thumb.jpg

eliptical roof.plan

  • Upvote 1

Share this post


Link to post
Share on other sites

Here it is done as an "old school" geometric construction.

the roofs are set to the angle of the construction, and then curved to the constructed radius.

This is done with auto roofs, so it's real easy, precise, and predictable.

Very good,  however two questions....

 

1:  if you wanted,  could you build manually?

2: Why don't you think the pitch of fascia of upper roof matches the pitch of ridge of the lower roof?    Don't you think these should be equal.  

 

But props for what you did.

Share this post


Link to post
Share on other sites

Here it is done as an "old school" geometric construction.

the roofs are set to the angle of the construction, and then curved to the constructed radius.

This is done with auto roofs, so it's real easy, precise, and predictable.

Actually,  if you turn on auto roofs,  you will get a gambrel roof which is not what Jon wanted.  So maybe you started with a gambrel roof,  but you then turned of auto roof build and made the roofs curved.  

 

Maybe you said that,  but it was not clear to me,  you did do some manual editing.  Maybe the manual editing is responsible for the lower ridge angle not matching the upper fascia angle.

 

Again props,  better than anybody else has done.

Share this post


Link to post
Share on other sites

Hi Scot,

Yes, these could be built manually, but it's easier to let Chief build it.

Yes, you get a gambrel.

The gambrel is based on the geometric construction of the ellipse. The construction gives me all of the information I need; the height of the roof change, the pitches, and the radius of each roof. I just plug this info into the DBX.

Share this post


Link to post
Share on other sites

Hi Scot,

Yes, these could be built manually, but it's easier to let Chief build it.

Yes, you get a gambrel.

The gambrel is based on the geometric construction of the ellipse. The construction gives me all of the information I need; the height of the roof change, the pitches, and the radius of each roof. I just plug this info into the DBX.

Thanks Bill,  I think I can almost do what you did....  but....  I can't figure out the geometric structure of the ellipse.  I am looking it up on line and I have not quite figured it out. I do not know how you did it.  I remember about 42 years ago,  on a wednesday,  my drafting instructor taught me the geometry,  but I think I was busy dodging spitballs to fully grasp what he was talking about.

 

Any hints?

Share this post


Link to post
Share on other sites

I think the point I want to make is that I can do this easily without guess work.

The construction is only an approximation of an ellipse, but it works well here, because we can only approximate and ellipse with chief roofs and walls.

This is a basic construction that I used to teach in first semester drafting when I was an instructor at City College in San Diego. I can copy it out of my old text book, and post it.

  • Upvote 1

Share this post


Link to post
Share on other sites

Bill has a winner!  The four plane roof with adjusted "Radius to Roof Surface".  Thanks Brudda!

 

jon

post-52-0-53421900-1440439498_thumb.png

Share this post


Link to post
Share on other sites

Please sign in to comment

You will be able to leave a comment after signing in



Sign In Now

  • Member Statistics

    27478
    Total Members
    6254
    Most Online
    Oilersfan20
    Newest Member
    Oilersfan20
    Joined