Need a book? Engineering books recommendations...

Return to index: [Subject] [Thread] [Date] [Author]

RE: Exposed Beam Roof Design

[Subject Prev][Subject Next][Thread Prev][Thread Next]
Chuck,
It still comes down to how the shank of the connector will perform (bending)
and any displacement (laterally) that will toss the heavy clay tile off the
roof. 
7-1/2" is not bad, but we need at least a 2x10 for BATT Insulation here in
the Mojave Desert for control of the energy calculations. Beyond this, we
also need a way to get the HVAC through, over or around the Heavy Timbers
unless there is enough room to bring it through the sides of the 38-foot
room. Going above, while not meeting the thin edge look that the Architect
created is the best choice for running electrical and mechanical through the
framing. 

Best Regards,
Dennis

-----Original Message-----
From: Chuck Ritter [mailto:riter(--nospam--at)jar.com] 
Sent: Friday, June 23, 2006 8:17 AM
To: 'seaint(--nospam--at)seaint.org'
Subject: RE: Exposed Beam Roof Design

Dennis

I built a (very-pseudo) Japanese country style house and have a thick roof 
above the exposed timbers. The roof is 12" full dimension rafters with 2x 
T&G decking over them and then SIPs (1/2" CDX both sides and 7 1/2" EPS 
sandwiched) --> roof deck is 10" thick. Looks fine to me, but I'm not your 
architect customer. If you want an apparent slim roof, what about adding 
some purely cosmetic laminations -on site - to glulams outside the line of 
the walls of the house.  You could then have a continous layer of rigid 
insulation over the rafters plus some more between them in the 
heated/cooled footprint. Re; Scott's observation on rafter heat conduction 
- with a heavy frost on the roof, I know exactly where the rafters are w/o 
looking for the tails. It does telegraph through - although it could be 
affected by incompletely foamed joints and/or those whopping big spikes for 
the SIPs as much as the rafter themselves.

regards

Chuck Ritter
JAR Associates, Inc
401-294-4589
401-294-3826 fax


-----Original Message-----
From:	Scott Maxwell [SMTP:smaxwell(--nospam--at)engin.umich.edu]
Sent:	Thursday, June 22, 2006 10:08 PM
To:	seaint(--nospam--at)seaint.org
Subject:	RE: Exposed Beam Roof Design

Dennis:

Two thoughts...

First, you raise the issue of the EPS foam in the core of the panel being
in shear due to the seismic mass of the roofing material (heavy clay tiles
in your case) being on the top layer of OSB but the shear resistance of
the SIP roof panels will be in the lower OSB layer.  Thus, for that
seismic mass/inertia load to reach the lower OSB layer, it will have to
trasmit down through the core of the SIP panel.  Very true.

I would offer, however, that many SIP applications have lumber splines
at the panel joints in addition to the foam in the core.  Thus, you will
have lumber at panel joints that will be much like stick framed
rafters/purlins, but likely at greater spacing than one would see in a
stick framed roof.  The point is that more than likely this splines would
server as "collectors" that then get the shear load from the tiles & top
OSB layer down to the lower OSB layer which is serving as the diaphragm.
The only reason why this would not happen would be if there was a desire
to use OSB surface splines, in which case there would not be lumber for
the full depth.

Now, I will admit that I have no testing data to support the notion that
SIPs with lumber splines at some spacing would be able to adequately
transmit high shear loads from something like a clay tile roof down the
lower OSB layer.  I can say that compression of the EPS foam due to the
weight of the clay tiles is NOT an issue.  SIPs have been and will be used
in areas where there are snow loads that would make the weight of clay
tiles seem small.  Typical EPS foam (i.e. that used for SIPs) has a
compressive strength (defined at 10% deformation, I believe) on the order
of 5 psi or greater.  5 psi is still 720 psf.  So, clay tiles will not
even come close to compressing the EPS foam in a SIP to any noticable
level.

The second item was the issue of the thickness of the roof.  This is an
issue, I would agree.  If the desire is to see a "thin" roof edge with
large timbers supporting it, then SIPs may not be the answer.  I would
agrue, however, that to some degree you are kind of outlining a rather
difficult situation.  You seem to want a thin roof edge with large timber
supporting the roof that are visible AND you want a rather high thermal
resistance (i.e. R value).  Now, depending on how you define large timbers
and what type of R value that R39 is (i.e. is it the insulation's R value
or the whole wall's R value), you may have a rather difficult time
achieving the desired result.  In order to achieve a "full wall" (full
roof in this case) R value of R39, then with SIPs you are talking at least
a 10" SIP panel likely without lumber splines (or at least minimum lumber
splines).  If you get too many lumber splines then you may be looking at
12" nominal roof SIPs.  Thus, in order to get a "whole wall" R value of 39
with stick framing and batt insulation, you are likely talking something
deeper than 12".  And in order to get your large timber beam look, you
would likely be talking very deep glulams or solid lumber going through
that whole depth plus some below the lower level of the "enclosed space"
of the insulated thickness...and such large pieces of wood would likely
lower the "whole wall" R value (at least I would think so based upon my
VERY limited understanding of thermal properties...my thermo class is WAY
in my past and not much remains in the dusty recesses of my mind).

Just some thoughts based upon musing after reading your response.

Regards,

Scott
Adrian, MI


On Wed, 21 Jun 2006, Dennis Wish wrote:

> Scott,
> I went off the digest mode because there are some topics that I wish to
> discuss and want to be able to receive replies quickly - the reason I am
> responding so quickly.
> I have concerns about the use of what I described as a SIP panel even 
though
> the lower plywood sheathing would be what transfers shear while the 
sandwich
> was intended to support and distribute the weight of the heavy clay tiles
> without causing damage or too much compression to the rigid insulation.
>
> The real problem with this is not so much in shear transfer as it is in
> appearance. The lines at the end of the roof should be such that the tile
> appears to sit above the heavy exposed beams and any framing (as I did
> suggest) would raise the exposed perimeter of the roofline and give the
> appearance that I had multiple roofs upon each other. I don't think this
> would play well to most architects.
>
> Years ago I tried using rigid insulation from below the structural 
diaphragm
> when I had smaller beams (4x) spaced at say 32" on center and the client
> wanted to give the appearance that the roof was T&G. What I did here was 
to
> create a modified panelized roof with the rigid insulation below the
> structural sheathing and secure T&G from below that would hide the
> insulation and not increase the visual thickness from the exterior of the
> home.
>
> The trouble with this is that it becomes difficult to work electrical and
> mechanical into the room and the labor becomes very costly. This is the
> reason I was looking for a more affordable solution.
>
> I'm not knocking SIPS - I've seen the details for long screw 
installations,
> but the lateral movement between layers of plywood inducing bending in 
the
> screws places the tile at risk during seismic (cyclic) motion and can 
create
> a hazard to anyone on the ground being a target for tile. While I 
understand
> that SIP's are glued and laminated, I am not aware of any testing for
> eccentricity or deformation of the skins due to the weight of the tile
> above. I would not have the same fear in lightweight roof coverings, but
> this would also change the mode of lateral control to wind (most likely).
>
> Best Regards,
> Dennis
> -----Original Message-----
> From: Scott Maxwell [mailto:smaxwell(--nospam--at)engin.umich.edu]
> Sent: Wednesday, June 21, 2006 6:48 AM
> To: seaint(--nospam--at)seaint.org
> Subject: Re: Exposed Beam Roof Design
>
> Dennis:
>
> While not intending to be a "company hack" just pushing a product, I 
would
> be remiss if not saying that SIPs would be ideal for such a situation.
>
> The only difficulty is that you don't have a ton of information as to how
> well SIP panel screws or panel spikes (big nails) will deal with the
> seismic diaphragm loads in a "desireable" way (i.e. ductile response
> similar to nails through structural sheathing in typical diaphragms).
>
> But, in theory, they should work.  You just may (or may not) have a tough
> time gettings code official to allow their use in high seismic zones.
>
> Regards,
>
> Scott
> Adrian, MI
>
>
> On Wed, 21 Jun 2006, Dennis Wish wrote:
>
> > I am looking for details showing how other engineers design typical
> exposed
> > beam ceilings / roofs with shear transfer. I have done a variety of
> > different designs over the years, but the cost of labor is beginning to
> make
> > me think more seriously if I am providing the best design. Here are 
some
> > specifics:
> >
> > Zone - California Mojave Desert with summer climates that can reach 130
> > degrees at an average of between 4% and 10% relative humidity and the
> > minimum roof insulation requirement of R38 according to the last time I
> > checked.
> >
> > In the past I would construct the roof in he following manner:
> > Exposed beams 6x material at 4-feet on center
> > 2x10 roof joists perpendicular and hung from the beams with Simpson
> hangers
> > Roof sheathing above (typically 5-ply 15/32" minimum OSB or Plywood)
> > unblocked
> > Clay tile roof where slope exceeds 2-1/2: 12 or 3:12
> > 5/8" or ?" gypsum ceiling at bottom of 2x10 joists (depth of 2x10's
> > determined for uncompressed R38 batt insulation
> >
> > Alternative:
> > 2x Joists parallel to exposed beams with 2x3 or 2x4 nailer secured to 
side
> > of exposed beams for nailing or securing of gypsum ceiling.
> >
> > Full depth blocking at double top plate set by flat cut (depth greatest 
at
> > exterior of bearing wall) beams and ripped 2x continuous blocking w/ 
A35F
> or
> > new flat clip - 2 per block at 4' lengths or 2 per block where spacing
> does
> > not exceed 24" on center and joist rather than exposed beams are used.
> >
> > 2x6 exterior DF Kiln Dried studs with double top plate not cut for 
exposed
> > beams. Birds cut or Corbel cut ends of exposed beam overhang with 1x 
T&G
> at
> > exterior  eave for southwest look. Boundary nailing of structural
> diaphragm
> > to solid blocking based on analysis or minimum 8D common nails @ 6" 
o.c. -
> > 10D common @ 6" o.c. if blocking is 3x or wider.
> >
> > Lately I have been reconsidering the cut and flush nailed roof rafters
> > because of cost (shrinkage no a factor where KD lumber is used). If I
> stack
> > the roof 2x above the exposed beams, I can save on the beam cost but 
will
> > need to block the rafters above the beams. The alternative may be to 
use a
> > T&G ceiling with a rigid insulation above, but I am not crazy about the
> idea
> > of using either a sandwiched OSB sheathing with a Rigid Insulation
> (similar
> > to an SIP panel).
> >
> > The latest design I am considering, the client will have a vaulted 
living
> > room ceiling with a second floor gallery for his legal library that 
over
> > looks the great room below. This means stacking a 38-foot support for 
the
> > gallery (open one side) and the use  installation of a heavy timber 
Ridge
> > Beam with exposed beams flush cut. The backside needs to change slope 
to
> > accommodate the 18 or 19- foot plate height to accommodate the book 
cases
> > flush to the wall and this can be flush framed to the ridge beam so the
> roof
> > diaphragm sheathing is flush to the top of beams and rafters. It does 
pose
> a
> > problem on the lower sloped backside if the lumber rafters are stacked
> above
> > the exposed beams.
> >
> > I would like some ideas and possibly if you wouldn't mind sharing some
> > details to show the means of detailing these types of ceilings, I would
> > appreciate the help.
> >
> > Finally, I am also considering the use of manufactured lumber for the
> studs
> > in order to minimize the labor necessary to shim the interior radius of
> the
> > crowned studs for a plumb and flush fit of interior drywall with a skim
> coat
> > of plaster. For the high wall (on the exterior side viewed from the
> gallery,
> > the stud wall will be platform framed since an exterior patio cover at
> least
> > 12-feet deep will brace the wall at 9-feet from out of plane buckling. 
At
> a
> > 38-foot clear span through the living room (minus the framing for the
> stairs
> > to access the second floor octagonal tower, library (gallery) and 
exterior
> > balcony, the designer has provided sufficient locations of interior 
walls
> on
> > the first floor to stack shear. The ends of the ridge beams will be
> > sufficiently solid to allow for shear transfer down to foundations and 
the
> > depth of the room will be such that the aspect ratio of this roof
> diaphragm
> > will not exceed a 2 to 1 aspect ratio. Shear transfer is pretty 
straight
> > forward, but some creativeness will be required for the octagon shaped
> tower
> > with exposed beam ceiling so as to flush frame the floor joists for
> > performance grade to minimum bounce and to provide for a sub-floor
> necessary
> > to create a 1-inch step from interior to balcony with a ?" per foot 
slope
> > for drainage at the balcony.
> >
> > I hope the explanation is clear. I am seeking some typical shear 
transfer
> > and framing details to see how others have address the ceiling framing 
to
> > accommodate a minimum of R38 roof insulation (batt or rigid) so that 
the
> > material and labor costs provide the best solution. Also I can use some
> > thought to the use of manufactured lumber in the exterior platform 
framed
> > walls to see if the cost saving in labor to straighten the wall 
justifies
> > the use of manufactured lumber which generally saves waste and does not
> need
> > as much labor and shimming.
> >
> > I'm working in AutoCAD and can read any DWG, DWF or DXF format. You are
> > welcome to send a PDF file as well and can attach directly to this list
> > replies as long as you purge your details to keep the size of file
> > considerate for those with slower Internet connections.  If the list
> refuses
> > the drawing format, zip the file but do not use EXE or self-expanding
> > formats.
> >
> > If responding from the SEAINT list, do not post attachments. Please 
send
> > directly to me at dennis.wish(--nospam--at)verizon.net .
> >
> > Thanks in advance for your assistance on this question.
> >
> > Dennis S. Wish, PE
> > California Professional Engineer
> > C-41250 Exp. March 31, 2007
> > Structural Engineering Consultant.
> >
> >
> >
> > ******* ****** ******* ******** ******* ******* ******* ***
> > *   Read list FAQ at: http://www.seaint.org/list_FAQ.asp
> > *
> > *   This email was sent to you via Structural Engineers
> > *   Association of Southern California (SEAOSC) server. To
> > *   subscribe (no fee) or UnSubscribe, please go to:
> > *
> > *   http://www.seaint.org/sealist1.asp
> > *
> > *   Questions to seaint-ad(--nospam--at)seaint.org. Remember, any email you
> > *   send to the list is public domain and may be re-posted
> > *   without your permission. Make sure you visit our web
> > *   site at: http://www.seaint.org
> > ******* ****** ****** ****** ******* ****** ****** ********
> >
> >
>
> ******* ****** ******* ******** ******* ******* ******* ***
> *   Read list FAQ at: http://www.seaint.org/list_FAQ.asp
> *
> *   This email was sent to you via Structural Engineers
> *   Association of Southern California (SEAOSC) server. To
> *   subscribe (no fee) or UnSubscribe, please go to:
> *
> *   http://www.seaint.org/sealist1.asp
> *
> *   Questions to seaint-ad(--nospam--at)seaint.org. Remember, any email you
> *   send to the list is public domain and may be re-posted
> *   without your permission. Make sure you visit our web
> *   site at: http://www.seaint.org
> ******* ****** ****** ****** ******* ****** ****** ********
>
>
> ******* ****** ******* ******** ******* ******* ******* ***
> *   Read list FAQ at: http://www.seaint.org/list_FAQ.asp
> *
> *   This email was sent to you via Structural Engineers
> *   Association of Southern California (SEAOSC) server. To
> *   subscribe (no fee) or UnSubscribe, please go to:
> *
> *   http://www.seaint.org/sealist1.asp
> *
> *   Questions to seaint-ad(--nospam--at)seaint.org. Remember, any email you
> *   send to the list is public domain and may be re-posted
> *   without your permission. Make sure you visit our web
> *   site at: http://www.seaint.org
> ******* ****** ****** ****** ******* ****** ****** ********
>
>

******* ****** ******* ******** ******* ******* ******* ***
*   Read list FAQ at: http://www.seaint.org/list_FAQ.asp
*
*   This email was sent to you via Structural Engineers
*   Association of Southern California (SEAOSC) server. To
*   subscribe (no fee) or UnSubscribe, please go to:
*
*   http://www.seaint.org/sealist1.asp
*
*   Questions to seaint-ad(--nospam--at)seaint.org. Remember, any email you
*   send to the list is public domain and may be re-posted
*   without your permission. Make sure you visit our web
*   site at: http://www.seaint.org
******* ****** ****** ****** ******* ****** ****** ********



******* ****** ******* ******** ******* ******* ******* ***
*   Read list FAQ at: http://www.seaint.org/list_FAQ.asp
* 
*   This email was sent to you via Structural Engineers 
*   Association of Southern California (SEAOSC) server. To 
*   subscribe (no fee) or UnSubscribe, please go to:
*
*   http://www.seaint.org/sealist1.asp
*
*   Questions to seaint-ad(--nospam--at)seaint.org. Remember, any email you 
*   send to the list is public domain and may be re-posted 
*   without your permission. Make sure you visit our web 
*   site at: http://www.seaint.org 
******* ****** ****** ****** ******* ****** ****** ******** 


******* ****** ******* ******** ******* ******* ******* ***
*   Read list FAQ at: http://www.seaint.org/list_FAQ.asp
* 
*   This email was sent to you via Structural Engineers 
*   Association of Southern California (SEAOSC) server. To 
*   subscribe (no fee) or UnSubscribe, please go to:
*
*   http://www.seaint.org/sealist1.asp
*
*   Questions to seaint-ad(--nospam--at)seaint.org. Remember, any email you 
*   send to the list is public domain and may be re-posted 
*   without your permission. Make sure you visit our web 
*   site at: http://www.seaint.org 
******* ****** ****** ****** ******* ****** ****** ********