Need a book? Engineering books recommendations...

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

Re: SIPS panels - Structural Insulated Panel System

[Subject Prev][Subject Next][Thread Prev][Thread Next]

First of all, a "disclaimer" (to be upfront)...I do part-time work for a
SIP manufacturer (who also does timber framing).  Thus, my knowledge of

Most SIPs are actually made with OSB skins rather than
plywood (at least that is what the "major" SIP manufacturer's use).  SIPs
can be in widths up to 8 ft and lengths of up to 24 ft (thus, the reason
for using can get bigger sheets easier).  FYI, there are two
basic types of SIPs...1) OSB (or plywood) skins glued to pre-formed EPS
foam and 2) OSB (or plywood) skins adhered to cast-in-place polyurethane
(I believe) foam.  The SIP panels will have a fair amount of strength both
in bending (i.e. roof or floor loading) or in axial/load-bearing values
(i.e. load-bearing walls) without splines.  Splines are used to connect
the panels together.  Splines come in what I call non-structural (i.e.
plywood or OSB surface splines) or structural (i.e. 2x material, LVLs, or
wood I-joists).

For load-bearing, panels can take a decent amount of load without any
structural splines (i.e. just use plywood or OSB surface splines) if
applied in a uniform manner (i.e. roof panels bearing on wall panels).
Panels can take some point loads, but for "significant" points loads you
will need a panel joint with 2x material or LVL "posts" in the wall.
Structural splines don't really increase the SIPs load bearing capacity
all that much for uniform load will get more capacity for
such situations by decreasing the floor to floor height or going with a
thicker panel.

For bending (i.e. roof or floor panels), structural splines can help quite
a bit, especially if you decrease the width of the panels to 4 ft or 2 ft
(i.e. more lumber in the roof...makes it more like a stick-framed
roof...but also increases the cost).

For racking loads (i.e. shearwall loads), all current panel ICC/ICBO
evaluation reports are only for wind or IBC seismic categories A, B, or C.
There is new acceptance criteria from ICC that will allow testing of SIPs
for ICC SDC's D, E, and F, but I don't believe anyone has an finalized
report done per that new acceptance criteria yet (I know that Premier has
done the testing, but I don't know if they have the finalized report
through ICC-ES yet).  Since you are in SDC B, any values you find in
manufacturer's ES reports will be useable for wind or seismic.

As both Eli and Joe pointed out, the major SIP manufacturers will have
ICC/ICBO ES reports that will give you allowable loads for bending,
load-bearing, and racking.  They will have their "basic" details in the ES
reports as well.  All those values are determined per an ICC/ICBO
evaluation criteria.  The three of the major SIP manufactuers are:

Insulspan (which I work for) ICC/ICBO report NER-520
Premier   ICC/ICBO report NER-633
R-Control   ICC/ICBO report ESR-1006

You can obtain the ICC/ICBO evaluation reports from the manufacturer's
websites typically or from (do a search on the manufacturer
or report number).

Now, as to your specific situation...

First, be careful with Hardi-plank.  If you don't want to void the
Hardi-plank warranty, take a good close look at the Hardi-plank
installation manual.  If you want to do "blind" nailing/screwing of the
Hardi-plank (i.e. the nail/screw heads are hidden under the overlapping
planks), then I believe that they require firring strips (2x2 if I recall
correctly) at some certain interval (something like 16 in OC or so).  If
you want to attach directly to the SIP skin, then you will have exposed
nail/screw heads that you will then need to caulk/seal.  Take a CLOSE look
at the Hardi-plank install manual or call them.  I do know that one of
their competitors (CertainTeed WeatherBoards Fiber Cement Lap Siding) does
have memo that does allow for direct attachment to the panel skin with
blind screwing.  The only problem is that their memo only talks about
polyurethane foam SIPs...should be fine for EPS foam SIPs (which is what
all three manufacturers listed above use...I believe)...I never got an
answer from their "Product Compliance Engineer" that issued the memo
(ironically, this "engineer" is actually a licensed architect).

As to the interior of the roof panels, you will likely want to still
install gyp board.  You seem to imply that it will be left rough.  I would
think an architect would not like this...most panels companies use a panel
seal tape on the underside of the roof panel joints.  Won't look so great.
Plus, you will have no fire rating if you don't have any gyp.  Most
manufacturers have details/requirements to achieve 1 hour fire-rating in
their ICC/ICBO report (requires gyp board).

As to your questions:

1)  What is your support structure that supports the panels?  Are you just
going from load-bearing wall to load-bearing wall?  Or will you have some
roof framing members that will support the roof panels on the interior?

Roof panels will be attached to either load-bearing walls or supporting
wood members (timbers or glulams) with either panels spikes (really long
nails) or panel screws (really long screws).  The number of screws/spikes
will be largely be dependent on the diaphram capacity that you need.  The
panels themselves will not likely buckle in a "global" manner (i.e. the
overall panel buckle).  The thing to be careful of is to make sure the
panel joints are properly done, otherwise you can get some
swelling/buckling at the panel seams.  This usually occurs due to moisture
building up inside the structure and trying to migrate out through the
panel seams and then causing the OSB at the joints to swell.  The chances
of this are minimized by properly foaming the joint (usually with an
expanding foam...we use a two-part foam that chemically cures rather than
the "Home Depot" type in the little cans which is a moisture based curing
foam) and applying panel seal tape to the INTERIOR of the panel seam.  It
is also necessary to make sure that there are proper mechanical systems to
handle the humidity (SIP structures are VERY air-tight if done properly
and thus can "trap" humidity from humans breathing) and carbon monoxide
(i.e. air quality).  This will mean at a minimum an air exchange
unit...A/C is usually helpful too.  You will also want to make sure that
any roof shingles that you use (assuming that you are using shingles) are
"approved" by the shingle manufacturer for SIP use (i.e. non-vented roof

2)  Yes, the skins are basically taking the load...assuming a uniformly
applied load.  The skins can take some load even if it is a point load.
Basically, in a global sense, you are just resisting a buckling load and
both skins are tied together with the foam to form an "I shape", so to
speak (the foam being the web).  So, in the case of a uniformly applied
load, it is largely the buckling capacity of the overall panel that will
govern the capacity.  You certainly do have the local crushing of the
skins to be careful of...mainly in point load situations, but also for
higher uniform loading.  SIP load bearing walls can handle point loads
from stick-framed roofing (i.e. 2x roof joists) in most cases...this is
handled typically by having a "bearing plate" (i.e. a 2x piece that is the
same width as the panel that sits in the edges of the skins and is
attached to the top let-in, which is a 2x piece that is inset into the
panel BETWEEN the skins).  It is only when the point loads get large that
you need to start putting in posts in the panels (at panels seams).

For your warm fuzzy feeling, get the ICBO/ICC reports.

3) And what is so different from nailing a SIP panel to a bolted down sill
plate than nailing plywood sheathing for an exterior wall for a
stick-framed shear wall?  It is the same "crap shoot".

If you are concern from a load-bearing point of view, the nailing of the
skins is mainly to hold it place somewhat like anchor bolts on gravity
column.  The load is really being "taken" through end bearing of the
skins.  FWIW, the typical detail is the sill plate is bolted down to the
foundation and is at least as wide as the overall panel thickness.  Then a
bottom let-in is nailed to the sill plate.  The let-in (2x material) is
what the panel is nailed to with the skins bearing on the sill plate (or
the floor decking).

As far as "relying on the insulation to do a whole lot of bracing of the
two, thin load bearing wythes," keep in mind that you only need about 2%
of the applied load to "brace" things typically.  Thus, the you don't need
a whole lot of resistance from the foam to cause the two skins to act
together.  Thus, the panel works well as a unit...much better than the
individual skin would.

If you have some specific questions, feel free to contact me offlist.  If
we happen to be supplying the panels on the project, then we are more than
able to help with design issues on the project (that is what they pay me
the "big buck" for)...and it is usually part of our scope of work to some
degree.  If there is not a manufacturer on board, then I can put you in
touch with one of our dealers/salespeople who can provide some assistance
(with the hope of being able to bid the job).  And regardless, I am
willing to help as much as I can (whether I am on the clock or not).


Adrian, MI

On Tue, 7 Dec 2004, John C. Jones wrote:

> I have a project that's a private high school dormitory.  The Architect is very interested in using products that can fulfill multiple tasks.  He has suggested using a SIPS panel as both the roof decking and the load bearing wall.  A SIPS panel is basically 2 pieces of plywood separated by insulation.  The panels are 48" wide and at each joint have actual studs or joist members.  The project is in central Alabama and is IBC Seismic Cat B.  Seismic will be relatively light for this project.  Wind will likely control.  The Architect wants this to be his structural deck (14' span or so), finish ceiling (yes, it's rough), insulation, and shingle nailer.  The walls would be the same except it would receive sheet rock on the interior and hardi-plank on the exterior.
> My main concerns are:
> 1.  With a conventional plywood roof on H-clips there is movement available in the roof.  This product doesn't have this option.  How is this accommodated for this system so that temperature doesn't cause visible roof bulges?
> 2.  What issues are there with load bearing?  It appears that the plywood is actually taking the load.  It really doesn't give me a warm fuzzy feeling.  What happens if there is a leak that causes plywood failure?  It doesn't seem like it would take much termite infestation to cause a problem.  Bulkier studs are more resistant to both of these.
> 3.  Bolting a plate down and then installing the panel and then nailing through the panel into the pre bolted sill plate just seems like a crap shoot.
> Has anyone out there used this product and can offer advice?  My gut feel is that it's probably pretty good as a roof deck/insulation/ceiling, but I'm not as comfortable with it as my load bearing wall.  It just seems that you're relying on the insulation to do a whole lot of bracing of the two, thin load bearing wythes.
> John C. Jones, PE
> Barnett Associates
> Pell City, AL
> 205-884-5334
> 205-884-0099 (fax)
> ******* ****** ******* ******** ******* ******* ******* ***
> *   Read list FAQ at:
> *
> *   This email was sent to you via Structural Engineers
> *   Association of Southern California (SEAOSC) server. To
> *   subscribe (no fee) or UnSubscribe, please go to:
> *
> *
> *
> *   Questions to seaint-ad(--nospam--at) 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:
> ******* ****** ****** ****** ******* ****** ****** ********

******* ****** ******* ******** ******* ******* ******* ***
*   Read list FAQ at:
*   This email was sent to you via Structural Engineers 
*   Association of Southern California (SEAOSC) server. To 
*   subscribe (no fee) or UnSubscribe, please go to:
*   Questions to seaint-ad(--nospam--at) 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: 
******* ****** ****** ****** ******* ****** ****** ********