Wednesday, July 6, 2011

2010 Wood Handbook

The Forest Products Laboratory (FPL) has recently released the printed edition of the 2010 Wood Handbook: Wood as an Engineering Material.  While you can download the entire book from the FPL website (see below), I encourage anyone serious about wood design to buy the printed edition as a desk reference.  At a cost of $60, this may be one of the least expensive engineering reference books you will find. 

If you are looking for a textbook on wood design, such as how to design beams, etc., this is not the book for you.  (I would recommend Design of Wood Structures by Breyer, Fridley, et.al.)  This book focuses on the physical and mechanical properties of wood and wood composites and provides information on wood preservation, fastening, finishing, and adhesives.  

This is taken from the FPL website:

The Wood Handbook is one of the most widely used documents in the wood literature. The 2010 edition is the first update of the 21st century and includes several new chapters. The original handbook was first published in 1935 by the USDA Forest Products Laboratory (FPL).
The Forest Products Society (FPS) is pleased to cooperate with the FPL to present this newly updated edition in hard copy. This 2010 edition is your one-stop source for information on wood as an engineering material and offers content such as:

• Properties of wood and wood-based products
• Design information for architects and engineers
• Wood and non-wood composites
• Wood-moisture relationships and wood durability

...plus NEW CONTENT on:
• Wood as an environmentally responsible, sustainable building material
• Heat-treating and sterilization procedures for wood infected by invasive insect species
• A special COLOR chapter on low-magnification micrographs of cross sections of commercial wood species
 

You can order your hard copy of the 2010 Wood Handbook here or you can download content or sample the material that is in the book here.

Tuesday, June 28, 2011

Deck Design Do's and Don't's: Part 1

I recently visited a wood framed observation platform that I am replacing with new designs.  The owner wants to keep a small section of the platform that is newer while replacing the older portion that is failing in some areas.  I will post some of the failure items over the next few weeks so as not to show everything at once.



For Part 1, we can discuss the importance of post caps when bearing a girder on a beam.  See the photograph above.  Reasons to use a post cap for this condition include:
  1. Increases lateral strength of the connection.
  2. Eliminates use of toe-nails, which when improperly installed can split the top of the post.
  3. Contains movement of the wood girders relative to the top of post that results from the shrinking, swelling and twisting of wood as it reaches its equilibrium. 
Ultimately, you do not want the condition in the photograph to result in the photograph below. 



For more information about deck design, please see my previous post here.

Friday, June 10, 2011

Residential Deck Design

In the last 6 weeks I have been researching and reading about deck and guardrail design and construction.  I am currently working on a project to develop two observation platforms for air monitoring equipment.  These platforms are stand-alone wood frames structures.  While they are not residential decks, they are very similar in many ways with the exception of design loads. 

For an introduction to deck design, I recommend Design for Code Acceptance-6: Prescriptive Residential Wood Deck Construction Guide published by the American Wood Council.  This guide covers design and detailing for several conditions including deck-ledger attachment, post-beam connections, and guard post attachments.

The guard post attachments are interesting because recent testing shows the inadequacy of many standard connections.  The International Residential Code requires that residential deck railings be designed for a live load of a 200lb concentrated load in any direction located at any point along the rail.  Upon reviewing local construction practices of guard-post attachments, many of which appeared questionable for meeting code requirements, researchers at Virginia Tech began testing these connections.  The results, which have been published in Wood Design Focus and the Journal of Light Construction, were somewhat surprising.  Typical bolted and lapped/bolted connections were unable to meet the code requirements. 

After attempts to redistribute load using typical connections were unsuccessful, the Researchers began testing Simpson HD2A anchors attached to deck joists.  This method of connection for guard-posts has subsequently been included in DCA-6.  While the 200lb concentrated load has been a code requirement for some time, until now there have been no prescriptive guidelines for meeting the code.  Further details for guard post attachment can be found here

Further Reading:



Friday, May 27, 2011

Repetitive Member Factor: An Introduction

An increase in the reliability of wood members when placed into assemblies with repetitive use components has been recognized for many years by the wood industry.  The term assembly is typically defined as 3 or more members spaced at 24” on center or less and attached with load distributing elements capable of supporting the design loads.  Examples include floors, walls, roofs and three-ply beams with load distributing elements of wood structural sheathing or transverse mechanical fasteners. 

Floors and walls are typically solid-sawn members attached with plywood or OSB sheathing.  Roofs, however, are solid-sawn members or prefabricated trusses that have the top chord and/or bottom chord connected with structural wood sheathing.  The sheathing creates load sharing between the members of the assembly.  Three-ply beams share load through the mechanical fastening system used to build the member. 

The capacity of the assembly benefits from an increase in lateral stability of the members, partial composite action where the load is distributed lengthwise along the member, and load sharing where the load is distributed across parallel members that have different stiffnesses.  The parallel members having different stiffnesses result in varying amounts of deflection under uniform loads.  The load distributing element reduces the effects of various deflections by distributing the load away from more flexible members to stiffer members.  Members that are more flexible tend to be weaker than stiffer members, resulting in load being distributed away from weaker members and into stronger members.

Load sharing is accounted for through the use of repetitive member factors.  Since 1968, the National Design Specification for Wood Construction (NDS) has specified a repetitive member factor of 1.15 for bending design values of dimension lumber.  Also, since 1970, ASTM D245 has recommended a load sharing factor of 1.15 for bending stresses in multiple member systems.

Tuesday, May 17, 2011

Lessons from a Tornado

The April 27 tornadoes left many opportunities to study structural performance of both commercial and residential buildings.  One area that I have observed was a residential development located just out of the path of an EF3 tornado that made its trail not far from my home.  With the exception of one building's collapse, damage to each home generally consisted of a portion of roof missing, as seen in the photo below. 


A close up inspection of the rafter heel condition at the top plate highlighted the failure of the roof system.  A birds-mouthed rafter with (3) toe-nails into the wall top plate is a common framing detail used in residential structures.  This detail was used throughout this development.  There appear to be two realistic modes of failure for this detail as highlighted in the pictures below.  The first mode is failure of the 2x rafter at the birds-mouth.  This failure mode occurred in a small portion of rafters witnessed. 


The second mode of failure is at the toe-nail connection.  Although this connection works reasonably well for lateral shear, the typical limit state for uplift loading is nail withdrawal from the wall top plate.  The nails in the picture below appear to have seen very little stress during failure.     


Athough this is a common detail, I avoid detailing a rafter/truss to top plate connection with toe-nails. A Simpson H2.5A anchor, available at local hardware stores, is inexpensive and provides a significantly improved connection capacity. This syle of clip results in a more consistent installation and brings a different connection failure mode that does not include nail withdrawal. For a small home with a simple framed roof, such as homes in this development, a small investment in H2.5A clips can have a major impact on structural performance.

Tuesday, May 3, 2011

Tuscaloosa, Alabama Tornado Meets Wood Truss

I pulled the picture below from The Birmingham Business Journal's coverage of the April 27, 2011 tornado damage.  I have been slow to look at online pictures and videos of the event for a myriad of reasons.  I lived in Tuscaloosa for several years while studying for my BS and MS.  Even though many people have seen the devestation via TV, Web, or first hand, I could not resist sharing this picture here on the blog. 



A close look at the stick of wood embedded in the bumper of this car, and the wood lying on the ground in front of the car, reveals that this "stick" is actually from a metal plate connected wood truss.   I do not know the context of this picture to understand how this may have happened or the location of adjacent buildings from where this truss may have lived.  Whatever the process, the end result is amazing.