The Perils of 3-Ply Plywood

Weak Earthquake Resistance of 3-Ply Plywood

Plywood made with only 3 plies tore in the 1994 Northridge Earthquake and should not be used

The Poor Earthquake Resistance of 3-Ply Plywood was proven in 1994 when shear walls made of 3-ply plywood tore exactly as shown in the photo to the right in the Northridge Earthquake.  For this reason, the City of Los Angeles downgraded the acceptable limits for 3-ply plywood to a maximum of 200 Lbs. per linear foot.

Footnote “n” of this table reads

 

This information comes out of the Los Angeles Building Code.

On page 10 of the Wood-Frame Subcommittee Findings Report published immediately after the Northridge Earthquake it says: “The performance of 3-ply construction has raised questions of its ultimate capacity.  Horizontal tearing has occurred on some outer face plies above the inner ply seam.  Values for all 3-ply panel construction were therefore reduced to 200lbs/ft maximum.”  In other words, the earthquake resistance of 3-ply plywood makes it too weak to consider.

This is what one of the field damage engineers with the Los Angeles Department of Building and Safety had to say.

My official civil service title was structural engineering associate. My unofficial title was training officer, one of six of us. I worked in the training and emergency management division, so we got to inspect all of the city’s problems, earthquakes and other disasters.”
 
“On more than one building, we saw horizontal tears in 3/8-inch plywood installed vertically.  
Some occurred at the interior joint of the outward plies; some occurred at notch points around openings, which were not at the inner ply joint.
I have heard some say they could tell the plywood itself was improperly manufactured by looking at the damaged panel after a major earthquake. I doubt that possibility since the earthquake forces were so large.  
2-inch-wide framing at joints frequently resulted in the plywood panel tearing for lack of proper edge distance. Studs also split due to too many nails closely spaced. This was a common occurrence. (My note: this is why the joint and sill framing requirements changed to 3x in the 1997 UBC)
Many engineers and inspectors saw the damage, but it mostly became apparent only after the wall coverings were removed during the repair or forensic processes. This was a more limited group of observers. 
Based on my field observations of damaged buildings, I believe that all current and past testing protocols do not adequately model the large dynamic forces imparted to wood shear walls during a major earthquake. 
My takeaways are that 3ply plywood and 3/8-inch plywood are low-strength materials.
There is no substitute for seeing what an earthquake does to a building. This is more important to properly design and construction than all the books and tests done to date.
Another engineer from Southern California who lived during the Northridge Earthquake had this to say.

“So, a little more history.

Part of the history includes that 3/8” plywood was typically made of 3 plies, and ½” structural 1 plywood was traditionally 5 plies.  Over time, to decrease the cost of making ½” plywood the number of ply’s was reduced to 4 ply’s and then 3 plies’ (now have less glue), but individual plies were now thicker to maintain the overall ½” thickness requirement. Yes, you could still get 5 ply struct 1 plywood, but it was more expensive than ordering ½” plywood with 4 or 3 ply’s and there were long lead times to get it as the mills had moved away from making the 5 ply sheets in ½” thickness.

After Northridge, the engineering community discovered that much of the ½” plywood being installed was just 3 plies.  Engineers had called for struct 1 plywood, but didn’t always say the number of ply’s required, as they had typically understood that ½” structural 1 plywood with exterior glue was typically made of 5 ply’s and didn’t realize the mills had moved away from manufacturing 5 ply ½” thick sheets.  After Northridge, 4 ply ½” structural 1 became the minimum requirement if you wanted to use the diaphragm and shear wall table design values.  If you had ½” 3 ply sheets, they became limited to 200 lbs. / ft.  You would also need the 4-ply minimum as well for ½” thick exposure 1 sheathing.

So, 3/8” sheets and ½” sheets made from 3 ply’s were given the same design value = 200 lbs./ft as their construction was basically the same- just 3 plies’.  Some of the issue was overdriven nails and gaps in the individual ply’s, such that once you drive the nail head through the first ply, you only have 2 ply’s left to resist the seismic forces.  If you over drive just one ply in 5 ply ½” thick sheets, you still have 4 ply’s left.  Having more glue in the 5 ply sheets makes it harder to overdrive the nails as well.   There was also the discussion of what is an overdriven nail versus properly installed nail:  Properly installed discussion depended on who you spoke with:

  1. Bottom surface of nail head rests on plywood surface or
  2. Top of nail head is flush with surface of plywood, and no deeper.

Each ply for 3 ply ½” thick sheets is approximately 3/16” thick.   2x 3/16” = 6/16” = 3/8” plywood.  So, the ½” plywood with 3 ply’s was treated as having similar behavior to 3/8” plywood, which may have overdriven nails as well.”

Thickness of 3-plywood makes no difference

This report applies to ALL thicknesses of plywood.  The poor resistance of 3-ply plywood does not change if it is 3/8, 1/2, or 5/8″ plywood.