Retrofit Strategies to Avoid
On this webpage we are showing real life retrofit strategies we have 
    encountered all over the Bay Area. These 18 examples show common
     —yet inappropriate— procedures that should never be part of a retrofit.

    The italicized comments are those of Thor Matteson, Structural Engineer
    (www.shearwalls.com). He reviewed these ineffective retrofit.as a service
    to the public.

    Some of the photos have been kindly provided by American Society of Home 
    Inspectors members Paul Rude and David Venable. They are conscientious 
    Berkeley based private home inspectors who are concerned about the 
    quality of the retrofit work they see every day. You can contact them at 
    Paul@SummerInspect.com and david.venable@sbcglobal.net

Example 1 - The illustration below shows a carpenter bolting a diagonal wood post to the floor framing.

"This sort of installation concentrates earthquake forces into the diagonal wood post.  It is possible that the post will be stronger than the floor framing it is attached to.  If the floor framing at the top end of the post breaks, or the connection fails, the post could punch up through the floor during an earthquake. If you install dozens of these, they might be acceptable. It is much more effective to connect the floor to the foundation by installing shear walls around the perimeter of the house."

Example 2 - This vertical piece of angle-iron has been bolted into the foundation and to the floor framing.

"During an earthquake the angle iron would put the floor framing into a condition known as 'cross grain bending.'

 

 

 

 


 

Wood has very little strength to resist cross grain bending, so little, in fact, that the building code does not even publish strength values for wood stressed in this manner.  In addition to probable damage to the wood, the steel angle will act as a lever to break apart the concrete. While there are cases that angle-irons could be effective, in the majority of houses it is impossible to use them cost efficiently."

Example 3 - This photo shows two metal straps connecting the beam, post, and concrete pier.  It is difficult from the photograph to see there are two straps here. The strap at the top is T-shaped; the vertical stem of the "T" is extended dowward with an additional straight strap to the concrete foundation pier.

"This installation is ineffective for several reasons: First, the 'T' strap is not installed in such a way that it can resist earthquake forces acting from side to side—the post will act as a lever and bend the strap. Second, if the straps are intended to keep the support post and beam from lifting up, the concrete pier that the lower strap connects to is not heavy enough offer much anchorage (uplift is not a serious concern anyway, because during earthquakes houses mostly sway from side to side rather than jumping up and down). Third, the slant in the bottom strap would tend to pull the post sideways if the post did actually try to lift up."

Example 4 - Each of these three straps is connected with a single bolt to the foundation.

"These straps are meant to join two pieces of wood together end-to-end.  As they are installed here, they can only resist forces acting upward. When an earthquake moves the floor from side to side, the straps will simply pivot around the bolts in the concrete."

Example 5 - "The manufacturer of these 'twist straps' does not intend for them to resist side-to-side forces generated by earthquakes. As they are installed here, they can only resist the floor trying to lift up off of the foundation. Side-to-side movement is the major cause of damage to buildings during earthquakes, and this installation is essentially worthless in preventing such damage."


Example 6 - This is a Simpson PHD hold down. This will keep the post from lifting up off the mudsill and foundation, but earthquakes do not cause this to happen. The purpose of hold downs is to keep plywood shear walls from tipping over. The diagonal braces on this post will perform in a manner similar to Example 1 at the top of this page.

 

 

Example 7 - This installation is also made of two straps and is similar to Example 3, only the "T" strap is turned sideways because a block obstructed its installation.

"This installation is worse than the previous one for two reasons: The "T" strap is wrinkled, which will allow it to kink or bend more easily.  Also, the manufacturer gives no load rating for the sideways installation."

Example 8 - The nails coming through the plywood from the other side were supposed to be driven into blocks of wood that sit on and are attached to the mudsill with nails.  When seen from the other side, this installation would look like it had been nailed into blocks. That is the side a building inspector would see.

"There appear to be about five or six nails in each space between studs.  Assuming that there is one nail into each stud, then this shear wall only has about one-sixth of its intended strength."

Example 9 - The "mudsill" is the wood member that bolts to the concrete foundation and supports almost all the weight of the house. Plywood shear walls must connect to the mudsill.  The "Nailed Blocking Method" is used almost exclusively throughout the Bay Area. This is an untested method, and no one knows how well it will work.  Other methods of making this connection are much more effective.  Below is an illustration of how the Nailed Blocking Method is constructed.

2"x4" blocks are nailed into the mudsill between studs. The nails often split the blocks. (See Examples 10 and 11) Bolts penetrate through the blocks and mudsill.

 

   Plywood is then       nailed into the         blocks and this         attaches the              plywood to the        mudsill.

Example 10 - The plywood on the other side of this wall is nailed to the short block. Short blocks tend to split very easily—when this happens, the strength of a block is severely compromised. Plywood on many retrofit shear walls covers blocks that look just like this one. This is why, when the Nailed Blocking Method is used, it is very important for building departments to inspect the framing before the plywood is installed.

"Not only do the nails that connect the block to the mudsill tend to split the blocks, but the nails driven from the plywood into the blocks will also tend to split them.  In most cases, once the sheet of plywood is nailed up, the blocks are completely concealed, and it is impossible to know if they have split."

/Example 11 - This is a longer block that has split. Often contactors switch to much thinner nails to prevent the block from splitting.  Unless color-coded, it is impossible to distinguish nail size from just the head of the nail.

"Using the correct quantity and size of nails is essential in providing the strength needed to resist seismic forces.  The nails in this block are also installed so close to the ends that they will likely split out the ends of the block during an earthquake."


Example 12 - This photo is a close-up showing the bottom edge of a piece of plywood that is supposed to be part of a shear wall.  This piece of plywood was nailed to the wall studs, without installing any blocks at all.

"Shear walls must have rows of nails along each edge of plywood. This installation will have only a small fraction of the strength of a properly constructed shear wall."

 

Example 13 - This is a Simpson "FSA" strap. The table, inset from the manufacturer's catalog, shows that the strap does not resist earthquake forces but only uplift forces (the arrow points to a dash where a value would be if this strap could resist lateral forces generated by earthquakes). The dash here means this hardware has zero resistance to earthquake forces.  These straps were bent into an "L" shape before installation

"The manufacturer intends for these straps to be installed straight, not bent as these were. With the bend in these straps they would not even resist uplift forces until the house had lifted up about two inches and the slack due to the bend in the strap was taken up."

Example 14 - This is the same piece of hardware as shown in Example 13, just not bent into an "L". 

"As installed, this hardware does not have any capacity to resist side-to-side movement, which is what causes most earthquake damage.  Additionally, the installation has almost no strength to resist uplift, because the bolt connecting it to the concrete footing is so close to the top of the footing that it could easily break out of the concrete."

 Example 15 - A few years ago the City of Berkeley recommended bracing cripple walls with "plumber's tape." This retrofit method is also seen in other parts of the East Bay Area. Plumbers tape is flimsy metal strapping used to suspend pipes underneath floor framing or attach water heaters to a wall—it is not designed to use in this way and will do practically nothing to resist earthquakes. A plywood shear wall should have been built here.

"Since those early recommendations, several manufacturers have developed many connectors specifically intended to resist earthquake forces. Thankfully we now have much more to choose from than plumber's tape. Even 30 years ago, a plywood shear wall could have been built here as a much more effective installation."


Example 16 - This is a Simpson "H3," which is primarily intended to resist wind uplift from hurricanes, not side-to-side forces generated by earthquakes.  It has minimal ability to resist such earthquake forces. It is also supposed to be installed with nails, not screws—and the hole at the bottom left is missing a connector all together.

"The manufacturer of this clip rates it as resisting 455 pounds of uplift, but only 125 pounds side-to-side force.  And those ratings are only valid if proper connectors (nails) are used, and if all the connectors are installed.  A different connector would be much more cost-efficient and better for resisting earthquake forces."

 
The Simpson "H10" and Simpson "L90", both shown here, can resist approximately four and six times, respectively, the amount of side-to-side force as the Simpson "H3". With modern tools it is just as fast and easy to install these far more effective connectors.

Example 17 - Note the gap between the steel plate and the wood mudsill.  In this situation the bolts will bend and eventually be torn out of the mudsill. To keep the upper bolts from bending, a continuous piece of lumber or plywood needs to fill the gap and be attached to the mudsill with nails or screws that have the same total strength as the lag bolts.  The lag bolts then need to be driven through the metal plate and into the piece of wood. When this is done, nails in the piece of lumber or plywood keep this piece of lumber from sliding on the mudsill. The lag bolts keep the piece of lumber or plywood from sliding along the inner edge of the plate.

 

 

Example 18 - When the mudsill slides on the foundation, as in an earthquake, the metal plate would spin around the single bolt at the bottom of the plate like a propeller. This rotation will cause the bolts to push up and break the mudsill. 

It is hoped that these examples of poor retrofit work impress upon you the need to check the work (that means YOU should crawl under your house) after a contractor has finished your retrofit but BEFORE you pay him.

To top of page

BuiltByNOF

Bay Area Retrofit
Residential and Commercial Seismic Retrofits
Structural Evaluations and Hazard Analysis
Howard Cook: 510-418-1676
Office and Fax: 510-548-1111
Email Howard Cook at
bayarearetrofit@aol.com
E-mail Jeff Bailey at jeffatretrofit@aol.com
This website and the articles within are copyrighted by Howard Cook
updated: September 13, 2008