Nels is giving sound advice, listen up because the tile is a different
animal than red brick.
The duct tape he mentions is the way to go, and I would add that you embed
the bolt through as many face shells as possible. The bulb of epoxy formed
on the back side of each face shell will act like a washer to spread the
tension load. If you can get away with exterior plates, use them. And if you
do, use a plate large enough to engage the bed joints. This way your plate
will be bearing on the "horizontal face joints" and not as likely to punch
through the exterior face shell.
I imagine there is even some bond going on that aids in shear but I have not
seen good "direct shear" test results on a bolt and this does concern me.
Torque testing the bolt to evaluate its shear capacity is bogus i.m.o. and
the code will allow you to do this. If you are going to test the bolts in
shear I suggest you install a wood block over the bolt and tighten down the
nut. Push the block with a ram, you may be able to push up, off an interior
wythe which forms a "ledge" at the ceiling line. You may also find an
interior brick masonry chimney (likely used for a heating stove in the old
days) that you can push off of horizontally. The wood block (nut tight) will
help relieve some of the bearing on the interior face shell for your test.
Without it, just pushing on the bolt will lead to early crushing of the
interior face shell.
Wall slenderness has always concerned me with this material. and I have seen
it laid up with independent wythes, but as Nels stated this does not seem to
be the rule. If aesthetics and cost will allow, I'll brace the wall for out
of plane loads every time.
If you have a situation where trusses rest on hollow clay tile pilasters,
provide secondary support. Even if you are in a zone that doesn't require it
(zone 2B if memory serves me this evening). I worked on a car dealership
after the Northridge quake that had the truss situation described above in
the service area. After the quake the pilasters looked like my children's
early attempts at building columns with their little wood blocks. The
individual tiles had displaced laterally perpendicular to the wall and
rotated relative to one another. It truly looked like a little kid had
stacked up a bunch of blocks.
Finally, when using screen tubes do a little research and use the tube with
the largest holes. I would even go so far as to watch the contractor squeeze
the epoxy out of a couple test screens. Make sure he knows the sufficient
volume of epoxy in each screen to produce the desired effect (bulbs) inside
the wall. And don't use cold epoxy.
Plan on using a lot of anchors since they have fairly low capacities. You
might check out an ICBO report for hollow cmu values, reduce them by 60% for
tile design purposes and test to verify.
----- Original Message -----
From: Nels Roselund <n.roselund(--nospam--at)worldnet.att.net>
Sent: Saturday, May 20, 2000 10:11 AM
Subject: Re: Hollow clay tiles
> The Uniform Code for Building Conservation, Appendix Chapter One provides
> the appropriate criteria for seismic retrofit for a hollow clay tile
> building. I expect that you will find that the 8" + 4" tile wall is
> actually made up of interlocked L-shaped or T-shaped tiles, or that the
> wythes of tile are interlocked with a periodically -spaced bond-course of
> tiles laid flat through the wall -- I would be surprised to find that the
> two wythes are independent.
> In-place shear testing in accordance with UBC Standard will need to be
> modified somewhat: The jack load will need to be distributed over the end
> the unit to avoid crushing (I like to use large, thick steel plates with a
> bed of gypsum plaster (USG Hydrocal, or equivalent) cast between the
> and the units to distribute the load. The typical mortar bed joint in
> construction approximated the unit wall thickness -- rarely were the
> bed joints laid full. In calculating the mortar strength per formula
> the Ab needs to be modified to use the net mortar joint widths rather that
> the full bed-joint area of the tile unit.
> Good wall anchorage capacities can be attained with resin-adhesive anchors
> (Hilti, Epcon, Simpson and others), but the design capacity will need to
> determined by testing -- no resin-adhesive manufacturer has ICBO approval
> for anchor capacities in hollow tile. I design the anchors to be made
> screen tubes that are partially wrapped with duct tape, to confine the
> adhesive, but allow it to form bulbs behind the masonry unit shells.
> There is a rumor going around that clay tile explodes, and clay tile walls
> can't be strengthened. It looks like a clay tile block explodes if you
> it with a hammer. But seismic loads aren't like hammer blows. I believe
> effective seismic retrofit can be accomplished using conventional URM
> techniques, and at costs comparable to a conventional seismic retrofit on
> URM building.
> Nels Roselund
> Structural Engineer