To: "INTERNET:seaint(--nospam--at)seaint.org" <seaint(--nospam--at)seaint.org>
Subject: Re: ACI 355.2
From: Mark Gilligan <MarkKGilligan(--nospam--at)compuserve.com>
Date: Tue, 20 Mar 2001 09:46:43 -0500
I have received input from John Silva of Hilti and Richard Klingner who is
on the committee that wrote ACI 355.2 and I have some comments on several
The manufacturers opposing ACI 355.2 claim that the test method is
un-necessary, that its adoption will require engineers to assume concrete
is cracked, and will require changes is practice.
This seems to ignore the fact that the ICBO Evaluation Reports for anchors
(Including ER-1372 issued to ITW RAMSET/RED HEAD) state that the values are
limited to installation in uncracked concrete. To me, this means that if
the engineer determines that the concrete is cracked the ICBO approval is
The status quo that the manufacturers opposing ACI 355.2 are proposing is
one in which the engineer is left with no guidance. One strategy would be
to not use post-installed concrete anchors where concrete is assumed to be
cracked, but somehow this does not seem to be what is implied by the
www.alertaci355-2.com web site. That web site seems to imply that we
should use these anchors in cracked concrete and not worry despite what
Unless you can change the ICBO acceptance criteria there is a need for
anchor values in cracked concrete and for anchors subject to earthquake
loading. It appears that ACI 355.2 provides us with a way forward. The
manufacturers opposing ACI 355.2 have not offered any alternative.
It appears that one of ITW's sister-operations in France, SPIT has an
anchor that can pass the ACI 355.2 criteria so we should be assured that
there will be more than one manufacturer that will be able to provide
anchors that comply with ACI 355.2.
The ICBO acceptance criteria will be discussed at the ICBO ES public
hearing on Tuesday, April 8 at the Furama Hotel near LAX.
On the question of whether any anchor failures have been reported Richard
Klingner, who monitors this list, states that there have been instances of
anchor failure. Apparently because the investigators were not looking
specifically for failures due to cracks in concrete the presence of cracks
were not reported. A post-installed anchor that fails as a result of
cracked concrete will manifest itself as an anchor that simply comes out of
a hole. Do we have a situation where because we have blinders on we do not
see a problem?
This is a partial answer but it would still be desirable to have a specific
example of and anchor failure due to cracked concrete.
It is commonly accepted that concrete cracks, and there appears to be no
dispute that many anchors in cracked concrete have lesser capacities, at
least when tested as per ACI 355.2. The question then becomes one of why
have we not seen more failures?
Some of the possible reasons that we have not seen more failures are:
--Cracking is often associated with factored loads which do not commonly
--The loads on anchors may be less than assumed.
--The use of multiple anchors in a connection may provide enough capacity
to prevent failure even if one anchor fails due to cracking.
--The test procedure may not reflect the installed situation.
I would be interested in specific reasons why the test procedure is
believed to be not reflective of field conditions.
Given that our codes are based on the acceptance of a certain number of
failures it could be argued that one could ignore the impact of cracks on
most connection designs. The problem then becomes one of making sure that
the tradeoff of failures against construction cost is reasonable. This
typicall consists of identifying the total annual loss due to failures and
comparing it with the cost of preventing the loss. I would welcome a
letter from one or more of the manufacturers showing a cost benifet
analysis of such a trade-off.
In the case of connections that have to carry load during and after an
earthquake I do not think that it is reasonable to ignore concrete cracks
in determining connection capacities. During earthquakes we will often see
loads on the concrete in excess of those used in design. In addition the
loads on the anchors during an earthquake may be considerably larger than
assumed. These realities combined with the relatively infrequent
occurrence of earthquakes, and thus limited experience with anchors in
these situations, argues that we should not ignore the impact of cracks on
anchor capacity when dealing with earthquake loading.
The manufacturers objecting to the adoption of ACI 355.2 state that:
"The European Commission has determined that the new code does not meet the
basic criteria of the Construction Products Directive, thus creating
unnecessary burdens for builders, producers and consumers alike! The
Commission's Directorate General for Enterprise directed the European
Office of Technical Approvals (EOTA) to determine a mechanism to address
these complaints and to report its proposed method to accomplish this
mandate, including substantive revisions to the ETAG. "
Apparently this issue was raised by ITW.
According to John Silva of Hilti, who on occasion has monitored this list
server, it is not clear that this review will result in any significant
There appears to be a difference of opinion as to the significance of any
changes to the ETAG documents. Can somebody list the specific changes and
comment on the significance of each.
A number of people have expressed concern about hearing of this problem at
the "last moment". It should be noted that there was aa Aug 28,2000
posting to this list by Kent Carlson in which he summarized the issues.