One
of my previous articles appearing in NAMS News, New Construction
Materials: How to Avoid Being Blamed for Someone Else's Mistakes,
evoked some rather hostile and predictable reactions. One respondent,
apparently preferring anonymity, sent me an E-mail message declaring
me a horse's ass and an ignoramus for pointing out some of the
weaknesses of foam core construction.
It
seems the gentleman owns a foam cored boat himself without nary
a problem. As a skeptic of foam cores, its been lonely out here
over the years in face of so much promotion and fanfare for
the material. However, he overlooked the point that I never
said that foam couldn't be used successfully; it can and it
is.
However,
I confess to feeling somewhat vindicated when, a few days later
I ran across an editorial in Professional Boat Building by
Henry Elliot, a composites consultant out of Swansea, Massachusetts.
Mr. Elliot points up the debates ongoing in the U.K., Sweden, Italy,
New Zealand and Austrailia about the International Measurement System
for racing sailboats that apparently prohibits honeycomb cores and
more or less mandates cross-linked PVC. Without being specific,
he alludes to some large scale disasters with the material.
Below:
Typical core bonding failure. In this case, due to failure to
achieve bonding during construction. Here, the cut away section
exposing the core shows that there never was bonding to the outer
skin.
He
also points out that PVC becomes unstable at increasing temperatures,
as well as pointing out the rather limited lab testing of such laminates,
a point that I have repeatedly made. He comments briefly that testing
small coupons or small pieces of laminate does not produce the same
results as testing a large panel.
Indeed,
in all of the test data that I have seen, I've never run across
tests of large panels or complex panel shapes such found in boat
hulls. Even a non-PHd engineer like myself understands that one-dimensional
testing yields only one-dimensional results. Yet even he overlooks
the effect of shape on cored parts or hulls, a factor as large as
any, and one that is routinely ignored.
To
explain briefly, curved and flat cored panels behave in vastly different
ways when stressed, depending upon size and shape. One of the reasons
that so much delamination of foam cored hulls has occurred (far
more with sail than power boats) is due to the fact that a thick
cored laminate has a sizable difference in the radius between inner
and outer laminate.
Like
two cars going around the inside and outside turns of a race track,
one has to go faster than the other because it is traveling a longer
distance.A similar thing happens when a cored panel bends: the inner
and outer panels do not bend at the same rate. This causes a shearing
action against the core that often results in bond failure. Failures
in sail boat hulls occur most often because of the hourglass shape
with a strong double radius, whereas powerboat hulls are
mostly flat.
I
have long criticized vendors of these products for not supplying
adequate test data, or merely selective specifications that put
their products in the best light. On the other hand, boat builders
have an obligation not rely on salesmen for their knowledge, but
to acquire their own knowledge of materials.
In the meantime, Mr. Elliot says that we (its not indicated who
"we" is) need to acknowledge what we don't know. "Designers
and vendors have to recognize that laminate and process information
generated in the lab often has little connection to what is achievable
on the shop floor."
Precisely.
Not because accurate data can't be obtained in the lab, but because
so much of the lab testing that is done is based on a faulty premise.
Dropping a heavy ball on a coupon suspended between two uprights
- which is the sort of data we most often see - constitutes nothing
more than an impact test of a coupon. It hardly approximates the
actual stresses imposed on a boat hull, or even a section of a hull.
Is this test supposed to approximate the stresses induced when a
twenty ton hull slams into a wave? Or falls off a wave?
There's
a vast difference between lab testing and conditions like these.
Here the author sea trials a 115' Crescent motor yacht in 6' seas
at full speed. Meeting a wall of water at 20 knots is a more realistic
method of load testing. The hull is sounded out after the sea trial
to check for evidence of delamination. See additional note below.
Yet
that is precisely what a marketing demo tape that I recently received
from a supposedly independent lab would have me believe about the
promotion of a new product.
In
this instance, dropping a shaft of steel on the coupon was supposed
to convince me of the strength of a composite when applied to an
entire hull. Most of this sort of so-called data appears to be little
more than self-serving promotion. A few of the tests that are usually
lacking are those that determine cleavage strength, i.e., the strength
at which the foam bonds to plastic resin; shear strength; effects
of temperature, and a host of shaped and flat panel stress tests.
Despite
my disdain for the material, even the weakest foams can be used
successfully when a structure is properly designed. But proper design
requires a tradeoff in terms of detailing that raises labor costs,
a tradeoff that most builders are not willing to accept. And so
the failures continue.
Proper
testing of materials and design need not be either particularly
expensive or difficult. Anyone with the most rudimentary engineering
knowledge could do it. But it does require a variety of tests on
appropriately large structural sections of a size commensurate with
differing hull shapes, whether on a scale models or a manufactured
cross section of panels. Unfortunately, the people promoting these
products and processes show little interest in producing reliable
data and so the failures and disasters continue.
As
I pointed out in my last article, recent hurricanes have put thousands
of hulls to some of the most rigorous tests imaginable, at a cost
of millions. I took this opportunity to compare hundreds of boats
with varying construction materials and methods, and just from a
cursory examination it was not difficult to discern which materials
and designs fared the best under catastrophic conditions.
The
proper means and methods of testing are known. We already know most
of what we need to know, its just a question of whether those who
design, manufacture and apply those products choose
to make proper use of that knowledge. The evidence shows that far
to many don't.
In
the meantime, its the role of the marine surveyor to serve as the
fourth estate of the boating industry by making sure that we are
fully up to date on our methods of surveying and ensuring that our
clients don't become the victims of ignorance - ours or anyone else's.
Footnote:
We
recently completed the survey of a 115' fiberglass yacht built by
Crescent Boat builders of Vancouver for a longtime client. This
yacht utilized a double layer of Airex with solid laminate
in between. With two surveyors on the job, we went over this hull
with a fine tooth comb and several heavy hammers and gave her a
thumbs up. The reason we did so, despite the extensive history of
problems with fully cored hulls, was the exceptionally good attention
to detail that we found throughout the hull. Just from observing
how she was framed out, we could tell that the designers and builders
knew what they were doing. And because they understood the weaknesses
of the material, they also understood how to use it. We did not
find a single area of incomplete bonding or bonding failure. And
this, on a 115' hand laid hull.
Posted
4/19/97 Copyright (C) 1997 David H. Pascoe
|
