Castings cut in half for examination and
measurement
Checking dimensions against a manufacturing
drawing
gun barrel was magnetized and a compass moved across its surface. By the
1920s, magnetic particles were substituted for the compass. British manufacturers then began building equipment
for magnetic particle inspection, and by
1934, Magnaflux followed in the U.S.
Identifying alloy castings for valves
presented a challenge from the start, with
a variety of methods developed to distinguish alloys by differences in reaction to
various mechanical and chemical tests.
The chemical testing used solutions that
changed color in response to the presence
of a particular element. These chemical
test kits are still available today, though
most are more sophisticated than during
the era when this was one of the technician’s main tools.
Methods that used the pattern of
sparks given off when a particular metal
was dressed with a grinder were first
described in 1909. Other methods of testing metals included chip testing and
determining color of the dressed surface.
A standard test procedure for the
various chemical and spark tests—
ASTM STP550—was last updated in
1986.
The definitive method for determin-
ing chemical composition of an object
used to be doing laboratory chemical
analysis on a small sample. These wet
chemistry labs are still seen today in
many third-world manufacturer facili-
ties, but are becoming rarer because of
the faster, less labor-intensive methods
now universally available. By the
1960s, the mass spectrometer, which
was first constructed in 1919, then
adapted for industrial uses during
WWII, took over routine analysis of
chemistry in industrial labs. The X-ray
fluorescence analyzer, which operates
on the same basic principles, was devel-
oped in the late 1960s and became fair-
ly common by the mid- to late-1980s,
along with the arc/spark OES (optical
emission) method commonly used in
Europe.
USER PERSPECTIVES
High-temperature failures and corrosion issues have always been present in
process services. In earlier years, users
tried to find ways of identifying impending failures. One of the first monitoring
techniques users adopted to identify corrosion status in piping was sentinel
drilling, a simple practice used in
British boiler services in the 19th century. It consisted of drilling holes into the
outside of a pipe, with the bottom of
each hole a certain distance away from
the inside wall. This distance, which was
supposed to be equal to the corrosion
allowance, provided an alert in the form
of leaking process fluid when the pipe
wall had corroded to that point. The
lack of precision, and more importantly,
the fact a leak had to be intentionally
created, put an end to this method.
Gate valve casting prepared for radiography
During the mid-20th century, some
users constructed their own facilities,
although in remote parts of the world,
contractors were used for most construction. As a consequence, both users
and contractors developed extensive
purchase specifications for manufacturing and examination of products. Often
these specifications were written under
the premise that a particular valve order
was created just for them, including
design of patterns. In many cases, this
was true because distribution consisted
of many individuals or small companies
representing only one manufacturer or a
manufacturer’s own branch offices in
major cities. The massive inventories of
multiple products that exist today were
largely unknown.
These specifications, which became
lengthier over time as new problems
were found and addressed by the specifications, were directed to the particular
needs of one user. Hydrocarbon users,
for example, had requirements about
high-temperature hydrogen, while
chemical users had requirements for
laboratory testing such as the intergranular corrosion test. But beyond these
specific cases, the idea that one product
met all user requirements was more or
less taken for granted, and the quality
witnessed by inspectors for commodity
valves often didn’t go much beyond factory hydrotesting.
This began to change during the era
when new manufacturers began coming
onto the scene. By the 1950s, there
were manufacturers from outside the
U.S. making products for the major oil
companies while being managed by
major contractors that had limited
experience with American standards
and practices.
The rise of a movement toward better
quality, which arrived in the process
industries by the early 1980s, meant a
large difference in customer requirements. While some industries, such as
nuclear power and the military nuclear
program, had very extensive documentation and testing requirements, extensive testing was somewhat less prominent in process work.
During this time, efforts were made to
add new requirements that addressed
