be referring to this device as the “XRF”
and the portable spectrometer as
“PMI.” By then, PMI had been adopted
by many major end users, and a policy
of performing PMI on “every alloy item
that came through the gate” was adopted by a large number of process plants
around the world.
The fate of valves was worse. As
multi-component items, they suffered
from interpretation in PMI specs that
implied or specifically stated that all
components were to be examined. This
situation was most troublesome early
on, when the ability to mask or examine
small surfaces by PMI was limited.
Eventually, however, a common-sense
interpretation was born that defined the
necessary components for examination
as body and bonnet. Some enthusiastic
or cautious users wanted to examine the
stem also, since it handily was visible at
the top of the valve and often marked
with its material identity. The examination of the stem sometimes depended on
whether the user considered that stem a
pressure-retaining part or not.
All cast surfaces exhibit some form
of surface segregation, and combined
with surface roughness, this led to much
difficulty at first with valve PMI that
was not present with wrought components. The exact number of incorrect
items that were discovered and items
that may have been improperly rejected
will never be known. But certainly, a
good deal of time was expended in retesting and in correspondence about the
exact nature of non-compliances.
Eventually, the practice came full
circle as it became common to rely on
PMI reports and not believe the mill
certificates. However, if you assume the
certificates are not fraudulent, this
reliance is an inherent error in judgment
since the certificates are made from
readings of more accurate analysis and
technique than the readings of the XRF
device. If you question this, ask a receiving inspector—especially one who operates a PMI machine.
MANUFACTURERS AHEAD
OF THE CURVE
A saying in Europe is that, when the
weight of the paper equals the weight of
a product, that product must be ready to
ship. U.S. practice has never been that
dramatic, but there has been creative
tension for a long time between users
and manufacturers. Manufacturers have
never been unwilling to supply documen-
tation, given the reality that costs are
involved. Some, however, have not been
set up to supply requirements on a rou-
tine basis, so it has been a matter of
inertia that NDE has not been more
widely performed on commodity materi-
als. It is more common that users have
been loath to pay for additional require-
ments as a standard, especially in the
old days when documentation was more
poorly regarded.
Hydrotest of finished valves, with gauges and
timers in sight
Wet mag examination of castings
(liquid penetrant testing) and wet MT
(magnetic particle testing) of all these
castings.
THE FUTURE
We are not at the end of any NDE
trend. The cycles we see now will continue into the future. NDE will continue to grow in terms of percentages of
evaluated valves shipped, and the likelihood of users reverting to the practices of old, where the NDE was of
minor or no interest, will probably
never come again. Only efficiencies of
design—such as using more forgings,
which inherently requires fewer NDE
practices than castings—will put a
dent in the growth of NDE. VM
RON MERRICK, P.E., is director of piping material
engineering at Fluor and a Fluor Senior Fellow.
Reach him at ron.merrick@fluor.com. Merrick
will be presenting on the topic of NDE at VMA’s
Technical Seminar, March 4-5, 2010, in San Antonio,
TX. For more info on this event or to register, visit
VMA.org > Events > Technical Seminar.
