Chemical Engineering June 2016 - 64

Field troubleshooting requires engineering...
* Review available documents (piping and instrumentation diagrams, standard operating
procedures, cut sheets)
* Review available data (tests, event logs)
* Contact vendors or use the internet to get cut sheets and parts lists
* Have the right equipment with you, and when you don't, improvise
* Break the problem into parts and isolate each issue
* Keep in mind that failure of a new component is frequently the result of its being the
wrong component, or improper installation or operation
* Understand that very few of the problems encountered in process systems are truly
random; if something happens more than once, there is a root cause
* Organize your work with a checklist and data sheet, and document what you have
already checked out
* Change one variable at a time - isolate cause and effects from each change
* Understand the chemistry of the process as it was designed, but also of any foreign
material or unexpected byproducts
* Run calculations to home in on the issues and prove or deny suspected root causes
* Get equipment model/serial number/job number nameplate data, by hand or photo
* Take good notes and always write up a trip report to close out the job
...and the right behaviors
* Actively engage the operators and maintenance personnel. Ask for their input, and be
aware that all personnel have the potential to answer key questions
* Ask for help - don't let your pride get in your way
* Be persistent and unafraid to ask the " stupid " question
* Ask what happened right before any unpredicted event
* Call the original equipment designers
* Call those who installed the system
* Call those who worked previously with the system
* Routinely use senses (eyes, ears, nose and touch) in addition to your brains
* Take photos for later review
* Write up notes at the end of each day or the beginning of the next day to keep up with
the data and spot errors
* Take action - don't be afraid to try things out
* Get out and around the equipment, and be willing to get dirty as you investigate
* Set aside time to think over information and discuss with others onsite
* Use the human factor to your advantage; people are a major resource of knowledge
Commissioning a pneumatic conveying
loop. While working as a
freshly minted engineer at Particulate
Solid Research in New York City, one
of us designed and built a pneumatic
conveyor system for industrial research,
made from 200 ft of Plexiglas
pipe with an inside diameter of 4 in. It
was instrumented with pressure taps
every five feet. During commissioning,
the goal was zero leakage.
The leak test method used a simple
tool: a rotameter. Both ends of
the conveyor pipe were capped off,
and compressed air was added via
a pressure regulator through the
rotameter, slowly raising the pressure
as leaks were found and sealed. The
rotameter reading provided visual
feedback on the size of the leak (for
instance, an open valve, a missing
pressure tap plug, or weld porosity).
Once the big leaks were found and
fixed, a persistent low-level leak remained.
The technicians used soap
and water to check multiple Victaulic
64
couplings used on the Plexiglas piping,
the threads on pressure taps,
and the valve stems - to no avail.
A change in personnel and sheer
persistence eventually located the
felonious fitting: a Victaulic coupling
in a downcomer just below a penthouse
floor (Figure 5) [3]. The coupling's
donut-like rubber gasket carried
concrete chips that sliced up the
rubber. With a fresh gasket, leakage
fell to effectively zero. A final safety
note: the compressed air approach
is not acceptable for leak-testing
at high pressures or large volumes,
due to the amount of stored energy
in these cases.
Preventing bypassing
Close cousins of leaks are bypassing
problems, in which gases, liquids or
solids go unexpected places inside
a system. One example concerns
a transportable high-temperature
incinerator that processed soil contaminated
with coal-tar creosote.
Notches from
ball clamping the
poppet stem
Figure 4. This 1-in. long brass poppet from an
unknown source became jammed inside a ball
valve in liquid propane service. It stopped the
valve closing fully, and so allowed backflow
When stack tests were run on the
incinerator, two runs passed the required
99.99% lower limit for creosote
destruction efficiency by a wide
margin. The third test, however,
showed far worse results. Interviewing
the operators revealed that the
failed run was " less stable " than the
other two. After further consideration,
the operators zeroed in on the
symptom that, on that run, the draft
in the kiln was not stable.
The hot solids from the 7.5 ft ×
45 ft kiln exited to a rotary cooler,
with the exhaust gas from the cooler
being ducted to a baghouse. This
exhaust should have contained only
steam, air and dust. However, when
the draft was momentarily unstable,
CO2 was detected in the duct as well.
This confirmed that incompletely oxidized
creosote vapors could bypass
the oxidizer along with the CO2. The
fix was to reroute the cooler fume
duct to the oxidizer inlet, thereby fully
treating the fume (Figure 6).
Upstream problems
If you don't look, you don't see, and
if you don't go inside, all you have is
speculation. Sometimes you have to
follow the process upstream to find
the source of the problem.
Ionizing wet scrubber. An incinerator
stack test showed particulate
emissions at about 60% over
ChemiCal engineering www.Chemengonline.Com June 2016
http://www.Chemengonline.Com
Chemical Engineering June 2016

Table of Contents for the Digital Edition of Chemical Engineering June 2016
