Hydrocarbon Processing - February 2022 - 17

Engineering
Case Histories
A. SOFRONAS, CONSULTING ENGINEER
http://mechanicalengineeringhelp.com
Case 115: A method for analyzing
catastrophic type failures
A portion of my consulting career involved investigating catastrophic
type failures on machines, pressure vessels and structures
in the industry. Causes and solutions were required quickly
so the equipment could restart safely, and production resumed.
While collecting and analyzing the damaged and mangled
parts, my involvement has been to propose and test theories on
the most probable causes. Some form of analytical analysis is
used based on the available data, and I share this information
with the investigating team and modify this as I obtain new information.
This helps focus the team on all the confusing debris
and input from interviews and increases understanding of what
might have happened. Testing the theory in this way will either
validate it or provide reasons to dismiss it. This is similar to
what is done using the scientific method. In the scientific method,
a theory, hypothesis or best guess on how something has
occurred is proposed based on the data available. As additional
data are gathered, such as interviews/metallurgical/material
evidence, they are tested to see if they agree with the theory.
It has been my experience that several events usually have
to occur to result in a failure-lack of or delayed maintenance,
overloads due to exceeding operating or design specifications,
faulty repairs or modifications are typical causes. Major design
deficiencies from the manufacturer are rare if built to their
specifications. A long history of successful operation before a
failure suggests this.
Consider theorizing the wreck of a 2,000-hp gas enginereciprocating
compressor as shown on FIG. 1.
The following theory was proposed after reviewing the data
shown in TABLE 1, which is only a small sample.
The slipper bolts had become loose over time, which was
the ticking heard by the operator. The dynamic load impacted
the body of the bolts and caused them to fail in shear. The
loose slipper wedged itself against the piston rod, and when the
crosshead came forward, it sheared the rod. This was the bang
the operator heard.
Others suggested several other causes; however, the data did
not support them. The solution was to install dowels to secure
the slippers in addition to the bolts. This eliminated the cyclic
load on the bolts, which loosened them. Periodic torque inspections
were recommended as a maintenance item. No such
failure occurred for 15 yr after the modification, at which time
the machine was replaced with a centrifugal compressor.
NOTE
Case 114 was published in HP October 2021. For past cases, please visit
www.HydrocarbonProcessing.com.
LITERATURE CITED
1
Sofronas, A., Analytical Troubleshooting of Process Machinery and Pressure Vessels:
Including Real-World Case Studies, Wiley, 2006
TONY SOFRONAS, D. Eng, was the worldwide lead mechanical
engineer for ExxonMobil Chemicals before retiring. He now
owns Engineered Products, which provides consulting and
engineering seminars on machinery and pressure vessels.
Dr. Sofronas has authored several engineering books and
numerous technical articles on analytical methods.
TABLE 1. Data to help verify theory
Observation
1. An experienced operator heard an
unusual continuous ticking and then
a bang just before the failure.
2. The compressor piston rod had
sheared off due to wedging
of the broken slipper.
3. The slipper bolts had all sheared
off due to dynamic impacting.
4. The slipper bolts had not been
torque checked for many years.
5. There were two bolts with
low torque on the six other
slippers out of 24 bolts.
6. The safety wire was still in place.
7. The slipper was cast iron and
broken into pieces, and the
3-in. diameter piston rod had
sheared. The nut was tight.
FIG. 1. Wreck of doghouse section
8. This type of failure had happened
to the same type of unit at
another division.
Source
Interview.
The calculation was performed
along with metallurgical
analysis-high wedging
force available to do this.
The calculation was performed
along with metallurgical
analysis. Loose hold-down bolt
torque and shear probable.
Maintenance records and
interviews.
Torque wench.
Observation.
Observation and
torque wrench.
Interview with a colleague
at a gas compression plant.
Hydrocarbon Processing | FEBRUARY 2022 17
http://www.mechanicalengineeringhelp.com http://www.HydrocarbonProcessing.com
Hydrocarbon Processing - February 2022

Table of Contents for the Digital Edition of Hydrocarbon Processing - February 2022
