Hydrocarbon Processing - May 2021 - 48

Maintenance and Reliability

FIG. 5. Simulation-using proprietary softwarea-of the hydrogen valve example from FLIR
camera imaging.

FIG. 6. Adjusted simulation-using proprietary softwarea-to account for isentropic effects for
hydrogen valve.

has been identified that the tee at Mach
velocity to a flare header can experience temperatures below the MDMT. At
this point, a mechanical stress analysis
should be performed to determine the
risk. In most cases, the mechanical stress
analysis passes. In some cases, it identifies the need to consider the risk of exceeding the stress analysis for a limited
amount of time while the PSV is opened,
requiring determination of whether the
entire pipe metal wall can sustain these
temperatures below MDMT. If the risk is
high enough, then a change in flare header design is merited.
Modeling of isentropic effects. It is
commonly assumed that process simulators are an accurate method of indicating
the temperatures and behaviors of a system. However, steady-state simulators do
not predict isentropic behavior. Proprietary softwareb can predict this behavior
but only in Dynamics mode and requires
user knowledge to implement the setting to report it. Any user can build it out
themselves. For example, the results of a
default simulation for the 99.9% hydrogen valve example is shown in FIG. 5.
48

MAY 2021 | HydrocarbonProcessing.com

The isentropic temperature for the
hydrogen valve example should be approximately 65°F, unlike the simulation
calculated value of 81°F. To measure the
temperature accurately by using steadystate simulation, a spreadsheet or other
embedded calculation should be created
to report the change in temperature via the
isentropic effect, along with the final temperature. An example of what this looks
like for the hydrogen valve is displayed
in FIG. 6, showing the 65°F, which aligns
closely with the FLIR camera imaging.
Proprietary flare modeling softwarec
can be used. However, the default settings
do not predict the behavior of isentropic
effects. The user must change the settings
to " include kinetic effects. " Other simulators have not been explored in detail.
Where to consider isentropic effects? Isentropic effects can create issues
with many elements of a system, including areas that have not yet been studied.
These include but are not limited to:
* Icing effect on the integrity of
metal and soft goods in PSVs,
flare headers and other
constituents

* PSV tailpipe and flare header
evaluations/metallurgy
* Pressure drop calculations that
can be marginally impacted
* Blowdown valves (potential
pipe embrittlement)
* Ejectors on top of vacuum
towers or systems
* Potentially on heat exchanger
performance impacting the heat
release curves (though very rare
due to velocities < 0.3 Mach for
standard exchanger designs).
There is also an ongoing study regarding
how hydrate formation may be impacted
at higher velocities.
Takeaway. To ensure systems are running safely, isentropic effects must be
considered where high-velocity gases occur or where lower than expected temperatures may impact design or operation.
While steady-state process simulators do
not easily simulate isentropic behavior,
the implementation of spreadsheets or
additional settings can aid in improved
predictions. By being able to determine
when and where isentropic effects are relevant and how to address them, any risk
to a system's integrity, safety and longevity can be reduced.
ACKNOWLEDGEMENTS
The authors would like to send special thanks
to Bart Carpenter, Parv Consulting, and Ben Day,
Tailwater Technical.
NOTES
Aspen HYSYS
b
VMG Symmetry by Schlumberger
c
Aspen Flare System Analyzer
a

ERIC PARVIN is the Founder and
Principal Consultant of Parv
Consulting in Denver, Colorado.
He has extensive experience in
upstream, midstream, refineries,
rail car facilities and petrochemicals,
with expertise in simulations,
towers/trays, flares, debottleneckiing and PSM/PSV
evaluations. With more than 24 yr experience, he also
serves as adjunct professor at the Colorado School of
Mines, is an editorial contributor of the " GPSA
Engineering Data Book " and member of API 520
and 521. Mr. Parvin earned a BS degree in chemical
engineering from Louisiana State University and is
licensed in multiple states and Alberta, Canada.
ADRIANA ROBINSON is a Process
Engineer at Parv Consulting. She has
recently completed her BS degree
in chemical engineering from the
University of Colorado Boulder, with
a minor in space and an emphasis in
energy process engineering.

http://www.HydrocarbonProcessing.com

Hydrocarbon Processing - May 2021

Table of Contents for the Digital Edition of Hydrocarbon Processing - May 2021