Heat Exchanger Tube Rupture Scenario Evaluation using Aspen HYSYS Dynamics

This presentation covers process safety considerations and when a dynamic simulation is required. We also provide a modelling approach and a case study on Coker Bottoms Steam Generator, which includes information on device selection and device sizing.

1. 03/01/13
1
Heat exchanger tube rupture
scenario evaluation using Aspen
HYSYS Dynamics
By: Francisco Da Silva, MSc. P.Eng.

2. Overview
•Introduction
•Process Safety Considerations
•When Dynamic Simulation is required?
•Modelling Approach
•Case Study: Coker Bottoms Steam Generator
–Device Selection
–Device Sizing
•Conclusions
•Questions & Answers
2017©COPYRIGHT PROCESS ECOLOGY INC.
2

3. Company Background: Process Ecology
•Founded 2003, Calgary, AB
•Key Competencies
–Engineering consulting,
process simulation
–Process engineering &
optimization
–Air emissions estimation
and management
–Software development
3
2017©COPYRIGHT PROCESS ECOLOGY INC.

4. Heat Exchanger Tube Rupture: Why Simulate it?
•Goodyear Synthetic Rubber Factory [1]
•Plant was shutdown for maintenance
•Isolation valve was closed (upstream of
relief valve)
•Steam was used to clean tube side
•Steam heated ammonia causing a
pressure buildup, ending in a heat
exchanger failure
•1 operator killed and 7 injured, loss of
containment, and thousands of dollars in
equipment damage
4
1
U.S. Chemical Safety Boards, CSB Investigating Causes of Fatal Rupture of Heat Exchanger at Goodyear Synthetic Rubber
Facility in Houston, 2008
2017©COPYRIGHT PROCESS ECOLOGY INC.

5. Process Safety Considerations
•API 521[1]: recommends the evaluation of tube
rupture scenario when the Maximum Allowable
Working Pressure (MAWP) of the Low Pressure
(LP) side is lower than 10/13 of MAWP of the High
Pressure (HP) side of the Heat Exchanger (HE).
•The Pressure Relief Device (PRD) needs to be
sensitive enough to react quickly and relieve the
pressure.
•Opening times for Pressure Relief Valves (PSV) are
in the range of 50-100 msand for graphite Rupture
Disks (RD) around 1-10 ms[2].
5
1
American Petroleum Institute, Pressure-relieving and DepressuringSystem. API Standard 521 6 ed., Washington: American
Petroleum Institute, 2014.
2
S. Nagpal, "Evaluate Heat-Exchanger Tube-Rupture Scenarios Using Dynamics Simulations," Chemical Engineering, pp. 48-53, 2015
2017©COPYRIGHT PROCESS ECOLOGY INC.

6. •Reactive systems,
•HEs in which the pressure
difference between the HP and
LP sides exceeds 7,000 kPa
(~1,000 psi), or
•HEs where the LP side is liquid-
full.
6
1
American Petroleum Institute, Pressure-relieving and
DepressuringSystem. API Standard 521 6 ed.,
Washington: American Petroleum Institute, 2014.
2
S. Nagpal, "Evaluate Heat-Exchanger Tube-Rupture
Scenarios Using Dynamics Simulations," Chemical
Engineering, pp. 48-53, 2015
2017©COPYRIGHT PROCESS ECOLOGY INC.
When is a Dynamic Simulation required?

7. Modelling Approach
•LP side of the HE needs to be
divided into sections
•Number of sections will depend on
HE dimensions. Each section is
modeled as a two-phase separator.
•The rupture is located at the HP
side outlet.
•HE duty is kept constant.
•Step time needs to be reduced to
0.1 -1 ms.
•Flash efficiencies is reduced to 5%.
7
Taken from: R. UrdanetaPerez and J. Oude Lenferink, "Design
pressure reduction in high-pressure heat exchanger with dynamic
simulation," Hydrocarbon Processing, pp. 37-41, 2015.
2017©COPYRIGHT PROCESS ECOLOGY INC.

8. 1
American Petroleum Institute, Pressure-relieving and DepressuringSystem. API Standard 521 6 ed., Washington: American
Petroleum Institute, 2014.
2017©COPYRIGHT PROCESS ECOLOGY INC.
•Tube failure is a sharp break in 1 tube [1].
•Tube failure is assumed to occur at the
back side of the tubesheet [1].
•HP fluid is assumed to flow both through
the tube stub remaining in the tubesheet
and through the other longer section of tube
[1].
•Flow estimation is based on Homogeneous
Direct Integration (HDI) method [1]
Tube Rupture Flow Estimation [1]
9
1
2
=
−2∫4633eυe
a
an
υ

2
Pto
=l

2
e−2u
4633e
l
a
an Pto

9. Case Study
Client:Confidential
ProcessUnit:CokerUnit
Service:CokerBottomsSteamGenerator
DesignDuty:88.5MMBtu/hr
TEMAtype:AKT
Fluids(tube/shell):CokerBottoms(oil)/BFW
Tubesideflow:5,100bpd
TubesideTemperature(in/out):700/500°F
TubesidePressure(in/out):85/81psig
ShellsideFlow:97,200lb/hr
ShellsideTemperature(in/out):458/458°F
ShellsidePressure(in/out):445/445psig
DesignPressure(tube/shell):550/177psig
RDPressureSet:140psig
Taken from: Gary, J.H; Handwerk, G.E.; and Kaiser, M. J.
Petroleum Refining: Technology and Economics (5
th
Edition).
CRC Press (2007)
2017©COPYRIGHT PROCESS ECOLOGY INC.
12

10. •Simplified model of Coker Fractionator
•Coker Fractionator Bottoms Pumps modeled using Performance Curves
•Pump discharge check valves (10% back-flow)
To Steam Generator
tube-side inlet pipe
2017©COPYRIGHT PROCESS ECOLOGY INC.
13
Case Study: Model Scope

11. •Tube Volume divided into 3 volumes (Inlet
head, tube bundle, and outlet head)
•Detailed piping was provided by client
•Evaluated Scenarios: 1 RD vs 2 RDs
From Coker Fractionator
Bottoms Pumps
Tube-side
Inlet Head
Tube Bundle
Tube-side
Outlet Head
Tubesheet
entrainment
Remaining tube
(long) entrainment
BFW @ shell
conditions
Top RD
Bottom RD
Flare Main
Header Tie-in
•No holdup considered in relief piping (i.e.
negligible volume, velocity >160 ft/s
•FlareMain Header pressure is constant (10
psig)
2017©COPYRIGHT PROCESS ECOLOGY INC.
14
Case Study: Model Scope

12. 12
•Due to uncertainty on tube
rupture flow, a sensitivity analysis
was performed to determine the
effect on the pressure rise time.
•It takes between 6 to 14 msto
reach tube design pressure,
therefore a rupture disk is
recommended
2017©COPYRIGHT PROCESS ECOLOGY INC.
Case Study: PRD Selection

13. 13
2017©COPYRIGHT PROCESS ECOLOGY INC.
Case Study:
80
90
100
110
120
130
140
150
160
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0
Pressure (psig)
Time (seconds)
HX Tube Bundle Pressure HX Inlet Head Pressure HX Outlet Head Pressure
DISK
RUPTURE
TUBE
RUPTURE
PRESSURE
PEAK
•Tube rupture occurs at 1 s.
•RD bursts 9 msafter tube rupture
(140 psig).
•A lower peak pressure (130 psig)
is reached 620 msafter the tube
rupture (oil pulse).
•The maximum pressure (150
psig) is reached 1,480 msafter
the tube rupture (steam/oil pulse)

14. •The pressure in the Steam Generator inlet will take from 6 to 14 msto reach the
design pressure in the tubeside, so RD was recommended for this application.
•The simulation model shows that the first peak reaches 130 psig which
corresponds to the RD burst pressure, and the pressure decreases suddenly when
the disk ruptures. At this moment, the system is full of liquid.
•A second pressure peak (149.9 psig) is reached at 2.44 seconds which corresponds
to a steam/oil slug when BFW flashes at tubesibepressure.
14
2017©COPYRIGHT PROCESS ECOLOGY INC.
Conclusions

15. •Once a stable condition is reached, pressures in the heat exchanger are in the
range of 116 psig.
•Based on the dynamic study, one 8 inch rupture disk is sufficient to protect the
steam generator.
•The model built in Aspen HYSYS Dynamics provided a basis for the assessment
on the steam generator tube rupture event allowing us to recommend measures
to avoid loss of containment, equipment damage, and possible human losses.
15
2017©COPYRIGHT PROCESS ECOLOGY INC.
Conclusions

16. 2017©COPYRIGHT PROCESS ECOLOGY
INC.
Question & Answers
16
?
????
??