Beruflich Dokumente
Kultur Dokumente
David Averous, Florian Picard and Gilles Aubert, Fives Cryo France
ABSTRACT
Aluminium Brazed Plate Fin Heat Exchangers have been frequently used in the cryogenics industry, in
particular in gas processing applications such as LNG plants. The complex technology of Brazed
Aluminium Plate-Fin Heat Exchangers (BAHX) offers a high-level heat transfer capacity and a large range
of allowable pressures, but requires highly skilled design in return.
Optimal design leads to some difficulties due to the fact that any hydraulic or thermal perturbation on a
stream is directly propagated to the others through the aluminium core matrix, and could also directly
impact the mechanical integrity of the exchanger. This is particularly applicable to off-design and
nonsteady state conditions.
The paper focuses on the implementation of technical resources in order to provide accurate expertise for
the operation of BAHXs, including thermo-hydraulic analysis and mechanical stress studies. Inhouse
computational tools are used to highlight the physical phenomena which are inherent to the behavior of
BAHXs.
Therefore, the reliability of BAHXs is directly based on such calculations in order to define and
continuously improve the standard design rules.
INTRODUCTION
Brazed Aluminium Plate-fin heat exchangers (BAHXs) are widely used in process industries such as gas
processing and petrochemical industries including cryogenic applications: industrial gases and
hydrocarbon separation are the most common ones, but many processes use this type of equipment,
including recovery of natural gas liquids, helium refrigerators and liquefiers, hydrogen purification,
ammonia and ethylene processes, nuclear engineering and synthetic gas production.
Maximizing thermal efficiency with high reliability is the main objective for such cryogenic processes.
BAHXs promote exchange between many streams simultaneously (cases with more than 12 streams are
common) allowing to save energy very efficiently.
However any perturbation on a stream entering the BAHX is propagated to the other streams and, as a
result, has an influence on all duties of its immediate surroundings. The comprehension of local and global
thermal stresses within BAHXs constitutes a primary objective to reduce the possibility of component
damage or failure during operation.
This document presents some keys both to prevent failure during operation, and also to offer the optimal
design according to drastic customers' requirements.
Figure 5 Block-in-shell
Description
A general description of a multi-stream brazed aluminium plate-fin heat exchanger is presented by
Figure 5.
General characteristics
Compactness
Lightness of aluminium
Good mechanical
characteristics at cryogenic
Temperatures
Multi-stream
Design flexibility
Energy integration
BAHX In operation
An academic example of BAHX (see Figure 8) is used here only for illustration of the complexity of such
technology, gathering some classical configurations for stream circulation. Five streams (1 to 5) are
distributed towards 160 stacked layers. These are extrapolated from four layer classes (A, B, C and D). An
individual layer class is defined as a unique functional geometry for a layer, including the specification of
all components which play a part in the flow arrangement as, for instance, distributors, length, width, type
of fins and bars. These layer classes are used to build the stacking arrangement of the core. Each layer
can define one or several stream circuits. A circuit is the way followed by a unique stream between two
associated ports, inlet and outlet, in a layer. Note that several streams can circulate independently within a
given layer class (for example streams 1 and 2 in the layer class A). Also one stream can circulate within
several layer classes which can be fed by the same header (stream 1 in layers classes A and B).
Intermediate distributors can be used to redistribute a fraction of a stream (stream 4 redistributed from C
to D). The stacking arrangement results from the sequence of layer classes, which are arranged with
accuracy in accordance with operating requirements. A part of the sequence pattern is represented in
Figure 8.
Results for sequences 1 and 2 are presented by the diagrams below (Figure 10 and Figure 11). The
maximum temperature difference diagram for sequence 1 shows a large temperature gradient in the
BAHX, which was confirm on site. It impacts directly on the overall thermal performance, and also could
obviously lead to severe damages due to too large thermo-mechanical stresses. Therefore, it is
preferable to reach balanced plate temperatures according to the stacking pattern, as it is shown in the
sequence 2 diagram. With the sequence 2 to define the stacking pattern, the BAHX reached the specified
performance.
CWT means Common Wall Temperature. This assumption imposes that the wall temperature is
considered constant at a given section according to the flow direction. It means all the plates of the BAHX
have exactly the same thermal behavior despite the stacking pattern.
Simulation is also very helpful to give efficient technical support in the event of BAHXs dysfunction on site.
For example, the following curves in Figure 12 present the analysis of the same case study (with
sequence 1) which had been characterized by confrontation between on-site experimental data and
calculated results.
Calculated plots represent the mean temperature profiles of the whole plates of the BAHX, according to
different flows-distribution in the stacking pattern. Three cases have been successively carried out by
simulation:
-
"Rated flow distribution" means that the mass flow rate is balanced between each layer,
according to each given stream ;
"Balanced flow distribution" means that pressure drops are balanced in each layer associated to
the circulation of a given stream ;
"Mal-distribution profiles" means that a mass flow rate profile has been imposed for the flow
distribution of each stream.
Plugging of a brazed aluminium plate-fin heat exchanger may seriously impact its thermal performance,
as a consequence of severe flow mal-distribution both in the stacking arrangement, and even across the
width of each layer. Therefore, the pressure drop of the plugged stream increases significantly, and, as a
result, leads to induced thermal stresses. These kinds of dysfunction can be fortunately evaluated by
software which will be presented below.
5
79
Width
Length
0.760 m
6m
Schematically, the stream pattern can be depicted on the box without a true representation of the header
positions (see Figure 15). As shown in figure 2, there is a redistribution of stream C: the layers made
available by D are used by C. In this zone only E is present and the others layers are dummy passages.
Stream A enters the exchanger in liquid state and must exit in vapor state due to the presence of a
compressor.
In Figure 17 the exit composition of A is shown. Due to the influence of stacking arrangement, its layers
exhibit a difference in their output quality and dynamics. The bold curve (A output) corresponds to overall
mass vapour fraction (adiabatic mixing of all output stream A layers). The other two curves (slowest and
fastest) correspond to the two extreme layers with respect to the response delay to the perturbation. The
stream starts leaving the exchanger below its dew point at t=121.7s.
Figure 20 Position and value of maximum plate temperature difference over time
10
However, it is interesting to see that, due to the difference in fluid response dynamics, the temperature
difference reaches a maximum before stabilization. This behavior is not uncommon and dynamic
simulation is the only way to track this kind of path. Position and value of maximum temperature difference
between two adjacent plates over time is shown in Figure 21.
The maximum is reached quickly in the distribution zone at the BAHX warm end. Due to the presence of
the headers, though, this zone may be more sensitive to mechanical stress created by the temperature
difference.
Figure 21 Position and value of maximum temperature difference between two adjacent plates
I
n conclusion, the use of the simulator leads us to a better understanding of the device and its dynamic. For
the manufacturer, it has helped in proposing better and safer designs as well as policy for start-up,
shutdown or case change. In a near future, additions should be made to the simulator to address
flowsheet level analysis and the dynamic behavior of commonly encountered processes.
11
12
Warm stream
Cold stream
kg.h-1
163109
153250
0.498
0.535
Temperature
-33.9
-37.1
Pressure drop
mbar
11.32
9.82
Total power
MW
39.62
39.62
Flow
Mass vapor fraction
13
To understand this notable feature, some clue may be obtained by representing the PROP stream profiles
of the relative difference between the mass flow rate calculated at a given x and the theoretical average
mass flow rate evaluated by assuming a homogeneous distribution of the flow along the entire width
(Figure 26). In spite of the geometric uniformity, coolant mass flow rate is not uniform across the core
layer. There is a maximum flow mal-distribution rate of 25%. Note that the largest flow mal-distribution
corresponds to the temperature imbalance described in the previous paragraph.
14
Figure 28 shows the thermal profiles of a parting sheet of the stacking arrangement, sandwiched between
two distinct layers A (above) and B (below), in which circulate streams OVHD and PROP respectively. The
plate closely retranscribes thermal phenomena induced by the flow circulation in each specific layer
geometry. Separating-bars and inactive zones of the two distributors could be considered thermal wells,
by which the energy preferentially transits.
Figure 29 illustrates the thermal profile of the top cap-sheet, which is directly in contact with the OVHD
stream. It is interesting to observe the influence of inactive zones on the two distributors implied in the
description of the layer A.
3D modeling is a highly improved method for BAHX study. The coupling of thermal and
hydraulicparameters on the global functioning of the core can be easily visualized and provides significant
clues to understand such a complex system, and to avoid near dry out condition or unstable flow regime,
or to check the maximum allowable liquid entrainment.
Mechanical Integrity
As with any pressurized heat exchanger, stresses in each component of a brazed aluminium BAHX have
to be maintained within allowable limits to prevent component damage or failure, due to pressure loads,
externally applied loads (e.g. piping forces and moments), and thermally induced loads producing
stresses.
Because components in BAHXs are relatively closely and rigidly connected, a large local metal
temperature difference may cause significant thermal stress, which could result in local mechanical
failure or damage during life time. Though BAHXs are very tolerant of large steady-state stream-tostream
temperature differences, they may be damaged if subjected to transient or continuously unsteady
operating conditions producing excessive thermal stresses.
A dedicated software tool SiTEME (4) has been developed to study the thermo-mechanical behavior of
BAHX. This tool integrates homogenization techniques that replace geometrical heterogeneities of
brazed fins and sheets and thus reduces considerably model size (see Figure 30). This allows first to
perform FE calculations and subsequently determine results inside the structure using localization
techniques. Therefore appropriate criteria can be chosen to validate the thermo-mechanical integrity and
the fatigue failure based on the pressure vessel codes interpretation of elastic analysis of the BAHX.
15
Figure 31 presents the main connections between SiTEME and Fives Cryo's in-house software, in order
to give overall needed information for the thermal stresses calculation. It includes geometrical information
(Layer drawing tool), external components' information (header...), stacking pattern information, 3D
thermal temperature map which have been calculated by ProSec3D, and the homogenization tool
Hompass. Therefore, Siteme is very helpful to validate BAHX mechanical strength for a given geometry
and operating condition. In-depth SiTEME studies could also lead to establish design rules which will be
used to continuously improve design of BAHX. Moreover, SiTEME gives high support for technical
expertises.
16
REFERENCES
1. ALPEMA. The standards of the Brazed Aluminium Plate-Fin Heat Exchanger Manufacturers'
Association. second edition. 2000. www.alpema.org.
2. Dynamic Simulation of Brazed Plate Fin Heat Exchanger. David Averous, Khaled Hammadi, Herv
Pingaud, Xavier Joulia and Philippe Guittard. s.l. : Computers and Chemical Engineering
Supplement , 1999.
3. ProSec : Modelling and Simulation in 3D of Brazed Aluminium Plate-Fin Heat Exchangers Core-inDrum. Florian Picard, David Averous, Xavier Joulia, Denis Barreteau. [ed.] Elsevier. Salvador :
s.n., 2009. 10th International Symposium on Process Systems Engineering - PSE2009.
4. Thermo-Mechanical Design of Brazed Plate-Fins Heat Exchanger Based On Finite Element
Modeling Using Homogenization Techniques. Johan Dib, Ivan Lewon, Boris Martin (Fives Cryo).
Prague, Czech Republic : s.n., July 26-30, 2009,. 2009 ASME Pressure Vessels and Piping Division
Conference. PVP2009-77251.
17