Being the only equipment that deals with two fluids with different properties (one on the shell side and the other on the tube side) – rubbing shoulders and yet not being in contact – shell-and-tube heat exchangers are the most difficult exchangers to operate.
By Haresh Sippy, MD & Chief Founder of Tema India Ltd.
Although the expertise lies in the hands of the licensor of the processing plant, and the process engineering, thermal and mechanical engineering on the EPC, the actual design concept involves probing in the dark and thrives on substantial safety factors and software. It is so because, in reality, there are several unaccounted variables in the manufacturing of customised equipment. More so for non-standard and proprietary products.
The design of a shell-and-tube heat exchanger
1) Properties of fluids
To work on the thermal design of the heat exchanger, the following properties of the two fluids under heat transfer are required: density, specific heat, thermal conductivity and viscosity. The right way is to get the values of these four parameters at different temperatures in the form of a heating or cooling curve for the application. The more we understand the physical properties of the fluids in question, the more precise the design of the heat exchanger will be.
2) Selection of exchanger construction and allocation of fluids
The very first and most important thing to determine is the allocation of the two fluids to the shell side and the tube side. Generally, the hotter, high-pressure, corrosive fluid is assigned to the tube side of the exchanger. Other factors such as the phase of fluid, allowable pressure drop, fluid velocity and heat transfer coefficient also influence this decision. Thereafter, the designer selects the type (fixed, floating or U-tube construction) of the heat exchanger by considering factors such as thermal expansion stresses and the need for bundle cleaning from the shell side. Later, the design temperature, pressure and maximum permissible pressure drop are established for the exchanger.
3) The energy balance Heat lost = Heat gained
Once the physical properties have been correctly defined, it is time to define the flow rate, and in and out temperatures. With values of two of the above parameters known, the third parameter is computed. The designer then chooses the geometry of the heat exchanger.
Tube OD x Thk x Length x Qty x Pitch and Pattern
Once the tube layout is finalised, the outer tube layout diameter and the shell diameter are determined. Based on the tube geometry selected, the heat transfer area can be calculated.
Q = Heat load (W)
M = Mass flow rate (kg/h)
c = Specific heat capacity
T1, T2 = Inlet and outlet temperatures of shell side t1, t2 = Inlet and outlet temperatures of tube side
A = Q / [U x LMTD]
A = Required heat transfer area (m2)
Q = Heat duty (W)
U = Overall heat transfer coefficient (W/m2K) LMTD = Log mean temperature difference (°C) [The average logarithmic temperature difference between the shell-side and the tube-side fluids over the heat exchanger length]
4) Thermal calculation
At this stage, the design engineer performs a thermal calculation. The objective is to obtain the shell-and-tube side heat-transfer coefficients. With the coefficients known, the overall heat-transfer coefficient can be calculated. Knowing this value, it becomes possible to calculate the total heat transfer area required.
Another important parameter is the pressure drop, which is calculated for the shell-side and tube-side fluids. The pressure drop is a function of the Reynolds number, the type of flow (turbulent or laminar flow) and the roughness value of the shell and inner tubes. The calculated required area is compared with the area provided and a check is made to see if the pressure drops are within the design limits. If the calculated required area exceeds the provided area, the geometry of the heat exchanger needs to be redesigned, possibly by increasing the tube length or the tube count. Likewise, if the calculated pressure drop exceeds the maximum defined, a new geometry must be designed to ensure a pressure drop reduction. The trial-and-error process is continued until a satisfactory design with suitable geometry is obtained.
5) Mechanical design calculations
With the heat exchanger geometry defined, the mechanical design calculations are performed as per the specified code of construction (such as ASME Sec VIII Div-1, 2) to determine the optimum thicknesses of each component of the exchanger for the applicable design conditions – pressure, temperature and corrosion allowance. FEA is used to qualify the design for non-standard constructions.
6) The manufacturing drawings
With all design dimensions of the heat exchanger defined, the manufacturing drawings can be prepared. The accuracy of these drawings is vital to ensure that no design issues –such as fouling – that can delay the project are discovered during manufacturing. Manufacturing drawing generation using 3D-based CAD software identifies and corrects fouling and other issues. An equipment manufacturer must ensure the material, size and quantity of all components specified in the Material Take Offs (issued for procurement of material) match the manufacturing drawings. This process is vulnerable to human error, particularly when revisions happen in the design and manufacturing drawings.
Codes focus only on the safety of external pressure boundary
The governing codes of construction are very particular about the equipment’s pressure boundary not bursting under pressure (i.e. safety of externals). However, these codes do not specify any rules for the construction of internals (being non-pressure parts) that determine the performance of the equipment. In the case of a shell-and-tube heat exchanger, its thermal and hydraulic performance is mainly dependent on the accuracy of its tube bundle. Apart from the tubes, baffles are vital in facilitating efficient heat transfer between the two fluids.
The shell side flow over the tube bundle is guided by the transverse baffles. Inaccuracies in baffle manufacturing can significantly compromise the effective heat transfer by allowing higher bypass/leakage streams between the tube-to-baffle hole and baffle-to-shell (refer to sketch below). Poorly finished baffle holes can cause scratch marks on the tube OD, during tube insertion and when the tubes move through the baffle holes due to thermal expansion during operation. These scratch marks can lead to failure of the tubes in a few years of continual service, especially in high pressure/temperature operating conditions. Hence, utmost care is needed in their manufacturing and assembly. However, baffles being a non-pressure internal part, their design and manufacturing requirements are not controlled by the codes. And so, the performance of a shell-and-tube heat exchanger is not addressed by the codes.
Baffles are baffling
The baffles in a heat exchanger are baffling simply because there is a limitless variety to choose from, and you can never be sure. Even when you are close to selecting the right one, the outcome will fluctuate according to circumstances such as manner, time and place. Nevertheless, having made a choice, what delivers results is workmanship. Refer to the following link for the lesson we learnt with helix baffles: https://bit.ly/3u0vGJW ( LinkedIn Article )
Despite being the most crucial item (after the tubes) in the entire Bill of Materials (BOM) of a shell-and-tube heat exchanger, not much thought is given to baffles in fabrication drawing. This happens primarily because, being a non-pressure part, most manufacturers outsource the work of baffle drilling and machining, and the quality-control inspection agencies do not give it the required importance. All pressure parts such as shells, tubes and tube-to-tubesheet joints are produced with stringent code requirements and are often the witness/hold points. All joints are subjected to NDT, and finally, helium/pneumatic/hydro-tested.
As a result, failures in the tube bundle – the core of the exchanger – are common. The most expensive part of the exchanger which can cost over half of a complete unit has to be periodically replaced. Tema India has developed standards for baffle-hole drilling and OD machining to substantially enhance the life of the tube bundle.
The selection of appropriate material for baffles is important. Incorrect material selection can lead to the following problems.
- In an electrolytic shell-side fluid, the baffle may form an electrolytic cell with the tubes and baffle holes may get corroded, leading to bypass of fluid through this clearance. This increases the possibility of autofrettage wear of tubes during vibration.
- The baffle hole may expand much more than the tube, leaving it unsupported and increasing the bypass of fluid through this clearance.
Interference of specifications in manufacturing
To reduce the risks in the procurement of critical equipment from less experienced equipment manufacturers, customers and EPC companies specify additional details in their datasheet and specifications which should otherwise be engineered by the equipment manufacturer. Of late, process licensors/customers have been interfering in the manufacturing methods of the tube bundles by insisting on specific constructional details that compromise the bundle instead of improving it. The following are some examples.
1. Insisting on a very small baffle-hole-to-tube OD clearance (regardless of tube OD, tube qty., baffle spacing). This creates hindrances for the insertion of tubes in the baffle; particularly, in cases where the baffle spacing is less and thereby inducing stresses in tubes.
2. Insisting on all-around welding of sealing strips/sealing rods with baffles as well as transverse baffles with the longitudinal baffle. Performing such unwarranted long welds at places where shorter weld lengths (~1in) are adequate, causes distortion of tube holes in the baffle and compromises its assembly with the tubes. Also, the resulting tight fit between tubes and baffles restricts the free movement of the tubes through these distorted baffle holes during operation and induces excessive stress in the tubes. Therefore, such construction causes more harm to the life of the bundle than improving it.
3. Insisting on specific dimensions for sliding strips/bundle skid bars or a specific number of tie rods makes the tube bundle more rigid and less maintenance-friendly as it takes more time for bundle insertion/removal from the shell.
4. Insisting on providing 3 x 45° chamfers on both sides tube holes in the baffles. Providing such large chamfers leads to the inside surface of the hole acting as a knife-edge to the tube and may lead to an autofrettage wear of the tubes. Removing sharp corners of the hole (i.e. deburring without insisting on a specific chamfer dimension) is enough to ensure tubes are not scratched during insertion.
Therefore, it is in the best interest of the life of the bundle that the construction of the bundle and its manufacturing process be left to the manufacturer to decide. Specifications will allow the required freedom for the equipment manufacturer to be able to provide designs that are more effective, manufacturing-friendly and economical.
The end-user and the end-maker
The end-user has first-hand experience of the performance and the reliability of maintenance; the most crucial feedback for the manufacturer. In its absence, we at Tema India believe that we have ownership of the equipment we sell. This automatically gives us an overview of the problems encountered in maintenance and the improvements that follow. The main lesson learnt is that the product’s quality should be beyond the code specifications. This varies from type to type if the equipment has to be built for trouble-free service throughout its lifecycle, with maximum efficiency and ease of maintenance.
How does this happen? With important quality parts such as baffles made in-house to the most stringent specifications, the entire tube bundle is produced as per the critical standards established by the manufacturer. The manufacturer assumes the dual responsibilities of checking out on the maintenance side and making changes in the design to accommodate the requirements of the maintenance team. Moreover, the manufacturing sequence is to be adopted and changes made according to the type of equipment. For example, a feedwater heater with a condenser inbuilt is prone to leakages that drastically reduce efficiency. The other example is that of a Hi-Hi screw-plug exchanger with internal leaks in the inner-shell-to-tubesheet joint. This not only reduces the efficiency but also allows the sulphur-dirty fluid to leak into the clean fluid.
Criteria for lowering cost and increasing efficiency
This being the end goal, the manufacturer must aim for:
- Optimisation of the size of the equipment
- Ease of procurement of raw materials
- Ease of manufacturing in terms of time taken and quality attained
- Ease of installation and maintenance at the user’s end
To do this, the manufacturer’s involvement must begin from the very onset of the project when the process parameters are defined. These include operating & design temperature and pressure, heat duty (total heat transfer rate), inlet/outlet temperature for hot fluid and cold fluid, maximum allowable pressure drops on the hot and the cold side, and the type of construction (as per Tema India standards or proprietary). Finally, the most critical parameter is the selection of fluids concerning the shell side and the tube side. The manufacturer plays an essential role in selecting the shell side or the tube side as hot or cold.
For example, in a Hi-Hi screw-plug construction, the side that is at a higher pressure is preferably the tube side. Correct stream allocation simplifies the design, manufacturing, operation and maintenance of the heat exchanger. How to achieve the end goal in the process design of a shell-and-tube exchanger is, therefore, an open-ended question. There are many more variables to choose from when setting the parameters. This calls for a trial-and-error calculation, such as iteration with the help of computational fluid dynamics or CFD.
About the author
Haresh Sippy – Founder & CMD Tema India Ltd – is an engineering graduate with over 40 years of experience in the field of Shell & Tube Heat Exchangers. He holds a patent for Shrink Ring technology in Screw Plug Heat Exchangers used in high-pressure and high-temperature oil & gas applications. Above all, he is a constant innovator with several patents pending.