Heat Exchanger Process Design
Heat Exchanger Process Design
JACOBS H&G
DOCUMENT NO: P WI 05
Rev No. Issue Date Pages Rev Description Prepared Checked Approved
By By By
0 30th June 2004 12 Issue for Use SPP RGP VVD
Att. 16
JACOBS H&G LTD
TABLE OF CONTENTS
1.0 PURPOSE
2.0 SCOPE
3.0 DEFINITIONS
4.1 GENERAL
4.2 CONTENT
6.0 RECORDS
7.1 STANDARD PROCESS DATASHEET FOR SHELL & TUBE HEAT EXCHANGER
7.3 SAMPLE PROCESS DATASHEET FOR SHELL & TUBE HEAT EXCHANGER.
/conversion/tmp/activity_task_scratch/850398962.doc
JACOBS H&G LTD
1.0 PURPOSE
The purpose of this procedure is to define the workflow, format and procedure for Shell & Tube
Heat Exchanger Design using HTRI Package.
4 2.0 SCOPE
This procedure applies to all Shell & Tube Heat Exchanger designs performed by the Process
Department.
3.0 DEFINITIONS
None.
4.1 GENERAL
It is the responsibility of the individual engineer to design a Shell & Tube heat exchanger in
accordance with this procedure.
Input flow rate, temperature & vapour fraction at inlet / outlet conditions and the allowable
pressure drop for shell & tube side.
For liquids, vapor fraction is “0”; for gas it is “1” and for two phase it is between 0 & 1.
4.2.2 Geometry
TEMA type As given in data sheet (if not mentioned, then it shall be decided
based on type of fluid, condition etc.)
Hot fluid side Hot fluid shall be either on the shell side or on the tube side
Tube type Plain or finned (for shell & tube generally plain tubes are used)
Tube length In design mode, enter the length & design the exchanger for various
shell ID’s. Standard tube lengths available in FPS units are 4, 6, 8,
10, 12, 14, 16, 18, 20, 22 and 24 feet. In MKS unit, tube lengths can
be used in steps of 500 mm. Sometimes, it may be required to use
FPS standard tube lengths in MKS unit (for e.g. 6096 mm, 3048
mm, etc.)
For small exchangers, a tube length of 1000 mm can be used.
Effective Tube length This is the value used for heat transfer and need not be entered.
HTRI calculates tube sheet thickness and also the effective tube
length.
/conversion/tmp/activity_task_scratch/850398962.doc
JACOBS H&G LTD
Tube Pitch Generally 1.25 times tube OD. Other values can also be specified.
Flow direction
Flow direction
Flow direction
/conversion/tmp/activity_task_scratch/850398962.doc
JACOBS H&G LTD
Tube material Tube MOC shall be selected from HTRI data bank.
Baffle type Various baffle types are Single, Double, NTIW and Rod. Baffle type
need not be specified for design case.
Baffle cut Baffle cut is specified with respect to shell inlet nozzle axis and can
be either vertical or horizontal (if the baffle cut is perpendicular to the
nozzle axis, then the cut is horizontal and if the cut is parallel to the
nozzle axis, then it is vertical).
Baffle cut need not be specified for design case.
Parallel pass lane Not required for design case (Refer HTRI help)
Perpendicular pass Not required for design case (Refer HTRI help)
lane
Sealing strips Not required for design case (Refer HTRI help)
Sealing strip reduces leakage in bundle and shell, which increases
the cross flow fraction.
If shell side nozzle inlet / outlet RV² is more than the allowable limit, then HTRI will consider
an impingement plate. For gas and two-phase flow, impingement plate is required. For liquid
phase, it depends on the value of RV². Generally, rectangular impingement plates are used for
exchanger.
There is some optional data, which is not required for design purpose. However, this data
should be corrected at the time of rating and fine-tuning, which is given below. Total tube sheet
thickness, floating head support plate, support at U-bend.
4.2.3 PIPING
/conversion/tmp/activity_task_scratch/850398962.doc
JACOBS H&G LTD
/conversion/tmp/activity_task_scratch/850398962.doc
JACOBS H&G LTD
4.2.4 Process
Fouling resistance Enter the value mentioned in data sheet. If this value is not available
in the datasheet, then the same should be taken from published
literature, reference books, like ‘Process Heat Transfer’ by D.Q. Kern.
Flash type Differential (separate and not in contact) or integral (well mixed and in
thermal and chemical equilibrium)
4.2.6 Grid Design
Geometry
a. Shell ID Enter the minimum and maximum shell ID and either the number of
steps or the step size in mm or inch
b. Baffle spacing Enter the minimum and maximum baffle spacing and either the
number of steps or the step size in mm or inch
c. Tube passes Enter the minimum and maximum number of passes and ‘odd’ or
‘even’ passes.
d. Tube length Enter the minimum and maximum tube length and either the number
of steps or the step size in mm or inch
e. Pitch ratio Enter the minimum and maximum pitch ratio and either the number of
steps or the step size in mm or inch
f. Tube diameter Enter the minimum and maximum tube diameter and either the
number of steps or the step size in mm or inch
g. Shell type Select any one of:
TEMA ‘E’, ‘F’, ‘G’, ‘H’, ‘J21’, ‘X’, ‘K’ type shell
h. Baffle type Enter any one from the following options:
Single
Double
No tubes in window (NTIW)
Rod
None
HTRI gives various designs with different shell ID with optimum baffle spacing for given tube
length and tube passes.
HTRI gives shell ID in standard inch format. It has to be fine tuned to the nearest round number
that is divisible by 5. This can be done by putting the programme on ‘rating’ mode.
/conversion/tmp/activity_task_scratch/850398962.doc
JACOBS H&G LTD
Constraints Vary hot fluid velocity, cold fluid velocity, pressure drop etc.
Options Generally we are not changing these options .If required we can vary
all options as given in the HTRI. i.e. Allowed over design range,
Baffle spacing options, Tube pass options etc.
Warnings If you required additional warnings other than HTRI’s standard
warnings, then we can specify here. i.e. Warnings for hot stream or
cold stream tube wall temperature, Allowed critical velocity ration,
Allowed vibration frequency ratio etc.
4.2.7 Control
Name Enter details like name of the Client, exchanger tag no., project
number, etc.
Case Give the case description (design case, rating case, with __ length,
__ type shell, etc.)
Methods, safety, user defined, vibration, F&J curve etc. are not required for simple design.
4.2.8 Once all the above data listed in sections 4.2.1 through 4.2.6 is entered, the programme shall
be run by clicking the ‘run’ option on the tool bar. The programme gives various designs
(output) for data entered. A typical design table is given below:
/conversion/tmp/activity_task_scratch/850398962.doc
JACOBS H&G LTD
All designs shall be analysed and evaluated for different factors like % over design, pressure
drop, number of shell(s) in series, number of shell(s) in parallel, area, heat transfer coefficient,
velocity, flow fractions, etc.
The best case shall be selected for rating. This shall be done by saving the file as ‘rating’ file.
Any modifications required shall be done and the file shall be re-run to get the final/desired
results.
Note: Generally, a good design shall have an over design range between 5-10% with pressure
drops on shell and tube side utilised to the extent possible. The final design shall also have B-
stream fraction greater (or at least equal to) 0.4. The different leakage streams
encountered/present and ways to reduce the same have been described below:
a. B-stream flow fraction shall be increased and the leakage stream flow fractions shall be
minimised. In actual practice, it is not possible or necessarily desirable to eliminate leakage
streams completely, as clearances are required to construct the heat exchanger. B-stream
fraction should be typically more than 40% (lower values are tolerated for viscous fluids
and shell side condensing fluids). Increasing the baffle cut and reducing the number of
baffles can increase B stream fraction.
b. A-stream is generated due to the clearance between the tubes & baffle, and hence the
/conversion/tmp/activity_task_scratch/850398962.doc
JACOBS H&G LTD
c. C-stream results due to the bundle to shell clearance. This stream can be reduced by using
sealing strips. The value of C-stream is high especially for floating head exchangers. HTRI
calculates the number of sealing strip pairs, but as a general guideline, one pair of sealing
strips for every 5 tube rows in overlap region can be specified.
d. E-stream is not thermally effective; hence a large E-stream has a large negative impact on
the exchanger’s thermal performance. E-stream is created because of baffle to shell
clearance. HTRI uses TEMA standard to calculate this clearance. This clearance can be
varied with the help of the fabricator.
e. The F-stream travels along tube pass partition lanes. As these bypass streams can
affect heat transfer and pressure drop performance, they must be modelled accurately.
Hence, a case shall never be run with this stream “minimise”. HTRI blocks this stream by
using a seal rod having OD equal to or less than the tube OD. We should not change or
input bundle to shell clearance or other clearances manually. HTRI calculates all
clearances on the basis of TEMA shell type.
4.2.10 Tips to decide optimum and good design of Shell & Tube heat exchanger
b. When specifying the number of baffles (or the number of crosspasses), ensure that shell
side inlet / outlet nozzle locations relative to each other are correct. For e.g., if the shell
side nozzles are on opposite sides, then the number of baffle crosspasses should be an
odd figure or the number of baffles should be an even figure.
c. Generally, for horizontal exchangers, shell side nozzles are on opposite sides. For vertical
exchangers, shell side nozzles can be either on opposite sides or on the same side.
d. Inlet / outlet baffle spacing is dependent on shell inlet / outlet nozzle ID. Baffle should not
foul with the inlet / outlet nozzle. Generally, inlet / outlet spacing is equal to or more than
the central baffle spacing.
Try to set a baffle spacing that is divisible by 5 in MKS unit or 1 in FPS units.
Compare % contribution of pressure drop in nozzle versus pressure drop due to cross flow.
If DP in nozzle is high, then go for a higher nozzle size.
e. If the shell side coefficient is very low, it can be increased by using a finned tube. However,
it must be ensured in such cases that the shell side fluid is clean.
f. If the tube side coefficient is governing, then check whether the entire pressure drop (DP)
available has been utilised or not. If the entire DP has not been used, then increase the
tube length and reduce the number of tubes such that the heat transfer area remains the
same. If the calculated DP is less, increase the number of tube passes.
A smaller tube OD can also be opted for; however the case has to be re-run in ‘design’
/conversion/tmp/activity_task_scratch/850398962.doc
JACOBS H&G LTD
g. For single phase on shell side, the baffle cut can be either horizontal or vertical.
For two-phase flow on shell side, vertical cut baffles are preferred. However, horizontal cut
baffles may be used with cutout at the bottom.
h. Nozzle orientation shall be such that the 1st tube pass has counter-current flow.
j. Nozzle location shall be specified (either at U-bend, or before U-bend or after U-bend) for a
U-tube exchanger.
k. Specify shell inlet nozzle location (options are programme decides, top, bottom or side).
l. NTIW (no tubes in window) type baffle eliminates vibration problem but increases the shell
diameter. When specifying NTIW design, care should be taken to keep the window velocity
value within limits as per HTRI guidelines.
Vibrations can also be eliminated/reduced by using intermediate support in between two
baffles. As a general rule, intermediate support cannot be used for single or double
segmented baffle.
m. For U-tube exchangers, the bend radius is normally 2 times the tube OD (HTRI takes 1.5
times the tube OD as default). To make this change enter pass lane width manually
equivalent to 3 times tube OD.
n. If there are more shells in series, then at the time of rating, shell side inlet / outlet nozzle
location shall be entered for each shell (i.e. at top or bottom, at front end or rear end).
4.2.11 Procedure for preparing Shell & Tube heat exchanger process data sheet
First copy standard shell and tube heat exchanger process data sheet from standard format.
a. Copy rows 9 to 58 from TEMA-1 Sheet and paste only the values at location given in
standard data sheet (Sheet 2).
b. Copy sketch of individual shells from individual sheets and connect to each other (sheet 3).
Check correctness of inlet and outlet nozzle locations for shell side and tube side. This
sheet is required when there is more than one shell in series or parallel.
If the design pressure, design temperature values and the material of construction are
different for different shells, then specify accordingly in sheet 3.
c. Read the following points when updating a standard process data sheet.
Marked cells to be checked / rewritten / deleted as per requirement.
Please note that for fixed tube-sheet exchanger mean metal temperature to be given for
various conditions as in sheet 4.
Copy tube layout from HTRI output on sheet 6.
When shells in series are more than one, then we are supposed to give detailed
interconnection drawing on separate sheet (sheet 3).
When shells in series are more than one and MOC for tubes and shells are different for
different shells then give table for MOC (sheet 3).
/conversion/tmp/activity_task_scratch/850398962.doc
JACOBS H&G LTD
There is no provision in HTRI to specify whether exchanger is IBR approved or not. This is
to be included under remarks.
d. Mean Metal Temperature: This is required for fixed tube exchangers only. If the number of
shells in series is more than one, then mean metal temperature is to be specified for each
shell. Specify nozzle schedule in sheet 4.
f. Copy tube layout from HTRI output (drawing – tube layout) and paste in sheet 6.
Remove shell side nozzle ID from tube layout. The nozzle sizes shall be specified in the
nozzle schedule on sheet 4.
None.
6.0 RECORDS
Heat exchanger calculations are quality records and must be filed as per standard procedure.
/conversion/tmp/activity_task_scratch/850398962.doc
JACOBS H&G LTD
ATTACHMENTS
/conversion/tmp/activity_task_scratch/850398962.doc