A Research Report Submitted by:
CHE 352
Spring Semester, 2016
Chemical Engineering Program
Abstract
In the experiment, many measurements were taken using the heat exchanger to observe the change in heat transfer from steam to water with respect to a change in water flow rate while trying to keep the inlet temperatures of water and steam constant. The flow rate was monitored using Brooks shorate meter and adjusted with valves. The flow rate has been veried from 1 bsm to 15 bsm. Two different control points were choose the measure the temperature of water and steam: the first one in inlet and the second one in outlet of a 0.19 inch pipe or 0.8225 inch tube. From obtained data it was noticeable that as flow rate increased, the heat transfer has been increased due to friction factor. The heat transfer coefficient has also been calculated using the experimental data and theoretical equations as well as two assumptions.
Introduction
The heat exchangers are wildly used as the energy transfer to ensure heating or cooling of different process fluids. The most common type of heat exchanger is a shel-and-tube exchanger, which could be used for closed circuit cooling of electrical equipment using demineralised water and for cooling water soluble oil solutions in quenching tanks (Bowman). The type of process fluid is chosen according to the final application requirements.
As fluid flows through a tube a heat transfer occurs due to the temperature gradient. It is best characterized by the heat transfer coefficient. The heat transfer depends on the materials used for tube design, the nature of fluids and the flow rates.
For the efficient and/or sufficient heat transfer for a given shell-and-tube exchanger it is important to understand the correlation between the heat transfer coefficient and flow rates. This experiment provides an opportunity to lay hands on the determination of such correlation and to practice technical description of the physical nature of heat exchanger. The obtained results are then compared to the predictive models to validate them.
Theoretical Background
The shall-and-tube heat exchanger design
The shall-and-tube heat exchanger vary in design (Figure 1-2). It is possible to make it straight or to shape it as a U- tube. Both types are commonly used. For example, U- shape is used in large heat steam generators in a nuclear power plant, while surface condensers in power plants are often 1-pass straight-tube heat exchangers.
Figure 1 U-Tube heat exchanger
Figure 2 Straight-Tube heat exchanger
The heat exchanges calculations
The basic heat exchanger equation (Kakac, 2002; Kuppan, 2000)
depict the relationship between heat transfer rate and the log mean temperature difference
Q=U*A*ΔTm
where ΔTm - the log mean temperature difference (°C) (°F);
Q – heat transfer rate (kJ/h) (Btu\h);
A – heat transfer area (m²) (ft²)
U – overall heat transfer coefficient (kJ/h.m².°C) (Btu/hr.°F)
The following equation is used to calculate the log mean temperature difference
ΔTm=T1–t2–T2–t1lnT1–t2T2–t1
where T1 - inlet tube side fluid temperature (°C) (°F;
t1 - inlet shell side fluid temperature (°C) (°F;
T2 - outlet tube side fluid temperature (°C) (°F;
t2 - outlet shell side fluid temperature (°C) (°F.
When used as a design equation to calculate the overall heat transfer coefficient, the equation can be rearranged to become:
U=QA*ΔTm
In heat exchanger heat is transferred from one fluid to another, so heat transfer rate can be calculated using the specific heat, which defined as the amount of heat energy needed to raise 1 gram of a substance 1°C in temperature, or, the amount of energy needed to raise one pound of a substance 1°F in temperature.
Q = m.Cp(T2–T1)
where Q – heat energy (Joules) (Btu);
m – mass of the substance (kilograms) (pounds);
Cp – specific heat of the substance (J/kg°C) (Btu/pound/°F),
(T2–T1) – the change in temperature (°C) (°F)
Specific heat depends on the initial temperature of the of a substance.(Water) Still in small-scale variation it can be considered constant. For example, at the temperature ranging from 200 to 600°F the steam specific heat is equal to1.93 (J/kg°C) or 0.46 (Btu/pound/°F) (Gasses) or for water under 200°C equal to 4,19 (J/kg°C) or 1 (Btu/pound/°F).
For a counter-flow designed heat exchangers the general tendency for change in fluid temperature along the heat exchanger surface is presented bellow:
Figure 3 Fluid temperature along the counter-flow heat exchanger surface
Materials and Apparatus
For this experiment a straight shell-and-tube heat exchanger has been used together with Brooks shorate meter to measure the flow rates with a maximum of 15bsm (1,46 liter per minute). The flow rate has been set using vents.
The photos presented bellow depict the apparatus.
Figure 4 Heat exchanger
The apparatus was preinstalled and operated by the system of vents.
The water and steam have been used without special pretreatment, thus considered to be the common pipe water and steam obtained via heating of such water.
Procedure
Water and steam are going through a tubes and shell side. The temperature of the steam was set higher than that for the circulating water. As the condensing steam was used their should not be any concern about counter-current flow matter, as condensed water is supposed to go downstream according to the basic physical laws. The steam is always present on the shell outlet ensuring that the heat exchange process was all over the surface of this heat exchanger. Three minutes after the each flow rate has been set and experiment started the inlet and outlet temperatures have been measured.
The experiment has been divided on two sets (15 and 9 measurements each). This was done prove the reproducibility
Before starting the experiment, every PPE for safety precautions were applied.
Results and Discussion
The flow of a fluid in a straight circular tube into either laminar or turbulent flow. For this experiment it has been claimed to be laminar. The deviations in water and steam inlet temperatures have been observe. However, they were rather small and have not exceeded 2°F for both fluids. According to the present data the water temperature was more stable. This fact should be also remembered when discussing the data which happen to be out of the line. Still, this factor was not considered to have a great final effect on the resulting heat flow assessment except for a specific starting time frame.
The data of steam temperature loss versus water temperature rise has been analyzed. The full data analysis is presented in Appendix I).
The tendency was clear for both replications, the hotter steam was giving heat away to water Figure 5 through a tube wall made out of copper. Still the first measurements were out of line for both replications with different starting flow rate point. The potential reasons for this are turbulent nature of the flow or the potential overheat when starting the heater and will be discussed later. At the beginning it also possible that the heat transfer through the metal transferring walls was not that much effective due to the low initial temperature, so it is highly possible that heat loss for heating of a copper walls influenced the final results.
Figure 5 Temperature change ratio
According to the graph the more or less stable date were obtained when changes in the temperature of water were less or around 40°F, or higher than 10°F for the temperature of steam.
The heat transfer area was calculated using a general mathematical equation applicable for cylinders:
A=2*π*Din*l
where Din – an innere dimeter of a tube
l – length of a heat exchanger
A=2*π*0.19*20.375=24,32inch2
Using the heat transfer are number and the principal equations for a heat exchanger it the heat coefficient has been calculated (Appendix I). The experimental heat transfer coefficient dependence on the water flow rate is presented bellow.
Figure 6 Heat transfer coefficient dependence on the water flow
The first and second replication showed the same tendency for heat transfer coefficient dependence on the water flow. The difference at the low flow rates is subjected to the start and the experiment and the apparatus bias. It is clearly seen that for every replication the process at the beginning was unstable, the steam was probably overheated during first minutes, so in the end the observed inner and outlet temperatures for water and steam has indicated simultaneous heating of both fluids. (Table. 1) this indicate the need either to wait until the apparatus starts to work in a regular mode (the fluid turbulent flow at the beginning could be than updated to a regular laminar flow) or to tune up the heating proses for steam. The former assumption is favorable, as the lower flow rates were less effected: the only one measurement was out of line for the first replication started with the rates under 0.10 liter per minute, while two have fallen out for the second replication started at the flow rates about 0.20 liter per minute.
Though this measurements were not accurate their results supports the main what makes the former assumption concerning the change from turbulent to laminar flow be close to the reality.
For the relationship between the flow rate and heat transfer coefficient the assessment the first measurements were not included. The equations were proposed using basic MS Excel software.
The linear equation has fitted here, but according to solely mathematical calculations it is not the best one. The polynomic assumptions was the most accurate for the presented data.
In order two calculate the theoretical heat transfer coefficient the following equation can be used:
General
U=1Din+dwall*1αoutDout+12λCulnDoutDin+1αinDin+Rdust
For a thin wall as Dout/Din<2
U=11αout+dwallλCu+1αin+Rdust
As the experiment design has allowed only to estimate evaluate the heat transfer via the temperature, to perform the theoretical calculations two assumptions have been made change are made:
1) the resistance of the copper tube is negligible and
2) the resistance of the condensing steam can also be ignored.
Thus the former equations are:
General
U=1Din+dwall*1αoutDout+1αinDin
For a thin wall as Dout/Din<2
U=11αout+1αin
For future mass flow calculation the cross-sectional cuts has been evaluated:
Fro the shall
A=π*Dout-dwall24-π*Din+dwall24
A=π*0.8225-0.24324-π*0.19+0.24324=0,12 inch2
Fro the tube
A=π*Din24
A=π*0.1924=0,03 inch2
This data can be used to modulate the heat flow change due to the mass flow change.
The mass flow in the discussed experiments was changed using the change in follow rates.
Conclusion
The presented experimental data has showed that the laboratory heat exchanger was functional despite small deviations at the heating period. The experimental data have clearly supported theoretical knowledge.
References
Sloley, A. Shell-and-Tube Heat Exchanger: Pick the Right Side. Chemical Processing. Oct 23, 2013 http://www.chemicalprocessing.com/articles/2013/shell-and-tube-heat-exchanger-best-practices/ (accessed Feb 11, 2016
Shell and Tube Heat Exchanger, Bowman http://www.ejbowman.co.uk/products/Shelland TubeHeatExchangers.htm (accessed Feb 11, 2016).
Gases - Specific Heats and Individual Gas Constants . http://www.engineeringtoolbox.com/specific-heat-capacity-gases-d_159.html (accessed Feb 11, 2016).
Water Vapor - Specific Heat Specific heat of Water Vapor - H2O - at temperatures ranging 175 - 6000 K http://www.engineeringtoolbox.com/water-vapor-d_979.html (accessed Feb 11, 2016).
. Kakac, S. and Liu, H., Heat Exchangers: Selection, Rating and Thermal Design, CRC Press, 2002.
Kuppan, T., Heat Exchanger Design Handbook, CRC Press, 2000.
Appendix I
