A Series-Parallel Heat Exchanger Experiment

A Series-Parallel Heat Exchanger Experiment KAREN A. FLACK Department of Mechanical Engineering United States Naval Academy

RALPH J. VOLINO Department of Mechanical Engineering United States Naval Academy

ABSTRACT An experimental apparatus has been designed to test the use of cross flow heat exchangers. Three identical fin-tube type cross flow heat exchangers are mounted on a board instrumented with thermocouples, flow meters, and a pressure transducer. The apparatus can be set to test the performance of a solo heat exchanger, two or three heat exchangers in series or parallel, or combinations incorporating both series and parallel configurations. The apparatus is relatively simple, inexpensive, and versatile. It may be used in a variety of configurations for several types of student laboratories ranging from demonstrations to design projects. The use of an apparatus such as this gives students hands on experience with experimental procedures and helps them to gain a physical understanding of heat transfer phenomena.

I. INTRODUCTION Most undergraduate heat transfer courses cover a large amount of material in a relatively short period of time. Students are presented with a barrage of new concepts and empirical correlations, and spend most of their time learning to apply them correctly to textbook problems. In the rush to cover the necessary material, care must be taken to insure that the students develop a physical understanding of the heat transfer phenomena they are modeling. The physical concepts are not difficult in most cases, but can become lost in the problem solving process. Students should also develop an appreciation for the meaning of quantitative answers. A solution of 1 kW or 1 MW may not mean much to a student unless it can be related with confidence to something familiar, such as the radiator in their car or the condenser on their refrigerator. Students should also develop the ability to apply the concepts learned in class for relatively simple geometries (flat plates, straight round tubes, etc.) to more complex situations (e.g. heat exchangers). In doing this they should understand the approximations they are making and the ways these approximations may influence their results. Physical tests to check these approximations provide a method to boost confidence and catch mistakes. To improve heat transfer education, students should be exposed January 1999

to more open-ended problems and physical experiments. The objective of the present work is to develop and present an experimental heat exchanger apparatus for use in undergraduate courses. The experiment is designed for flexibility with a number of possible configurations for fluid mechanics and heat transfer experiments. The apparatus is portable, allowing classroom demonstrations. It may also be modified easily by students, for use in design problems. The apparatus is also well instrumented and laid out so that students can see qualitatively and measure quantitatively what is happening to the various flows in their experiments.

II. EXPERIMENTAL APPARATUS Three identical, copper fin-tube, cross flow heat exchangers are mounted on a board as shown in figure 1. The heat exchanger dimensions are 12.7 cm x 12.7 cm x 2.9 cm. Each compact heat exchanger has 6 passes of 7.9-mm diameter tube through 85 fins. Hot water, heated to a controlled temperature by a resistance coil in a 25-liter tank, is pumped through 7.9 mm ID Tygon™ tubing which connects the tubes of the heat exchangers. The pump supplies a flow rate ranging from approximately 0.02 to 0.04 kg/s through the heat exchangers, depending on the experimental configuration. Flow meters are located at the inlet of each heat exchanger. A pressure transducer measures the pressure drop across heat exchanger #3. Air is forced over the fins at a constant flow rate using small electric cooling fans that are attached to the front of the heat exchangers. The air flow rate is calculated from an energy balance using the measured inlet and outlet temperatures of water and air and the water flow rate. The apparatus is instrumented with thermocouples on the water side at the inlet and outlet of each heat exchanger. Additionally, a thermocouple is located on the exit side of the fins to determine the air outlet temperature. This thermocouple can be traversed across the heat exchanger in order to better estimate the bulk temperature of the air. The inlet air temperature is assumed to be the ambient room temperature. The thermocouples are connected to digital voltmeters which, using electronic reference points and thermocouple calibrations, display the measured temperatures in degrees C. Ball valves control the flow of water through the system (figure 1). The apparatus can be set to test the performance of each heat exchanger operating alone, two or three heat exchangers in series or parallel, or combinations incorporating other series and parallel configurations. A total of thirteen different configurations can be tested. The apparatus can produce a large range of conditions, including a three-fold increase in temperature drop with flow rate and a fourfold increase in temperature drop based on flow configuration. The heat exchangers, tubing and instrumentation are mounted on a 1.5 m x 1 m vertical wooden board. All temperatures, pressure Journal of Engineering Education 27

transducer voltages and flow rates can be read directly from the front of the board. Thermocouple wires and electrical connections are on the back of the board. Electrical power is supplied though a single power strip mounted on the back of the board. The board is mounted on a cart along with the heating tank and pump. The entire apparatus has dimensions 1.5 m x 0.8 m (1.8 m in height). The apparatus may be easily moved into any classroom and plugged into a standard wall outlet. The experimental heat transfer apparatus was constructed in house using relatively inexpensive, readily available materials. Total cost for materials and instrumentation was approximately $3,000. The apparatus was designed and built by two high school students who were participating in a summer intern program at the Naval Academy. It is simple to operate and versatile due to the variety of configurations available. Although this project was completed by high school students, it would also be well suited as a design project in an undergraduate course.

III. LAB POSSIBILITIES Several heat transfer laboratories may be used to illustrate concepts associated with heat exchangers and internal flow predictions. Students could be asked to predict the overall heat transfer coefficient for a heat exchanger given the flow rates of water and air. Since the geometry of the heat exchangers is fairly complex, several assumptions and approximations would be necessary. The water flow may be approximated as flow through a straight pipe, ignoring the effect of the bends. The air may be treated as flow over flat plates, or as channel flow with a significant entry region effect. Alternatively, students could consult a text on compact heat ex-

changers1 to find a better estimate of the external heat transfer coefficient. Once predictions are made, the apparatus can be used to measure inlet and outlet temperatures. From the temperatures and mass flow rates, the heat transfer coefficient can be determined using the Log Mean Temperature Difference (LMTD) method. A second experiment might involve prediction of exit conditions from the heat exchangers given the mass flow rates of air and water and the inlet temperatures. If heat transfer coefficients are given, the calculation is a straightforward application of the Effectiveness Number of Transfer Units (e-NTU) method of heat exchanger performance prediction, as found in undergraduate heat transfer texts .2 Comparisons between prediction and experiment will be good, although some differences will occur due to uncertainty in the heat transfer coefficients and the measured temperatures. If the students are asked to analyze a combination of heat exchangers in series or parallel configurations, there will also be heat losses from the Tygon™™ tubing that may or may not be taken into account. The differences provide good opportunities for discussion of uncertainties in the experiments and simplifying assumptions in the predictions. Moving beyond prediction of specified cases, design problems are also possible. Students may be asked to design a system that will produce a specific outlet temperature, or a specific outlet temperature and mass flow rate. Once the design is complete, it can be verified experimentally. In all of the above laboratories, the connection between the predictions and experiments is key. The goal is to make the link between the theory covered in lecture and the physical experiment as clear as possible. In doing this, ample time should be allowed at the completion of the experiment for return to the predictions and reanalysis in light of the experimental results. First the significance of

Figure 1. Schematic of the Heat Exchanger Board 28

Journal of Engineering Education

January 1999

differences should be considered with respect to experimental uncertainty. Next, the students should be asked to determine and demonstrate quantitatively the impact of various refinements of their models on improving their predictions. A. Example Lab In the first use of the apparatus, students were given the temperature of the water in the heating tank, the room temperature, and estimated flow rates through the heat exchangers for various configurations. Water and air side heat transfer coefficients were also supplied, although these could have been left to the student to determine. The students were asked to predict water and air exit temperatures for several configurations. Table 1 shows some typical results where a, w, i, and o indicate air, water, inlet and outlet, respectively. The students in this case performed a short sensitivity study and noted that changing the water inlet temperature had a noticeable effect on the air exit temperature, but that changing the air inlet temperature had much less effect on the water exit temperature. This is expected since the water side has the higher heat capacity. The students noted that the predicted temperature drops were quite close to the experimental results. A few students (who had errors in their calculations) found larger differences between prediction and experiment, which they attributed to losses in the lines between heat exchangers. Such students could be directed to the experimental data (table 1), which shows an average Tw between heat exchangers of less than 0.1°C. The students could also be asked to estimate the heat loss from the Tygon™ tubing analytically. They would find that the dominant thermal resistance is due to the external free convection, with a secondary conduction resistance from the tube wall. A rough estimate shows a water temperature drop of about 0.05°C for a 1 m long tube.

IV. IMPACT ON STUDENTS’ UNDERSTANDING OF HEAT TRANSFER The primary purpose of this experimental apparatus is to demonstrate the principals of compact heat exchangers. Its use can be two-fold: (i) to give students a hands-on experience to gain an understanding of typical heat exchanger systems and (ii) to provide experiment data for comparison to theoretical calculations. The experiment demonstrates the importance of each mode of heat transfer in heat exchanger operation. Conduction and forced convection are the dominant modes of heat transfer in this type of heat exchanger, allowing for simple, fairly accurate computations. More

complex problems can be solved accounting for other modes of heat transfer including radiation (which accounts for less than 2% of the heat transfer from each heat exchanger) and natural convection. The ability to take physical measurements to compare to calculations tests the accuracy of assumptions made in the calculations. The comparison between prediction and experiment will depend strongly on the approximations and correlations used in the predictions. Any reasonable approximations will give results of the correct order of magnitude, but results will vary depending on the methods used. This should be illustrative to the students, showing both the utility and limitations of the theory they have learned. In the end, the students should see the utility of very simple models for making useful initial estimates, and their ability to develop more sophisticated models for making more accurate quantitative calculations. The flexibility of the apparatus for multiple configurations allows the investigation of both series and parallel operation. These configurations produce varied flow rates in the lines, demonstrating the effect of mass flow rate on the amount of energy transferred. The apparatus can also be used to demonstrate or test calculations of multiple pipe networks, a subject covered in most fluid mechanics courses.

V. IMPACT OF THE EXPERIMENTS ON THE LEARNING PROCESS Experiments such as those described above improve student learning by providing hands on experience with the concepts covered in lecture. The heat exchanger board does this in several ways. First, it illustrates concepts in a very simple and clear manner. There is no complicated machinery, and in no way does the apparatus resemble a “black box.” Everything is exposed and very “physical.” The students can see the path the flow moves through, and can even see the water circulating though the transparent Tygon™ tubes. Temperatures are measured with thermocouples, which are clearly visible, and students can also feel the temperatures of the various streams with their hands. The heat exchanger board also provides an opportunity for students to use several experimental techniques. Thermocouples, pressure transducers and flow meters may all be utilized. Students might be introduced to these instruments and the theory behind them in a measurements course. Using them again in a heat transfer course, would provide reinforcement of material covered in other courses and give the students an opportunity to use what they have already learned. Hands on experience and experimental measurements complement the theory covered in lecture. Comparison of calculated and

Table 1. Predicted / Measured Results from Sample Lab January 1999

Journal of Engineering Education 29

experimental results is particularly instructive. Differences can lead to a better understanding of the accuracy of correlations, the impact of simplifying assumptions on final results, and the need for consideration of experimental uncertainty. The apparatus can be used to give students practice in solving complicated problems involving several concepts in both fluid mechanics and heat transfer. The instructor may choose which information to give the students, what to have them calculate or estimate on their own, and what they should measure. As problems become more involved, they can also be designed to be more open ended. Students tend to like specific answers and are often uncomfortable with open-ended problems. Overcoming their resistance to this type of problem is necessary for their education in design. If designs are to be completely tested with the heat exchanger board, some limitations are necessary. The board was designed to be flexible, but also self sufficient and easy to reconfigure. It was not, therefore, designed with the intention of replacement or addition of parts. Students are restricted to the three heat exchangers provided and the existing pump. It is fairly easy to drain the working fluid from the board, so water could be replaced with some other fluid for test of a design, although care should be taken to use only those fluids which can later be flushed from the lines. Designs that go beyond the capabilities of the board are also possible, if the board is used for testing of parts of the final design instead of the entire design.

REFERENCES 1. Kays, W.M. and London, A.L., Compact Heat Exchangers, 3rd ed., McGraw-Hill, New York, 1984. 2. Incropera, F. P. and D.P. DeWitt, Introduction to Heat Transfer, 3rd Ed., John Wiley and Sons, New York, 1996.

VI. CONCLUSION The introduction of experiments, such as those described in this paper, into a course aids the learning process by improving students’ motivation. Students are likely to take an added interest in problem solving if they know that their solutions will be subject to a physical test. Laboratories may also be set up as contests, providing students or student teams the chance to compete with each other to provide the best predictions or design. Problems may be made sufficiently complex so that there are several possible ways of arriving at an estimate of performance. The best solution would be that which most closely matches the results of the physical experiment. It should also be noted that the heat exchanger board can be used in the learning process in ways other than problem solving exercises. The board is relatively small and can easily be wheeled into a classroom for demonstrations and incorporation into lectures and discussions. Many faculty members are beginning to incorporate interactive computer simulations into their classes. The heat exchanger board provides the opportunity to do the same with interactive experiments.

VII. ACKNOWLEDGEMENTS The authors gratefully acknowledge help with the construction of the apparatus by Louise Becnel of the Technical Support Division of the U.S. Naval Academy, and summer interns Raegan Burroughs and Andrew Holt, participants in the Science/Engineering Apprentice Program sponsored by the National Science Foundation and administered by George Washington University.

30

Journal of Engineering Education

January 1999