Experimental study on frosting suppression for a finned-tube evaporator using ultrasonic vibration
Dingyuan Wang,Tangfei Tao,Guanghua Xu,Ailing Luo,Shaoying Kang
DOI: https://doi.org/10.1016/j.expthermflusci.2011.03.002
IF: 3.2
2012-01-01
Experimental Thermal and Fluid Science
Abstract:An experimental study was conducted to investigate the possibility and the effect of frost release from a finned-tube evaporator by using ultrasonic vibrations in natural convection. A compressor was adopted as cold source, and a finned-tube evaporator was placed horizontal on the frames inside a chamber. The aluminum fins surface was covered with hydrophilic coatings whose contact angle of the water drop of the surface is 37.7°, and the fins had wing-cases to enlarge the heat-exchange area. The thickness of a fin was approximately 0.1 mm, and the gap distance between any two fins of the evaporator was about 1.55 mm in average. An ultrasonic transducer with resonance 28.2 kHz and rating power 60 W was adopted as a vibrator to excite the evaporator. A microscopic image system was used to observe and record the whole frosting process with/without ultrasonic vibrations, a method of frost thickness measurement on the double sides of a fin was present and realized through digital image processing technology. From microscopic view, the deposited frost nearly block the fins gaps when the compressor worked for 32 min without ultrasonic vibration, while almost 2/3 of the gaps between two neighbor fins were unblocked when the compressor worked for 92 min with ultrasonic vibration, which shows that the frost deposited on the fins surfaces could be suppressed with high frequency ultrasonic vibrations. From macroscopic view, the frost growing on the fins are being vibrated to fall off and accumulated under the evaporator to form frost piles. Although the fins above the braze tubes were deposited with frost, the fins between any two neighboring tubes were unblocked. Experiments show that the basic ice layer on the fins cannot be removed with ultrasonic vibrations, but frost crystals and frost branches on the ice layer can be fractured and removed effectively, the frost growing on the evaporator can be suppressed by using ultrasonic vibrations. The main mechanism of ultrasonic frost suppression maybe mainly attributed to the high frequency ultrasonic mechanical vibrations that can break up frost crystals and frost layers, then frost will fall off with the gravity, not to ultrasonic cavitation effect or heat effect. Keywords Evaporator Frost suppression Defrost Ultrasonic vibration Nomenclature BLT bolt-clamped Langevin-type transducer W * L * H Width(mm) * Length(mm) * Height(mm) s second min minute H hour Th frost thickness on the double sides of fins (mm) °C temperature Tf fins’ temperature RH relative humidity (%) T∞ ambient temperature V voltage (V) W power μm the amplitude of ultrasonic vibration f frequency (Hz, kHz) 1 Introduction Frost accumulation on the surface of a finned-tube evaporator is an undesired but commonly seen phenomenon in an air-heat-pump system, especially for an air conditioner, a refrigerator, a fan and other refrigeration system. If the deposited frost on the evaporator is not removed in time, it will block the air channels and reduce the heat-exchange area, and the air flow resistance will increase dramatically, such condition will drastically decrease the heat-exchange efficiency, even worse, the system may break down. Therefore, periodical defrosting is an essential procedure to ensure the normal function of an air-pump system. Currently, defrosting methods using heat to melt frost suffer some shortcomings, e.g., resulting in more energy consumption, reducing comfortable feelings, lowering the quality of goods and so on. Therefore, a method with low energy consumption to help preventing frosting or reducing frost accumulation on the evaporator to ensure the normal heat exchange is urgently required. Cheng and Shiu [1] adopted a microscopic image system to study frost formation and frost crystals growth on a cold plate, they found that frost thickness grows in two-dimension at the leading edge, the thickness and the density of the frost layer both increase with time flows and the ambient humidity increases. Frost crystals and branches like trees with different shape and size consist of a frost forest. Some previous experimental studies on normal frosting have concluded that the crystallization of frost and the growth of the frost layer depend on several factors in the frost formation process, e.g., the ambient temperature, the ambient relative humidity, the velocity of the surrounding air, the temperature of the cooling surface and the surface nature (hydrophilic or hydrophobic). Frosting velocity is slow when the relative humidity is less than 40%, however, when the ambient humidity is high and there is a big temperature difference between the ambient air and the cold surface, it will be easier for frost deposited on the cold surface. Frosting on a cold surface can be dived into two kinds, one is the direct deposition of frost on the cooling surface, and the other is: condensate water droplets → ice layer → frost crystals → frost branches → frost layers. Frosting on the evaporator of an air conditioner and a refrigerator can be classified into the second mode. Hayashi et al. [2] divided the frost formation growth into three periods: (a) crystal growth period, (b) frost layer growth period, and (c) frost layer fully growth period. Bascom et al. [3] studied the ice adhesion to hydrophilic and hydrophobic surfaces, they found that the reduction of the adhesion shear strength did not correlate with the water contact angle on the coatings but was usually due to air entrapment at the ice/coating interface of to cohesive failure of the coating itself. The authors have observed some heat-exchangers with different fin gap distances, the fin gap distance of a refrigerator (about 5 mm) is larger than an air-conditioner (about 2 mm) in commercial use, an evaporator with such a large fin gap distance will have to decrease the heat-exchange efficiency and increase the volume of the evaporator. In addition, the authors also observed that part of the accumulated frost could be vibrated to fall off from the surface of an evaporator by tapping the outdoor unit of an air conditioner in winter. Under such consideration, it may be a feasible method for frost removal or reduction of the accumulated frost on the surface of an evaporator by using mechanical vibrations, meanwhile, an innovative evaporator with small fin gap distance and good heat-exchange performance without defrosting too frequently maybe widely used in refrigeration field. However, it brings forth the question to study which form of vibration to exert and how to exert the vibration in practical defrosting applications. Some studies have investigated frost suppression or defrosting technology for cold surfaces with low frequency mechanical vibrations. Wu et al. [4] mentioned that if an additional field force was exerted to a low-energy cold surface, frost crystals might fall off from the cold surface. Such additional fields could include mechanical vibration or acoustics affect, and electrical field. The contact angle of a cold surface with hydrophilic coatings is near to zero, and such a hydrophilic surface has a big surface energy, while the cold surface with hydrophobic surface has a big contact angle and a lower surface energy. Frost crystals have a small adhesion force to the low energy surfaces. A semiconductor refrigeration chip was used for frosting in order to examine the effect of defrosting with low frequency mechanical vibration. The experiment shows that frost cannot be tapped off in case of the hydrophobic cold surface no matter when to tap and how to vibrate the surface. Cheng and Shiu [5] found that the low frequency mechanical vibration with frequency varying from 100 to 200 Hz, and amplitude varying from 40 to 100 μm could not remove the frost accumulated on a cold surface. On the other hand, the vibration could change the state of condensate water droplets, and affect the behavior and process of frosting. In other ways, some researchers have studied the possibility of frost suppression or defrosting by using high frequency ultrasonic vibration or ultrasound with ultrasonic actuators, e.g., piezoelectric plate, ultrasonic transducer, and the ultrasonic transducer widely used in this field of technology is a kind of piezoelectric bolt-clamped Langevin-type transducer (BLT), which can be driven to vibrate at a high frequency with an ultrasonic power supply. Power ultrasonic or power ultrasound is a kind of ultrasonic with frequency more than 15 kHz, both ultrasound and ultrasonic vibration can induce mechanical vibrations. Small particles in the ultrasonic field will be excited to vibrate with low amplitude, high frequency and high acceleration. Power ultrasonic has a large output power density, which is mainly used for process strengthening and ultrasonic machining. Lecture [6] maybe one of those very first experiments studying ultrasonic frost suppression. Adachi et al. [6,7] exerted ultrasonic vibrations to an aluminum plate (20 mm * 92.6 mm * 1.9 mm) for about 90 min to examine the effect on frost with ultrasonic vibrations. The experimental result shows that ultrasonic vibrations with frequency of 37 kHz and amplitude of 3.1 μm can decrease the frosting by almost 60%, and the mechanism of this effect cannot be attributed to the heat generated by ultrasonic vibrations with high amplitude, but may have some relationships with the mechanical vibrations or acoustics to the frost crystals, that is, frost crystals cannot stay on the plate to grow further under high frequency vibrations when the vibration amplitude is high. Yan et al. [8] exerted the ultrasonic with parameter of 15 kHz/30 W and 20 kHz/30 W to a refrigeration fan, after working continuously for 48 h, the fins were not blocked by frost. The result shows that it is feasible to use the ultrasonic defrost technology. Barelli et al. [9] connected some piezoelectric plates to the fins of an evaporator, and an electronic circuit was used to drive the piezoelectric plates. They found that ultrasonic vibrations using piezoelectric plates can suppress and decrease the frost deposition during the refrigeration operation, as well as increase the period between two consecutive defrosting phases. Dong and Zhenqian [10] investigated the frost phenomenon on a flat surface with and without 20 kHz noncontact ultrasound wave under natural convection conditions for 10 min. The experimental results show that the frost growth on the flat surface was remarkably restrained by ultrasound. Frost suppression or defrosting using ultrasonic vibrations or ultrasound is a kind of new and challenging work, and it is a promising area with direct potential applications in refrigeration system. However, until now, only a few open literatures exist in such field. Although a certain amount of thought-provoking conclusions have been derived in early literatures, the overall study is small in scale and far from being sufficient. The mechanism of frost suppression or defrosting with ultrasonic is still not clear, how to build a mechanical structure on the evaporator with good ultrasonic vibration transfer performance and how to exert ultrasonic vibration are also not understood. The features of the early studies on ultrasonic frost suppression or defrosting are as follows: Firstly, using an evaporator as a target for frost suppression or defrosting study with vibrations is not extensively investigated. Furthermore, an aluminum plate (or a copper plate) is commonly used for experimental study, and some piezoelectric plates are affixed to the fins of a finned-tube evaporator which has a large gap distance for frosting and defrosting study in early days. Such experiment structures are significantly different from the commercial use. Secondly, early studies are mainly concerned with the possibility of frost release from cold surfaces with macroscopic study, as well as the thickness measurement for frost layers by using traditional measurement devices, e.g., precision balance [6,7] or vernier calliper [8] . It is not easy to define the boundaries of frost, and the error of the measurement results is large. In addition, it is complicated to operate a laser instrument or a precision balance in frost measurement. Macroscopic study on frosting (e.g., traditional calipers measurement method, high accurate balance method) can only obtain the changes of frost thickness during the period of frosting on the fins surface, thus making it difficult to understand the effect on tiny frost crystals with ultrasonic vibrations in microscopic view. The effects on the early patterns of frost in microscopic view under the influence of ultrasonic vibrations have not been observed and described in previous studies for some reasons; especially the effects on the formation and growth of tiny frost crystals and frost branches on the fins of a finned-tube evaporator have not been studied before. Thirdly, where and how to install an oscillator on an evaporator to excite the evaporator vibrating at high frequency and low amplitude are still not clear, and how to design the mechanical vibration transfer structures for the evaporator with high performance of ultrasonic transfer is still not clear. Finally, the mechanism of frost suppression or defrosting by using ultrasonic vibrations is not clear. Some studies conclude that frost crystals cannot stay on the cold surface to grow further under high frequency vibrations, the others ascribe the effect to the high frequency vibrations of vapor that close-by the cold surface, the question that which feature of ultrasonic is the main function for frost suppression or defrosting is still to be controversial. Compared with previous studies, our study has some new points. This paper is intended to study frost removal for a finned-tube evaporator whose fins covered with hydrophilic coatings by using ultrasonic vibrations in natural convection; the experimental condition is more close to a commercial refrigeration system. The ultrasonic vibration transfer structure for the evaporator will be described to show where and how to install the ultrasonic transducer. Dynamic microscopic process of frost crystals’ formation and growth will be investigated through a microscopic image system. Then, the changes of frost thickness on the double sides of the fins are obtained through microscopic measurement software to examine the effect of frost suppress with ultrasonic vibrations. Finally, the mechanism of ultrasonic frost suppression or defrosting will be summarized. 2 Experimental setup Fig. 1 shows the experimental setup, which is mainly composed of a chamber, a set of custom-built refrigeration frosting system, a set of ultrasonic vibration system and a microscopic image system. The chamber the temperature and humidity kept constant is used to offer a controllable experimental condition with stable temperature and humidity, the compressor and the condenser are put outside of the chamber, the evaporator is placed horizontal and fixed on the inside frames of the chamber, a braze pipe and a capillary pipe pass through a whole on the left side of the chamber, one side of the braze pipe or capillary pipe is connected to the evaporator, and the other side of the pipe is connected to the condenser. The refrigeration system includes a compressor, a finned-tube condenser, a fan, a finned-tube evaporator, a controller, and brazes tubes. The refrigeration system uses a China R600a refrigerator compressor (ESD65A, Capacity: 100 W, C.O.P: 1.40 W/W) as cold source for frosting on the evaporator. The evaporator is comprised of single row braze tubes and aluminum fins, the size of the evaporator is 270 mm * 300 mm * 12 mm ( W * L * H ). As shown in Fig. 2 , the initial fin gap distance from one side of a fin to the next side of its neighboring fin is 1.55 mm. In addition, the arising part of a fin is defined as fins wing-case or wing-sheath, which is used to enlarge the heat-exchange surface of fins. The thickness of a fin is about 0.10 mm, the fins surface is covered with hydrophilic coatings whose contact angle of the water drop of the surface is 37.7°, the surface roughness of the fin is about 0.313 μm. A set of ultrasonic vibration device is used to excite the evaporator to vibrate at a high frequency. The power ultrasonic system includes an ultrasonic power supply and an ultrasonic transducer. The maxim output power of the ultrasonic power supply is 300 W, and the frequency of the power supply can be adjusted in 15–60 kHz, the power supply has a function of auto frequency tracking, and the resolution of the frequency is 0.01 Hz. The BLT adopted in the experiment is a commonly used ultrasonic transducer for ultrasonic cleaning application with resonance 28.2 kHz, the bandwidth of the BLT is 2 kHz, the rating power of the BLT is 60 W, and the stationary capacitor of the BLT is 4200 pF. There are two metal aprons on the left and right side of the evaporator with their length of 300 mm and height of 33 mm. The aprons are bent in 90°; the board width of the bent apron is 10 mm, the thickness of the board is 1.5 mm. The aprons and fins are connected through braze tubes in expand tube joint style. In order to excite the evaporator to vibrate at a high frequency with a BLT, a vibration transfer structure and a suitable installation place for the BLT on the evaporator are required. With the consideration, the center of the evaporator may be the best place for installing a BLT, however, it is difficult to build the structure and it will modify the evaporator to a great extent. The apron of the evaporator is selected as the vibration transfer structure, so there is no need to modify the mechanical structure of the evaporator. Fig. 3 is depicted to show how and where on the evaporator the BLT is connected to. A BLT is fixed on the center place of the right apron metal board along the distance direction. The diameter of the BLT is 60 mm, only about 1/6 part of the radiating surface of the BLT is contacted to the narrow board, therefore, there is only a part of ultrasonic vibration can be transferred from the BLT to the evaporator. Ultrasonic vibration is a kind of mechanical vibration whose amplitude decreases with the distance and the vibration transfer structure. The finned-tube evaporator has a complex structure with braze tubes and aluminum fins, if the evaporator is excited with an ultrasonic transducer fixed on a metal board which connected to the tubes, the vibration amplitude on the fins will decrease after the ultrasonic vibration transferred through metal board to tubes and through tubes to fins. The vibration amplitude at the radiating surface of the BLT is lower than 5 μm when it works without any load, when the BLT works with load, the vibration amplitude will decrease, especially when it is fixed on an evaporator. As a result, the vibration amplitude on the fins should be less than 5 μm. The vibration on the fins is difficult to measure for its finned-tube complex structure. A BLT is fixed on a metal board of a finned-tube evaporator, when the BLT is working, ultrasonic cavitation phenomenon can be induced on the metal board with ultrasonic vibration; however, cavitation phenomenon cannot be induced on the fins, which shows that the ultrasonic energy density on the fins is lower than on the board. PSV-400 (Polytec Scanning Vibrometer, made in Germany) is adopted to measure the vibration on the evaporator when the BLT is working. The metal board and the surface of fins are selected to measure the vibration. Fig. 4 a shows the structure of the evaporator and Fig. 4 b shows the distribution of the measuring points, a BLT is installed on the center of the board approximately with a bolt, and the distance between the center of the BLT and the top edge of the metal board is a bout 143 mm. According to the requirements of the measurement, 3 points on the evaporator are selected for measuring the vibration, point A is on the board, point B and C are on the fins surface, the distance between point A and the center of the BLT is 107 mm; the distance between A and B is 100 mm, the distance between point C and B is 10 mm. Point B locates above a part of the tube. Point B and C locate under the lens of the microscope for observing the frosting process. To ensure the laser’s reflection strength, reflecting films are pasted on the surface of the measuring points. Then, ultrasonic power supply is turn on and the BLT start to work, from PSV-400, the vibration signal can be acquired, then from FFT analysis of the acquired signals, it can be seen that there is only one main spectrum line in the graph and the vibration resonance frequencies and amplitudes can be obtained. The resonance frequency is 28.641 kHz, and the approximate amplitude and acceleration of ultrasonic vibration at the measuring points on the evaporator are stated in Table 1 . As can be seen from the measurement, the vibration amplitude is low (about 1 μm), but the acceleration is large. In addition, the amplitude at point A is larger than point B or C, which shows that the mechanical vibration has been decreased after the ultrasonic vibration transferred from metal board to tubes and then from tubes to fins. Therefore, the vibration transfer structure can influence the ultrasonic vibration transfer efficiency, as well as the vibration amplitude on the fins. As shown in Fig. 5 , a microscopic image system is adopted to observe and record the dynamic microscopic frosting process with and without the influence of ultrasonic vibrations to examine and measure the effect of ultrasonic frost suppression, the lens of the microscope is vertical to the surface of the evaporator. A LED light source is used as a cold light source to illuminate the target fins under the lens to get a clear microscopic view. The distance between the center of the lens and the ultrasonic transducer is 100 mm. The microscope is comprised of a three head optical zoom stereo microscope, an industrial CMOS camera with resolution 3.0 mega-pixel, a PC and camera application software. The magnification of the microscope is from 7× to 28×, the microscope and the camera are connected with a 0.8× lens; microscopic images can be transmitted to PC for monitor and analysis through camera drivers and application software. Besides, a digital hygrothermograph is used to monitor the experimental condition, the uncertainty of the temperature measurement is 0.1 °C, and the uncertainty of the relative humidity measurement is 1%. A portable thermocouple temperature tester and a portable infrared radiation thermometer are used to measure the temperature of the evaporator; with the uncertainty of the measurement is 0.1 °C. 3 Frost thickness measurement procedure Wang et al. [11] introduced a non-contact frost thickness measurement method based on a microscopic image system and image processing technology, the method is adopted in this experimental study to measure the frost thickness on the double sides of the fins for a finned-tube evaporator. In order to obtain the changes of frost thickness varying with time during the frosting process, a method of frost thickness measurement for frost growing on the double sides of the fins is adopted based on a microscopic image system and digital image processing technology. Fig. 6 a shows the frost crystals growing on the fins in the early frosting period, and Fig. 6 b shows the frost layers on the fins at the latter frosting period. Frost thickness is defined as the length or thickness of frost growing on the double sides of a fin. From the experiment, we can capture clear images of tiny frost crystals and frost layers, as well as can observe the dynamic growing process and the fracture phenomenon of frost crystal with/without ultrasonic vibration. The measurement procedure is as follows: (1) Calibration of the measurement system. Firstly, the microscope should be rightly placed and adjusted to make the barrel of the microscope vertical to the surface of the evaporator, then the magnification and focal length of the microscope are carefully adjusted to obtain a clear amplified vision of fins; a standard circle with its diameter D (the maximum diameter is 1 mm, the minimum diameter is 0.1 mm) is acquired and saved to PC with a camera application software. Secondly, image processing on the standard circular is applied to get the count number of pixels X , then the distance of each pixel P ( P = D / X ) can be obtained. (1) Thickness measurement for the frost on the double sides of the fins. Firstly, in the case of a constant magnification of the microscopic image system, frosting images of fins with different frosting condition at different time are acquired by using a microscopic camera. Secondly, Frost boundaries can be divided into two types, one is the boundary of tiny frost crystals growing on the fins, and the other is the boundary of frost layer growing on the double sides of a fin, as shown in Fig. 7 . Both kinds of frost boundaries are not ideal linear and easy to be undermined by heat or other direct measurement methods. Image edge detection is used to obtain the frost boundaries, which is a non-contact method, and it will not undermine the frost on the fins. After image filtering and edge detection, the edges of frost images become more distinct and clear than the original frosting images, then the pixels of the distance between two edges Y can be obtained. In order to get a high precise result, 3–6 typical points along the frost boundaries are selected to calculate the average edge distance of tiny frost crystals and frost layer. Finally, frost thickness on the double sides of the fins which includes the fins thickness can be calculated through Th = Y * P . 4 Experiments and discussion With the restrictions of the camera and the microscope, the experiments were conducted in normal convection to avoid the lens of the microscopic system becoming steamed up. Besides, the blast fan is not working in the chamber to prevent the air flow melting the frost on the fins, so air velocity is not considered in this experimental study. The evaporator was placed horizontal and fixed on the inside frame of the chamber, the initial ambient temperature around the evaporator was 9.2 °C, the relative humidity was 62%.The cooling surface temperature will drop down gradually varying with time, and it needs about 6 min to drop down the temperature of the evaporator from the ambient temperature to the lowest temperature, the lowest temperature of fins was −9 °C, the temperature drop curve of the fins surface is shown in Fig. 8 . The authors are interested in the influence on frost crystals and frost layers under the influence of ultrasonic vibrations during the whole frosting process, so we only concern about the initial temperature and the lowest temperature of cooling surface, the changes of the temperature of the cooling surface varying with time will be discussed in our follow study. Ultrasonic defrosting experimental studies under the influence of different air velocity, temperature of cooling surface, ambient temperature and ambient humidity in force convection will be carried out in future for the restricted experimental devices at present. Experimental frosting tests were carried out respectively with and without ultrasonic vibrations; the influence on the frost crystals, frost branches and frost layers with ultrasonic vibrations was expected to be found through microscopic observation of the whole experiments. All the experiments started with the initial fins surface in dry state. When the compressor started to work, the microscopic image system began to record the frosting process. The vibration frequency of the ultrasonic transducer was about 28.64 kHz, and the practical working power was about 30 W in average. The frosting experiments had been carried out for several times to ensure the reparability and the credibility of the experimental study, one of the normal frosting experiments without ultrasonic vibrations lasted for 31 min, the other frosting experiment with the influence of ultrasonic vibrations lasted for 92 min. 4.1 Normal frosting without ultrasonic vibrations Fig. 9 shows the frosting process in microscopic view without the presence of ultrasonic vibrations. When the compressor worked for 45 s, quite many condensate droplets appeared. At 90 s, the condensate water started to freeze and became a layer of ice. At 118 s, the authors observed some frost crystals began to appear on the surface of ice layer. At 3 min, there was evident appearance of the patterns and sizes of frost crystals, meanwhile, some big frost crystals grew up to frost branches. At 9 min, frost crystals began to form big feather-like branches; however, the branches were still not overlapping. Some time later, more and more frost branches formed an overlapping growing pattern, and the basic frost layers on the fins became thicker and thicker gradually while the frost density grew larger. At 31 min, frost branches between any two neighboring fins touched and overlapped with each other; the basic ice layers on the fins surface became quite thick and compact in such a way to form a high density ice layer strongly adhering to the fins surface, and the accumulated frost approximately blocked the fins. The pictures of the normal frosting process of the evaporator without the influence of the ultrasonic vibrations in macroscopic view are listed in Fig. 10 . From Figs. 9 and 10 , the normal frosting process can be divided into 4 stages in detail. The first stage is the formation of condensate and its freezing, during this stage, the basic ice layer forms, and the basic frost thickness including fins increase a bit level. The second stage is the formation and the growth of tiny frost crystals, in such stage, quite many tiny frost crystals formed and start to grow quickly, the quantity and the sizes of frost crystals increase a lot in a short time, meanwhile, the frost thickness of the fins increases. The third stage is the overlapping growth of frost crystals on the fins, such growth makes frost crystals density increase, and more deposited frost crystals accumulated on the fins. The forth stage is the growth of frost layers, when the fins are blocked, frost layers formed on the surface of the evaporator evidently, with time flows, the thickness of frost layers become thicker and thicker. Through the normal frosting experiment, we can see that the patterns of frost crystals on the surface of an evaporator are quite complicated, there are two features during the frosting, the first one is frost crystals growing in two-dimension, frost crystals grow in vertical direction to increase the thickness of frost on the double sides of fins, and growing along with fins to increase the height of fins. The second one is that the boundaries of frost is nonlinear, and appear to be a sawtooth pattern. 4.2 Frosting under the influence of ultrasonic vibrations Fig. 11 shows the microscopic views of the fins frosting under the influence of ultrasonic vibrations. The initial fins thickness was about 0.1 mm, and the narrow gap distance between any two neighboring fins was about 1.55 mm. When the compressor worked for 110 s, condensate droplets could be seen on the fins surface from microscopic view. At 160 s, the condensate freezing water started to form a thin ice layer on the fins surface. At 180 s, the basic ice layer became thicker than before, and then, frost crystals began to form on the ice layer and started to grow. At 220 s, frost crystals appeared to be more evidently, but tiny frost crystals could not be broken with ultrasonic vibrations. Then from 225 s, it could be seen from the microscopic image system that many frost crystals were gradually fractured at the joined point of ice layer and fall off with the influence of gravity, part of frost crystals that not broken with vibrations will keep growing, and the thickness of the basic ice layer increased a little. At 5 min, most frost crystals on the fins were fractured and fall off, then, frost crystals growing and being broken gradually, the basic ice layer increased a bit, which had a high density and difficult to remove with ultrasonic vibrations. At 50 min, the basic frost layer was about 0.61 mm in thickness, and the gap distance between two fins decrease to about 0.95 mm, which is 3/5 of the initial fin gap distance. Apart from the basic frost layer on the fins surface, there were only little quantity frost crystals remained on the basic frost layer, the fins were not blocked, however, some new frost crystals that could not fall off with the gravity for the blockage of fins wing-case began to form frost accumulation on some place of fins, especially on the fins above the braze pipes. At 70 min, there was more frost deposited on some fins that above the tubes. At 92 min, the basic frost thickness was about 0.66 mm, the interval distance of fins was about 0.89 mm which was about 60% of initial fins distance, and some place of evaporator was accumulated with more evident frost which might be caused by the effect of ultrasonic vibration nodes. Fig. 12 shows the macroscopic views of the frosting process of frost growing on the surfaces of aprons and fins. At 10 min, there were only a thin ice layer and little frost crystals on the surface of aprons and fins. At 13 min, frost accumulated on the surface of apron in interval phenomenon, which reflects the apron’s flexural vibration. At 16 min, more frost crystals appeared on the fins surface, and more frost accumulated on the place of the apron where vibration nodes exist. At 50 min, frost accumulation on the vibration nodes of the apron appeared to be more evident, and frost deposition on the fins above the braze tubes could be found, however, there was a little frost accumulation on the fins between any two tubes, and the gap distance of fins that between any two tubes changed little. At 70 min, there was more frost deposited on the surface of fins above the braze tubes, the deposited frost were mainly the fallen frost caused by vibrations and some new growing frost crystals. At 91 min, there was more frost accumulated on the surface of the apron and the fins above the tubes, even the place between vibration nodes were connected with deposited frost, however, the gaps of fins were still unblocked with frost, the phenomenon that frost accumulation with fallen frost induced by ultrasonic vibrations and gravity could be seen under the evaporator. It is evident that ultrasonic vibrations can suppress frosting and remove frost crystals. Through observations of the frosting experiment with ultrasonic vibrations, there is frost deposited on the apron board every certain distance. The accumulated frost on the apron surface appeared to be a sight of vibration nodes effect, which reflects a kind of flexural vibration on the apron with ultrasonic vibrations. The vibration nodes are stationary places, therefore, the place where the vibration nodes exist have a poor ability for frost suppression, and frost will accumulate on such place. However, there is only a little frost accumulation on the place between any two vibration nodes. Such phenomenon may be attributed to the strength (amplitude and acceleration) of ultrasonic vibration. From vibration measurement for an evaporator by using PSV-400, we find that the amplitude and acceleration on the fins above the tubes is smaller than on the fins between any two tubes, therefore, the fins above the tubes have a poor defrosting ability. The vibration transfer structure adopted in the experimental study is not ideal, and the amplitude of the vibration is small, we infer that the effect of ultrasonic frost suppression/defrosting may be better if the vibration amplitude increases. In addition, the evaporator is placed horizontal, frost on the fins that between any two tubes can fall down from the fins gaps, and accumulated under the evaporator. However, frost on the fins above the tubes cannot fall off because of the blockage of tubes, the narrow gap distance of fins and fins wing-case, with time flows, more and more removed frost will accumulated on the tubes and block the gaps of fins where vibration nodes exist, and the fins above the braze tubes will be accumulated with incompact frost. Although the accumulated frost has a low surface adhesion force to the fins, the ultrasonic vibration cannot release the frost from tubes for its low vibration amplitude and acceleration. Such condition will decrease the ability of defrosting. To ensure the normal heat exchange for an evaporator, fins gaps should not be blocked. Without ultrasonic vibration, the tubes and fins will frost quickly, and then frost will block the fins gaps. Even worse, the accumulated frost may change to be a layer of ice to cover the whole evaporator with strong adhesive force, which is difficult to remove except for using heat to melt the ice layers. Frost and ice on the evaporator become the resistance of air flow; therefore, the heat exchange efficiency will decrease a lot. If the accumulated frost is not removed in time, air flow cannot work normally through fin gaps, the refrigeration system will work in low efficiency or breakdown, and the compressor may be damaged. However, if an evaporator is placed horizontal under the influence of ultrasonic vibration, frost crystals can be fractured and dropped through the gaps of fins with the gravity during the early frosting period, frost will not block the fins, and there is only a little frost accumulated on the fin gaps, normal air convection can be ensured, then the removed frost will form a certain frost accumulation under the evaporator. In the latter period of the experiment, frost blockage will happen at some place of fins above the tubes because of the frost accumulation on the gaps of fins or on the surface of the evaporator where the amplitude of the vibration is too small. With ultrasonic vibration, although the frost will accumulate on the fins above the tubes, the frost is incompact and easy to be released with air force in a short time (such method will be adopted to improve the defrosting effect in future). The whole heat resistance of the evaporator with ultrasonic vibration is lower than the condition that working without ultrasonic vibration, the refrigeration system can work in a longer time without defrosting, that means, ultrasonic vibration can prolong the time of defrosting period. Frosting data is calculated through microscopic image measurement software, the plots of frost thickness varies with time with/without ultrasonic vibrations are obtained. As can be seen from Fig. 13 , ultrasonic vibration has an evident effect of frost suppression for a finned-tube evaporator, however, there are still some remnant frost left on the fins that could not be removed by ultrasonic vibrations for the strong surface adhesion to the fins, it means that ultrasonic vibrations cannot make completely no frost accumulated on the surface of fins. Ultrasonic vibration can break up frost crystals, frost branches and frost layers from the initial ice layers growing on the fins surface, new frost crystals will grow on the ice layer, and then new frost crystals and frost branches will be fractured and fall off, however, not all frost crystals and frost branches can be fractured, in such condition, the thickness of frost layers on the fins surface will increase slowly with time flows. We can find that fin gaps between any two fins are not blocked at the end of the experiment, and the fins above the tubes are accumulated with loose frost, the density of the accumulated frost is not large, and its adhesive force is not so large. With the influence of ultrasonic vibration, the thickness of ice layers increases a bit; however, it is not means that the density of frost on the evaporator increases a lot with ultrasonic vibration. During the early frosting period, frost crystals’ length varying quickly for the frost crystals’ growth, fracture and fall off, then at the later period, the thickness of frost layer changes slowly. However, the frost thickness is increasing. From microscopic observation, frost is being vibrated and fractured, that means frost is influenced with ultrasonic vibration during the whole experiment. Ultrasonic vibration cannot ensure completely no frost accumulated on the evaporator, but it can decrease frost accumulation on the evaporator, the frost will fall off and accumulated under the evaporator with the influence of ultrasonic vibration. 4.3 The effectiveness of ultrasonic frost suppression study in microscopic view As can be seen from the experiments, frost growing in two-dimension during the period of frost crystals growth, frost boundaries are very nonlinear which makes the frost thickness measurement based on a microscopic image system has some shortcomings, firstly, during the early period of frost growth, frost crystals are being fractured and fall off from the fins with the influence of the gravity, the quantity of frost crystals decrease a lot, however, the frost thickness on the double sides of fins changes a little, the method only using the variation of frost thickness to examine the effect of frost fracture is not enough. Secondly, in the latter frosting period, some place of fins will accumulated with frost, it is not scientific to evaluate the effect of frosting with ultrasonic vibrations by measuring the changes of frost thickness on such place where vibration nodes exist. Withstanding its limitation, it is still effective in examining the effect of frost suppression in the early frosting period before fins blocked. The dynamic influence on frost crystals and frost branches on the fins of a finned-tube evaporator can be observed clearly with the help of a microscopic image system. 4.4 The mechanism of ultrasonic frost suppression for an evaporator It needs some time to dropt down the temperature of fins from the ambient temperature to a stable low temperature in normal convection, firstly, a basic layer of ice will be formed on the fins surface, and then, frost crystals will grow on the surface of ice layer. With the comparison of frosting experiments for a finned-tube evaporator with and without ultrasonic vibrations, ice layers appeared on the fins surface at the early period of frosting, and with time flows, the thickness of ice layer will increase to some extent. However, the difference between two conditions is that frost crystals on the basic ice layer will be fractured and fall off under the influence of ultrasonic vibrations, the growth of frost crystals can be suppressed. The main functions of power ultrasonic are cavitation effect, mechanical vibration effect and heat effect, but which function is the most possible mechanism of frost suppression is still controversial. The cavitation effect is mainly take place in the liquid which has a certain temperature, as can be found from the experiment, the condensation water on the surface of fins and apron is little and it only take a short time to freeze the condensate droplets to a basic ice layer, there is no suitable condition to generate cavitation phenomenon, thus, during the early period of frosting, the mechanism of ultrasonic frost suppression cannot be attributed to the cavitation effect induced by ultrasonic vibrations. An ultrasonic transducer will turn hot when works for a long time, from the experiment, we can see that the place of apron which has a short distance with the transducer will accumulate less frost, which shows the heat effect of the transducer, however, the heat effect is not the main function to suppress frost, the fins under the lens of microscope has a long distance to the transducer(about 80 mm), the heat effect will has a little affect on the fins, and from microscopic observation, frost crystals are being fractured and fall off from the fins surface, and after a certain time, the removed frost will accumulated under the evaporator, such phenomenon is quite obvious. Therefore, the cavitation effect and the heat effect can be excluded from the main functions in ultrasonic frost suppression, the most possible mechanism may be attributed to the effect of mechanical vibrations on frost crystals induced by the ultrasonic transmit in the fins and evaporator, as well as frost crystals and ice layers. Through the microscopic observation, the basic ice layers formed on the surface of fins cannot be released with ultrasonic vibrations, and the thickness of the basic ice layer increase with time flows, however, frost crystals and frost branches on the fins surface can be fractured and fall off. It shows that the basic ice layers have a strong adhesion force to the cold surface with hydrophilic coatings, while frost crystals have a poor adhesion force to the basic ice layer. The result of our experiments is corresponding to and proves the mechanism hypothesis of ultrasonic mechanical vibrations as previously described by Adachi et al. [6,7] . The amplitude of ultrasonic vibration is very low, but the acceleration induced by ultrasonic vibration is very large for the high frequency of ultrasonic vibration. Ultrasonic vibration can induce a big shearing force to frost crystals and frost branches. When ultrasonic transmit in fins, the basic ice layers and frost crystals, the speed of ultrasonic will change at the interfaces because of the different response to the ultrasonic frequencies, then, the three different interfaces will vibrate asynchronously, so as to form high frequency relative vibrations. The difference of ultrasonic speed form relative shearing forces between the basic ice layer and fins, as well as between the basic ice layer and frost crystals/layers. With the thin ice layer and high surface adhesion, the basic ice layer cannot be removed by using ultrasonic vibrations, however, frost crystals have a small area to adhere to the ice layer which makes a weak surface adhesion, frost crystals will be fractured and fall off with the influence of gravity under the powerful shearing force, in all, conclusion can be drawn that the power density of the mechanical vibrations induced by ultrasonic transmit in the evaporator is large enough to fracture frost crystals. From the above analysis, the mechanism of ultrasonic frost suppression/ defrosting technology may be mainly attributed to the mechanical vibrations of power ultrasonic to fracture frost crystals and frost branches from the interface of frost crystals and basic ice layers, and with the influence of the gravity, the broken frost will fall off and form piles of frost accumulation under the evaporator. 5 Conclusions A refrigeration experimental system based on a compressor was custom built, and frosting experiments with/without the influence of ultrasonic vibrations were conducted. As can be seen through the microscopic observation and macroscopic view, the basic ice layers on the fins surface that formed with frozen condensate droplets cannot be removed by using ultrasonic vibrations for the strong adhesive force, and the thickness of the basic ice layer increases with experimental time flows. Frost crystals and frost branches growing on the basic ice layer can be vibrated and fractured to fall off from the fins surface with ultrasonic vibration, which shows that ultrasonic vibration can suppress the frost growing on the surface of a finned-tube evaporator evidently, especially in the period of frost crystals’ formation and growth. However, it is not so effective to exert ultrasonic vibration when the fins are blocked when the adhesive force of the frost to the fins surface is thus greatly increased. Therefore, the authors suggest that ultrasonic vibrations should be exerted from the early frosting period. The mechanism of power ultrasonic frost suppression is mainly attributed to the mechanical vibration effect, the vibration with high frequency and large acceleration can form a stupendous shearing force to fracture frost crystals and frost branches, so as to suppress the frost growing on the fins surface. The fractured frost will fall off with the gravity and form frost accumulation with frost piles under the evaporator. Although ultrasonic vibration can suppress frosting on the fins to some extent, it cannot ensure completely no frost accumulated on the evaporator, some other methods (e.g., air force) should be exerted to coordinate with ultrasonic vibration to improve the performance of frost suppression or defrosting. The mechanical structure for transferring vibration will influence the amplitude and distribution of vibration, so as to influence the frost suppression/defrosting effect, therefore, the vibration transfer structure should be improved to enhance the performance of ultrasonic vibration transferring in the finned-tube evaporator in the latter study. Only a BLT with a fixed frequency is not so effective, the method of frequency sweep or vibrations cooperation with two ultrasonic transducers may be useful to improve the effect of frost suppression/defrosting. In addition, a suitable working mode of ultrasonic vibration should be studied for energy conservation. It is feasible to observe and distinguish the effect on frosting/defrosting before fins blocked with frost by using a microscopic image system, however, with the influence of ultrasonic vibration, there is only a little frost deposited on the fins between any two neighboring tubes, but some fins above the tubes will be accumulated with frost balls at the later frosting period. It is not sufficient to examine the effect on frosting only by using the changes of frost thickness. Therefore, this study does suggest that the technique of image difference detection may be used to obtain the differences of different frosting images between and after ultrasonic vibrations to evaluate the effect of ultrasonic frost suppression. Besides, ultrasonic defrosting experiments on the fins with hydrophobic coatings should be studied in future. Maybe, it will have a better frost suppression effect than the fins surface is covered with hydrophilic coatings. Acknowledgment The authors wish to acknowledge the financial support and the technical collaborations with Henan Xinfei Electric Co. Ltd. (Xinfei) for their help in developing the experimental apparatuses. References [1] Chin-Hsiang Cheng Chiuan-Che Shiu Frost formation and frost crystal growth on a cold plate in atmospheric air flow Int. J. Heat Mass Transfer 45 2002 1289 4303 [2] Y. Hayashi A. Aoki S. Adachi K. Hori Study of frost properties correlating with frost formation types J Heat Transfer 99 1977 239 245 [3] W.D. Bascom R.L. Cottington C.R. Singleterry Ice adhesion to hydrophilic and hydrophobic surfaces J. Adhes. Oct. 1 1969 246 263 [4] X.M. Wu L. Ralph R.L. Webb Investigation of the possibility of frost release from a cold surface Exp. Therm. Fluid Sci. 24 2001 151 156 [5] ChinHsiang Cheng ChiuanChe Shiu Oscillation effects on frost formation and liquid droplet solidification on a cold plate in atmospheric air flow Int. J. Refrig. 26 2003 69 78 [6] Kazunari Adachi, Kazushi Saiki and Hiroki Sato, Suppression of frosting on a metal surface using ultrasonic vibrations, in: IEEE Ultrasonics Symposium, 1998, pp. 759–762. [7] Kazunari Adachi Kazushi Saiki Hiroki Sato Ultrasonic frost suppression Jpn. J. Appl. Phys. Part 1-Regular Papers Short Notes Rev. Papers 42 2003 682 685 [8] Qinlao Yan Lin Zhu Ningxia Yan Study on ultrasonic defrost technology of refrigeration fan J. Agric. Machine 34 2003 74 75 [9] Linda Barelli, Gianni Bidnni, Simona Moraglia, Development of an innovative defrosting system for commercial chiller evaporators through piezoelectric elements application, in: Proceedings of IMECE04. 2004 ASME International Mechanical Engineering Congress and Exposition, 2004, pp. 21–26. [10] LI Dong CHEN Zhenqian Effect of ultrasound on restraining frost formation on flat surface CIESC J. 60 2009 2171 2176 [11] Dingyuan Wang, Tangfei Tao, Shaoying Kang, Guanghua Xu, Non-contact frost thickness measurement by using a micro-camera and image processing technology, in: 2010 IEEE International Conference on Mechatronics and Automation (ICMA), 2010, pp. 288–293.