Impingement heat/mass transfer to hybrid synthetic jets and other reversible pulsating jets

Abstract

This study focused on round, hybrid synthetic (non-zero-net-mass-flux) jets impinging on a wall. To complete this study, two additional variants of reversible pulsating jets were investigated, namely synthetic (zero-net-mass-flux) jets and mixed pulsed jets (pulsed jets containing an additional blowing component). For comparison purposes, the continuous jet was used. The working fluid was air. The Reynolds numbers ranged from 4000 to 6000, the dimensionless stroke lengths were 14–16, and the dimensionless orifice-to-wall distances were 2–16 (both related to the orifice exit diameter of 8 mm). The experiments used flow visualization, single-sensor hot-wire measurements, and mass transfer measurements on the wall using the naphthalene sublimation technique. The local heat transfer coefficient, expressed as a Nusselt number, was evaluated using the heat/mass transfer analogy.

All tested jets exhibited relatively flat, nearly top-hat velocity profiles. An increase of the Reynolds number by an additional blowing component resulted in a heat/mass transfer enhancement. The flow oscillations (for the present geometry and driven parameters) caused a heat/mass transfer enhancement of 12–40%.

The main outcome was that the 26% larger flow rate of the hybrid synthetic jet, versus the conventional synthetic jet, resulted in an 18% increase in the heat transfer rate. These gains were caused by a partial rectification effect of incorporated fluidic diodes, without consuming any additional energy and without introducing movable parts.

Introduction

Submerged impinging jets (IJs) and impingement heat and/or mass transfer on exposed walls have been widely studied in the past. Because of their high heat and mass transfer values in single-phase flows, there are large number of applications for IJs. This has resulted in significant research. The most important early results were collected in monographs, such as the outstanding book by Dyban and Mazur [1] and the distinguished work by Martin [2]. Since then, several comprehensive reviews have appeared, e.g. [3], [4], [5], [6], [7]. The majority of previous studies focused on continuous (steady-flow) IJs. Despite the ability of steady IJs to achieve very high heat fluxes, further intensification of the transport processes seems possible through an incorporation of unsteadiness effects. However, pulsations do not automatically lead to enhanced rates of heat transfer; the effect can sometimes be found to increase the transfer, sometimes to decrease it, or to have a negligible effect – see, e.g. Herman [8], Herwig et al. [9], and Persoons et al. [10].

There are many ways to generate a pulsatile character of IJs, which utilize mechanical vibration/motion, alternate blowing, and fluidic oscillators with no moving parts. For example, when Page et al. [11] used a straight circular pipe nozzle with a collar to generate a self-oscillating IJ, they concluded that its use resulted in a 45% enhancement of the impingement heat transfer at a rather high orifice velocity of 153 m/s. Camci and Herr [12] used a fluidic oscillating nozzle to generate a self-oscillating IJ with a periodic flapping motion, which, based on their conclusions, resulted in a 40–70% greater heat transfer rate than that from a continuous IJ.

Liu and Sullivan [13] investigated the heat transfer to excited round IJs at relatively small nozzle-to-wall spacings. They obtained either an enhancement or reduction of the local heat transfer by controlling the development of the vortex structure by forcing the excitation frequency to be near the natural frequency of the unexcited free jet or its subharmonic frequency, respectively.

Gau et al. [14] concluded that excitation at the natural frequencies of the unexcited free jet (at the first or second subharmonic frequency) can enhance impingement heat transfer by increasing turbulence intensity, while other frequencies can reduce turbulent intensity and decrease heat transfer.

Hwang et al. [15] investigated the flow characteristics and heat transfer to an actively controlled axisymmetric IJ. The frequency of the excitation was chosen to correspond with the natural frequency of the unexcited free jet. The harmonic or double harmonic excitation caused the promotion or suppression of vortex pairing, and measurable enhancement or reduction in heat transfer, respectively. Correspondingly, a region of maximum heat transfer moved further or closer to the orifice.

The properties of a round IJ with the target placed relatively close to the nozzle and under simultaneous excitations at two frequencies was investigated by Vejražka et al. [16]. The sensitivity of vortex roll-up processes to excitations was found to occur at Strouhal numbers ranging from 0.56 to 2.4.

Trávníček and Tesař [17] measured acoustically excited annular IJs. Two different regimes of the time-mean flow were found, each differing in the size of the recirculation region. Either a small recirculation bubble with the central stagnation point on the impingement wall or a large recirculation bubble with the stagnation circle on this wall occurs. The acoustic excitation tends to transform the small recirculation bubble into the large one over the entire investigated Strouhal number range of 0.38–2.47. The effective stabilization of the large recirculation bubble leads to a large increase, compared with an unexcited IJ, in the average heat/mass transfer, with the maximum augmentation of 23% occurring at a Strouhal number of 0.94.

A synthetic jet (SJ) is a fluid jet flow that is created (synthesized) during an oscillatory process of suction and blowing between a SJ actuator cavity and its surroundings – Smith and Glezer [18]. If one boundary of the cavity is a deformable diaphragm or a piston, the pressure generated when the volume of the cavity is reduced by the motion of the boundary results in a flow out of the cavity; conversely, the external ambient pressure produces a flow into the cavity when the volume is increased. The time-mean mass flux through this orifice is zero, therefore a SJ is frequently known as a zero-net-mass-flux jet – Cater and Soria [19]. Another term used for the same phenomenon is the oscillatory vorticity generator – Yehoshua and Seifert [20].

The first SJ actuator, as we think of it today, was most likely a laboratory air-jet generator designed and used by Dauphinee in 1957 [21]. Another, more recent study from 1986 was a conference paper by Yassour et al. [22] focusing on convective heat transfer on a wall exposed to an annular impinging SJ. The term synthetic jet was introduced by James et al. [23]. Since that time, experimental, theoretical, and numerical investigations have been undertaken – e.g. Kral et al. [24], Lee and Goldstein [25], Mallinson et al. [26], Glezer and Amitay [27], Gallas et al. [28], Smith and Swift [29], and Kordík and Trávníček [30]. Formation criteria and flow field regimes of SJs were studied by Holman et al. [31], Zhou et al. [32], Xia and Zhong [33], Trávníček et al. [34], and McGuinn et al. [35].

One desirable feature of SJ actuators is its relative simplicity, as neither blower nor fluid supply piping is required. SJs are promising alternatives for various applications of active flow control such as jet vectoring and exciting (or shaking or flapping) and flow control in external and internal aerodynamics. Another potential field is SJ application in heat transfer [36], [37], [38], [39], [40], [41], [42], [43], [44]. The majority of these studies have focused on air SJ impinging on the wall [37], [39], [40], [41], [42], [43]. The wall was electrically heated, and heat transfer characteristics were evaluated from the heated wall as it cooled. Trávníček et al. [36] investigated another geometry, namely an annular impinging SJ, by means of mass transfer experiments, and taking into consideration an analogy between heat and mass transfer processes. Another arrangement, called the “SJ ejector”, was investigated by Mahalingam et al. [38]. A channel with two heated walls was cooled by SJs operating as a jet ejector, i.e. SJs placed alongside the walls induced a channel flow. Note that SJs were the sole driver of the fluid motion in all the studies [36], [37], [38], [39], [40], [41], [42], [43]. A fundamentally different case was proposed by Trávníček et al. [44], namely heat transfer characteristics of a continuous jet impingement under an active flow control. The main flow was the impinging continuous jet driven by a compressor, while the control flow was generated by a system of SJs. Particularly promising fields of heat transfer applications can be found on a micro scale, such as the cooling of highly loaded electronic components in microchannels – see Timchenko et al. [45], Lee et al. [46], [47] and Trávníček et al. [48].

A hybrid synthetic jet (HSJ) is a fluid jet flow produced partly by the same principle as a SJ and partly by a fluid pumping from valveless pumps [49], [50], [51]. A key element of the valveless pump is a fluidic diode. This fluidic element has a hydraulic resistance in one (forward) flow direction that is basically smaller than in the opposite (reverse) direction – Priestman and Tippetts [52], Tesař [53]. This feature causes a partial rectification effect: if the fluidic diode is connected to a source of symmetrical pressure oscillations, it partially rectifies flow into the forward direction. This effect enhances the volume flux during the actuator extrusion (pump) stroke without any additional power consumption. The resulting HSJ is a non-zero-net-mass-flux jet – see Trávníček et al. [54], [55], [56], [57], Hsu et al. [58], and Kordík and Trávníček [59]. The advantages and application possibilities of HSJs are qualitatively the same as those mentioned above for SJs. Moreover, HSJs achieve higher volume fluxes, thus higher momentum. Therefore, HSJs can achieve higher effects in applications of active flow control and heat transfer. Another HSJ advantage can be useful for cooling applications: the entrained fluid is taken via fluidic diodes from the colder surroundings far from the heat exchange zone, i.e. another fresh fluid without preheating is introduced into the cooling process.

Recently, several types of active heat sinks for highly loaded electronic components equipped with SJs have become commercially available. One motivating factor behind this study is the authors’ opinion that non-zero-net-mass-flux pulsating jets, namely an impinging HSJ, can be a useful alternative in various heat transfer applications. However, the heat transfer characteristics of impinging HSJ cannot be discovered in current literature, except for the preliminary results presented by Trávníček and Vít [62]. The principal objective of this study was to investigate the local heat transfer coefficient (expressed as a Nusselt number) of a HSJ impinging on a wall. To complete this approach, two variants of pulsating IJs, with a large, reversible velocity component, were investigated, namely a synthetic jet (SJ) and a mixed pulsed jet (MPJ) – i.e. a pulsed jet containing an additional blowing component – see Béra et al. [60]. In addition, a continuous jet (CJ) was used for comparison purposes. To characterize the flow fields of the jets, flow visualization and hot-wire anemometry were used. Local mass transfer on an exposed wall was measured using the naphthalene sublimation technique. Nusselt number distributions were evaluated using the heat/mass transfer analogy.