Professional Documents
Culture Documents
2014 K Tang Heat Transfer of Laminar Oscillating Flow in Finned Heat Exchanger of Pulse Tube Refrigerator
2014 K Tang Heat Transfer of Laminar Oscillating Flow in Finned Heat Exchanger of Pulse Tube Refrigerator
a r t i c l e i n f o a b s t r a c t
Article history: In order to characterize the heat transfer of heat exchangers employed by pulse tube refrigerators, an
Received 1 September 2012 experimental apparatus was built to investigate the heat transfer performance of a water-cooled finned
Received in revised form 6 June 2013 heat exchanger operating in laminar oscillating flow. The test results summarized into the Nusselt
Accepted 23 November 2013
number with respect to the maximum Reynolds number and the Valensi number were presented. The
Available online 18 December 2013
increases in the maximum Reynolds number and the Valensi number both lead to a rise in the Nusselt
number. The comparisons of the experimental results and the available typical correlations were
Keywords:
conducted and discussed. A new correlation of the Nusselt number to the maximum Reynolds number
Oscillating flow
Heat transfer
and the Valensi number was proposed, with which the results have a maximum deviation of 6.3%
Heat exchanger compared with the experimental values.
Nusselt number Ó 2013 Elsevier Ltd. All rights reserved.
Pulse tube refrigerator
0017-9310/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.ijheatmasstransfer.2013.11.067
812 K. Tang et al. / International Journal of Heat and Mass Transfer 70 (2014) 811–818
Nomenclature
number correlation to Reynolds number. Wakeland and Keolian exchanger. As schematically shown in Fig. 1, the experimental
[18] introduced a concept of ‘‘effectiveness’’ to analyze the heat apparatus consists of a linear compressor, a tested heat exchanger,
exchanger’s performance in an oscillating flow. The space-cycle a needle valve and a reservoir.
averaged Nusselt number is consistent with the conventional The pressure-wave generator is a linear compressor of type
understanding of the Nusselt number, and can be directly used 2S132W, fabricated by QDrive for driving a Stirling-type pulse tube
for designing the heat exchangers working in an oscillating flow. refrigerator. The mean electricity input power of the compressor is
Additionally, the visualization techniques of temperature field 500 W, powered by an AC power supply of type PCR2000M fabri-
[19–22] have been developed and applied in the study of the heat cated by Kikusui Electronics Corp., with the operating parameters
transfer in an oscillating flow. The variation in the temperature of AC 1–270 V, 20 A, 2000 VA, and 40–500 Hz. The jacket of the
field during one cycle could be measured to summarize the compressor is cooled by water to prevent the overheating,
space-cycle heat transfer characteristics. especially for the case with large working current. The maximum
Following the second strategy mentioned above, this study
focused on the heat transfer of a finned heat exchanger usually ap-
plied as the ambient-temperature heat exchanger in a pulse tube
refrigerator. An experimental apparatus was built according to
the operating conditions of the pulse tube refrigerators, which will
be presented in next section in details. Based on the measured
data, the space-cycle averaged Nusselt numbers (shortened as
‘‘Nusselt number’’ for simplicity in the following context) were
calculated and plotted with respect to the maximum Reynolds
number and the Valensi number, which are the important and
widely used similarity criterions for the oscillating flow
[3,4,8,11,14,16–18]. Upon comparing our results with some typical
reported works, a new expression was proposed to correlate the
Nusselt number to the maximum Reynolds number and the
Valensi number.
of the heat exchanger. The constant current supply and the voltage
signals of the thermometers (T3, T4 and T5) are transmitted be-
tween the atmospheric surroundings and the high pressure work-
ing gas through the electric connectors. Additionally, the K-type
armored thermocouples (T6, T7, T8 and T9) of 0.5 mm in diameter
are inserted into the flanges of the heat exchanger to measure the
wall temperature. The distance from the tips of thermocouples to
the fin base is 4 mm. The symmetrical arrangement of these ther-
mocouples allows us to calculate the fin-base temperature with
the average of T6, T7, T8 and T9. The temperature differences
among all these nine thermometers are corrected to ±0.1 K by cal-
ibrating them simultaneously in a variable-temperature water
bath, since the present measurement is quite sensitive to the
accuracy of temperature differences instead of the absolute tem-
perature accuracy.
The pressure measurement employs two UNIK-5000-type pres-
sure sensors, P1 and P2, fabricated by DRUCK Ltd., with the range
of 0–5 MPa and the accuracy of 0.2% FS (10 kPa). Since the pressure
amplitude in the reservoir is the key parameter for calculating the
velocity amplitude in the tested heat exchanger, a dynamic pres-
Fig. 2. Schematic of the finned heat exchanger. sure sensor of model 106B manufactured by PCB PIEZOTRONICS
Inc., with an accuracy of 0.57 kPa, is used to measure the dynamic
pressure in the reservoir (i.e. P3 in Fig. 1). In addition, the flow rate
working current is restricted below 10 A in our experiments to en- of the cooling water is measured by a mini turbine flowmeter with
sure safe operation. an accuracy of 0.005 l/min.
Fig. 2 presents the detailed structure of the tested copper heat All electric signals of abovementioned sensors are transmitted
exchanger. The cooling water enters the heat exchanger from the into the computer via CompactDAQ devices of National Instru-
top, and is then divided into two streams, which flow through ments and 34970A Data Acquisition Unit of Agilent.
the channels surrounding the heat exchanger, converge at the bot-
tom and flow out of the heat exchanger. The parallel fins, fabri- 2.3. Calculation of maximum Reynolds number, Valensi number and
cated by electrical discharge machining, construct the channels Nusselt number
for the oscillating working gas. The fins have a thickness of
0.75 mm and a uniform length of 20 mm, and their heights range The maximum Reynolds number Remax is expressed as
from 2.6 mm to 13.6 mm. The fin-to-fin gap is 0.75 mm. The tested
heat exchanger, as well as the connecting tubes for the working gas umax dh
Remax ¼ ; ð1Þ
and the cooling water, is insulated by polystyrene foam to reduce m
the influence of the heat transfer with the surroundings. where umax is the velocity amplitude of the oscillating working gas
A combination of needle valve and reservoir, usually for mea- in the channel of the tested heat exchanger, dh is the hydraulic
suring the acoustic power output from a thermoacoustic engine diameter of the channel, and m is the kinematic viscosity of the
[23,24], is adopted to adjust and measure the velocity amplitude working gas. For the parallel-plate channels constructed from the
in the heat exchanger. The tests of the heat transfer under different fins, dh is twice the fin-to-fin gap. umax can be calculated by
maximum Reynolds numbers are conducted by adjusting the
opening of the needle valve. The velocity amplitude in the heat ex- U max pA3 xC res 2pfpA3 V res
umax ¼ ¼ ¼ ; ð2Þ
changer can be calculated from the measured pressure amplitude Aflow Aflow Aflow cpm2
in the reservoir.
where Umax is the volume velocity amplitude of the oscillating
working gas, Aflow is the flow area in the heat exchanger, pA3 is
2.2. Measurement the pressure amplitude in the reservoir measured by pressure sen-
sor P3, x is the angular frequency, Cres is the acoustic compliance of
Nine thermometers, three pressure sensors and one flowmeter the reservoir [23], f is the frequency, Vres is the volume of the reser-
are employed in the experiments, whose arrangement can be voir, c is the specific heat ratio of the working fluid, pm2 is the mean
found in Figs. 1 and 2. operating pressure in the reservoir measured by pressure sensor P2.
The temperatures of the cooling water flowing into and out of The Valensi number Va is formulated as
the heat exchanger are measured by two sheathed PT100 platinum
2
resistance thermometers of T1 and T2, which are immerged into xd2h 2dh
2
dh
Va ¼ ¼ pffiffiffiffiffiffiffiffiffiffiffiffi ¼2 ; ð3Þ
the cooling water through two Swagelok tee-fittings, respectively. m ð 2m=xÞ
2 dm
The temperatures of the working gas at the so-called inlet and out-
let of the heat exchanger, as well as the temperature of working which is the ratio of the time scale of viscous penetration to the
gas in the reservoir, are measured by three PT100 platinum resis- oscillation period, also twice the squared ratio of the hydraulic
tance thermometers of T3, T4 and T5 with 2 mm in diameter and diameter to the viscous penetration depth dm.
7 mm in length. Two cone chambers are adjacent to the edges of The Nusselt number is defined as
the heat exchanger to connect the heat exchanger and the connect-
hdh
ing tubes, and the working gas from all of the flow passages Nu ¼ ; ð4Þ
constructed by the fins is mixed in the cone chambers. The ther-
k
mometers T3 and T4 are located inside these two cone chambers where h is the convective heat transfer coefficient of the oscillating
to measure the temperature of the mixed gas, which is regarded flow, k is the thermal conductivity of the working gas.
as the mean temperature of the working gas flowing into and out The convective heat transfer coefficient h can be expressed as
814 K. Tang et al. / International Journal of Heat and Mass Transfer 70 (2014) 811–818
0.04
tanhðmHcor Þ
gfin ¼ ; ð11Þ
mHcor
0.03
sffiffiffiffiffiffiffiffiffiffiffiffiffiffi -5 2 -2
y=-5.43x10 x +3.556x10 x-5.77
2h
m¼ ; ð12Þ 0.02 2
R =0.977
kCu dfin
dfin 0.01
Hcor ¼ H þ ; ð13Þ 300 305 310 315 320 325 330
2
th (K)
where dfin and H are the thickness and the height of the fin,
respectively. Fig. 3. (Qh Qc)/Qh as a function of th.
K. Tang et al. / International Journal of Heat and Mass Transfer 70 (2014) 811–818 815
9 9
8 8
7 7 3
2
6 3 6 2
5 5
1 1
Nu
Nu
4 4
3 3
2 5 2 5 2.5 MPa, Va=250
1 2.5 MPa, Va=150 1 3.0 MPa, Va=250
4 4
3.0 MPa, Va=150 3.5 MPa, Va=250
0 0
0 200 400 600 800 1000 1200 1400 0 200 400 600 800 1000 1200 1400
Remax Remax
Fig. 4. Nusselt number Nu as a function of the maximum Reynolds number Remax Fig. 6. Nusselt number Nu as a function of the maximum Reynolds number Remax
with the Valensi number Va = 150. with the Valensi number Va = 250.
9 9
8 8
3
7 7
3
6 6 2
2
5 5
1
Nu
Nu
4 1 4
3 3
2 5 2 5
1 2.5 MPa, Va=200 1 3.0 MPa, Va=300
4 3.0 MPa, Va=200 4 3.5 MPa, Va=300
0 0
0 200 400 600 800 1000 1200 1400 0 200 400 600 800 1000 1200 1400
Remax Remax
Fig. 5. Nusselt number Nu as a function of the maximum Reynolds number Remax Fig. 7. Nusselt number Nu as a function of the maximum Reynolds number Remax
with the Valensi number Va = 200. with the Valensi number Va = 300.
o
4
3 With Ao = 2Remax/Va and Rex = Va, Eq. (23) can be rewritten as
2 5 Nu ¼ 0:036Re0:85
max Va
0:27
: ð24Þ
1 Nsofor et al. [17] investigated the heat transfer of a finned heat
4 3.5 MPa, Va=350
exchanger in a thermoacoustic refrigerator under the conditions of
0
0 200 400 600 800 1000 1200 1400 the mean operating pressure from 0.3 MPa to 0.8 MPa and the fre-
quency from 300 Hz to 450 Hz with helium as its working fluid.
Remax
The parallel-plate channels were constructed from the copper fins
Fig. 8. Nusselt number Nu as a function of the maximum Reynolds number Remax
of 0.15 mm in thickness with the fin-to-fin gaps of 0.95 mm. Based
with the Valensi number Va = 350. on the 24 tested Nusselt numbers, a correlation was obtained as,
Nu ¼ 0:61Re0:31
rms Pr
0:11
: ð25Þ
3.2. Comparison between our experimental results and the reported pffiffiffi
For the root mean square Reynolds number Rerms ¼ Remax = 2,
correlations
Eq. (25) can be expressed as
The calculation methods of the Nusselt number for the oscillat- Nu ¼ 0:548Re0:31 0:11
max Pr : ð26Þ
ing flow, reported in the existing literatures, were developed
mainly via three approaches, i.e., citing the available steady-flow The calculation results of Eqs. (19), (20), (22), (24), and (26) are
heat transfer correlations [10,11,29], developing simple physical also presented in Figs. 4–8, with the corresponding labels of 1–5,
models [30] and conducting experimental tests of the oscillating respectively.
flow [14,16,17]. The correlations of TASFE and RMSRe models based on the stea-
Typical approaches based on the steady-flow heat transfer cor- dy flow can reflect the rise in the Nusselt number with the increas-
relations are so-called ‘‘Time-Average Steady-Flow Equivalent’’ ing maximum Reynolds number. However, omitting the influence
(TASFE) approximation [29] and ‘‘Root Mean Square Reynolds of Valensi number leads to an increasing deviation with the rising
Number’’ (RMSRe) model [10]. The former takes the time average Valensi number. The higher tested Nusselt numbers than the re-
of the known steady-flow Nusselt number correlation over one cy- sults of TASFE and RMSRe models may imply the heat transfer
cle to obtain the correlation for the oscillating flow. The latter sub- enhancement by the oscillating flow, compared with the steady
stitutes the Reynolds number figuring in the known steady-flow flow.
correlation with the root mean square Reynolds number calculated In contrast with TASFE and RMSRe models, the correlation
with the oscillating fluid velocity. Piccolo and Pistone [11] pre- based on Swift’s simple boundary layer model can reflect the im-
sented two Nusselt number correlations, as given in Eqs. (19) pact of Valensi number due to the function of thermal penetration
and (20), based on the Hausen correlation [25] for laminar flow depth and frequency. A rise in Valensi number derived from the in-
in the tube entrance region for TASFE and RMSRe models, crease in the frequency means a reduction in the boundary layer
respectively, thickness, which enhances the heat transfer indicated by the
increasing Nusselt number. However, the influence of maximum
Z p=x ( )
x 0:0668ðdh =LÞPrRemax sin xt Reynolds number is ignored so that the deviations of Nusselt num-
Nu ¼ 3:66 þ dt; ð19Þ ber with the variation of maximum Reynolds number are visible.
p 0 1 þ 0:04½ðdh =LÞPrRemax sin xt
2=3
The results of the correlation by Zhao and Cheng, Eq. (24), can
pffiffiffi also indicate the rise in Nusselt number with the increasing max-
0:0668ðdh =LÞPrðRemax = 2Þ imum Reynolds number. The negative exponent of the Valensi
Nu ¼ 3:66 þ pffiffiffi 2=3 : ð20Þ
1 þ 0:04½ðdh =LÞPrðRemax = 2Þ Number leads to a slight decrease in Nusselt number with an in-
crease in Valensi Number. This variation is opposite to our test re-
Swift [30] suggested an approximate expression of the sults. The correlation by Nsofor et al. shows better agreement to
convective heat transfer coefficient as our experimental results than Eq. (24), however, the impact of
1 k the Valensi number is not yet considered.
h pffiffiffi ; ð21Þ Among the abovementioned correlations, Swift’s simple bound-
2 dj
ary layer model, i.e., Eq. (22), can achieve a relatively better agree-
based on a boundary layer conduction heat transfer model. k and dj ment to our experimental results with a maximum error of 21.5%,
are the thermal conductivity and the thermal penetration depth of however, it is not yet satisfactory. For application convenience in
the working fluid, respectively. From Eqs. (4) and (21), the Nusselt the heat exchanger design, we adopt a typical expression as Eq.
number can be expressed as (27) to fit our experimental Nusselt number,
hdh 1 dh 1 dh 1 pffiffiffiffiffiffiffiffiffiffiffi m
Nu ¼ bRemax Van : ð27Þ
Nu ¼ ¼ pffiffiffi ¼ pffiffiffi pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi ¼ VaPr: ð22Þ
k 2 jd 2 2m=ðxPrÞ 2
We obtain m = 0.0876 from the linear fitting of log(Nu) and
Zhao and Cheng [14] correlated the Nusselt number to the log(Remax) with a coefficient of determination of R2 = 0.946, based
dimensionless oscillation amplitude of fluid Ao (Ao = xmax/d, where on the data for Va = 250, as shown in Fig. 9. Then, all 84 data points
xmax is the peak-to-peak displacement amplitude, d is the inside are linearly fitted for logðNu=Re0:0876
max Þ and log(Va) with a coefficient
K. Tang et al. / International Journal of Heat and Mass Transfer 70 (2014) 811–818 817
0.90 4. Conclusions
2.5 MPa, Va=250
3.0 MPa, Va=250 An experimental apparatus was built to investigate the heat
0.85 3.5 MPa, Va=250 transfer of a finned heat exchanger operating in the laminar oscil-
lating flow, aiming for the application of the ambient-temperature
heat exchangers of pulse tube refrigerators. The experiments
focused on the Nusselt number Nu with respect to the maximum
log(Nu)
0.80 Reynolds number Remax and the Valensi number Va. The test
y=0.0876x+0.6 results indicate that for Remax = 200 1200 and Va = 150 350,
2
R =0.946 Nu ranges in 5.0–8.3, and that the increases in Remax and Va can
0.75 both lead to a rise in Nu. The comparison between our experimen-
tal results and the typical correlations reported in the literatures
shows that the boundary layer conduction heat transfer model,
0.70 Eq. (22), achieves a relatively better agreement with our experi-
2.2 2.4 2.6 2.8 3.0 3.2 mental results (with a maximum deviation of 21.5%). A new
log(Remax) correlation of Eq. (27), including the effects of both Remax and
Va, was proposed and achieved a maximum deviation of 6.3%
Fig. 9. log(Nu) as a function of log(Remax) with Va = 250. compared to the experimental results, which means the feasibility
in aiding the design of the heat exchangers employed in pulse
tube refrigerators.
0.70 More working fluids (such as nitrogen, argon and carbon diox-
2.5 MPa ide), as well as different operating pressure ratio, are expected to
3.0 MPa be tested in the future work to verify the applicability of the above
0.65 3.5 MPa correlation, including the effect of Prandtl number on the heat
transfer performance. The investigation of the impact of heat
)
0.0876
0.55
Acknowledgement
0.50 y=0.405x-0.367 The project is financially supported by the National Natural Sci-
2
R =0.964 ence Foundation of China (51176167 and 50890182).
0.45
2.1 2.2 2.3 2.4 2.5 2.6 References
log(Va)
[1] P. Nika, Y. Bailly, F. Guermeur, Thermoacoustics and related oscillatory heat
Fig. 10. logðNu=Re0:0876 and fluid flows in micro heat exchangers, Int. J. Heat Mass Transfer 48 (18)
max Þ as a function of log(Va).
(2005) 3773–3792.
[2] J. Liu, S.L. Garrett, Relationship between Nusselt number and the
thermoviscous (Rott) functions, J. Acoust. Soc. Am. 119 (3) (2006) 1457–
8.5 1462.
[3] Y. Chen, E. Luo, W. Dai, Heat transfer characteristics of oscillating flow
2.5 MPa
8.0 regenerator filled with circular tubes or parallel plates, Cryogenics 47 (1)
3.0 MPa
+10% (2007) 40–48.
7.5 3.5 MPa [4] Y. Chen, E. Luo, W. Dai, Heat transfer characteristics of oscillating flow
regenerators in cryogenic temperature range below 20 K, Cryogenics 49 (7)
(2009) 313–319.
7.0 [5] Y. Wang, Y.L. He, Q. Li, G.H. Tang, Numerical simulations of gas resonant
-10%
Nucal
oscillations in a closed tube using lattice Boltzmann method, Int. J. Heat Mass
6.5 Transfer 51 (11–12) (2008) 3082–3090.
[6] A.A. Kornhauser, J.L. Smith, Heat transfer with oscillating pressure and
6.0 oscillating flow, in: W.D. Jackson (Ed.), Proceedings of the 24th Intersociety
Energy Conversion Engineering Conference, IEEE, New York, 1989, pp. 2347–
5.5 2353.
[7] A.A. Kornhauser, J.L. Smith, Application of a complex Nusselt number to heat
5.0 transfer during compression and expansion, J. Heat Transfer Trans. ASME 116
(3) (1994) 536–542.
[8] P. Bouvier, P. Stouffs, J.P. Bardon, Experimental study of heat transfer in
4.5
4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0 8.5 oscillating flow, Int. J. Heat Mass Transfer 48 (12) (2005) 2473–2482.
[9] G.W. Swift, Thermoacoustic engines, J. Acoust. Soc. Am. 84 (4) (1988) 1145–
Nuexp 1180.
[10] G.W. Swift, Thermoacoustics: A Unifying Perspective for Some Engines and
Refrigerators, Acoustical Society of America, American Institute of Physics,
Fig. 11. Comparison of the proposed correlation prediction and the experimental
New York, 227, 2002, pp. 99–103.
data. [11] A. Piccolo, G. Pistone, Estimation of heat transfer coefficients in oscillating
flows: the thermoacoustic case, Int. J. Heat Mass Transfer 49 (9–10) (2006)
1631–1642.
of determination of R2 = 0.964, as shown in Fig. 10. As a result,
[12] F. Meng, M. Wang, Z. Li, Lattice Boltzmann simulations of conjugate heat
b = 0.43 and n = 0.405. transfer in high-frequency oscillating flows, Int. J. Heat Fluid Flow 29 (4)
The calculated Nusselt numbers with Eq. (27) are compared (2008) 1203–1210.
with the experimental data in Fig. 11, which shows that the devi- [13] W.L. Cooper, V.W. Nee, K.T. Yang, An experimental investigation of convective
heat transfer from the heated floor of a rectangular duct to a low frequency,
ations are less than 10%, and the absolute value of the maximum large tidal displacement oscillating flow, Int. J. Heat Mass Transfer 37 (4)
deviation is 6.3%. (1994) 581–592.
818 K. Tang et al. / International Journal of Heat and Mass Transfer 70 (2014) 811–818
[14] T.S. Zhao, P. Cheng, Oscillatory heat transfer in a pipe subjected to a [22] A.J. Jaworski, A. Piccolo, Heat transfer processes in parallel-plate heat
laminar reciprocating flow, J. Heat Transfer Trans. ASME 118 (3) (1996) exchangers of thermoacoustic devices – numerical and experimental
592–597. approaches, Appl. Therm. Eng. 42 (2012) 145–153.
[15] J.R. Brewster, R. Raspet, H.E. Bass, Temperature discontinuities between [23] K. Tang, Z.J. Huang, T. Jin, G.B. Chen, Influence of acoustic pressure amplifier
elements of thermoacoustic devices, J. Acoust. Soc. Am. 102 (6) (1997) dimensions on the performance of a standing-wave thermoacoustic system,
3355–3360. Appl. Therm. Eng. 29 (5–6) (2009) 950–956.
[16] K.C. Leong, L.W. Jin, An experimental study of heat transfer in oscillating flow [24] M.E.H. Tijani, S. Spoelstra, A high performance thermoacoustic engine, J. Appl.
through a channel filled with an aluminum foam, Int. J. Heat Mass Transfer 48 Phys. 110 (9) (2011) 093519.
(2) (2005) 243–253. [25] F.P. Incropera, D.P. DeWitt, T.L. Bergman, A.S. Lavine, Fundamentals of Heat
[17] E.C. Nsofor, S. Celik, X.D. Wang, Experimental study on the heat transfer at the and Mass Transfer, sixth ed., Wiley, New York, 2007. pp. 147–155, 512–
heat exchanger of the thermoacoustic refrigerating system, Appl. Therm. Eng. 513.
27 (14–15) (2007) 2435–2442. [26] P. Merkli, H. Thomann, Transition to turbulence in oscillating pipe flow, J. Fluid
[18] R.S. Wakeland, R.M. Keolian, Effectiveness of parallel-plate heat exchangers in Mech. 68 (3) (1975) 567–575.
thermoacoustic devices, J. Acoust. Soc. Am. 115 (6) (2004) 2873–2886. [27] M. Hino, M. Sawamoto, S. Takasu, Experiments on transition to turbulence in
[19] C. Herman, E. Kang, Experimental visualization of temperature fields and study an oscillatory pipe-flow, J. Fluid Mech. 75 (2) (1976) 193–207.
of heat transfer enhancement in oscillatory flow in a grooved channel, Heat [28] T.S. Zhao, P. Cheng, Experimental studies on the onset of turbulence and
Mass Transfer 37 (1) (2001) 87–99. frictional losses in an oscillatory turbulent pipe flow, Int. J. Heat Fluid Flow 17
[20] C. Herman, M.P. Cetinguel, Experimental visualization of the thermoacoustic (4) (1996) 356–362.
effect, J. Heat Transfer Trans. ASME 131 (8) (2009) 080905. [29] P.D. Richardson, Effects of sound and vibration on heat transfer, Appl. Mech.
[21] L. Shi, Z. Yu, A.J. Jaworski, Application of laser-based instrumentation for Rev. 20 (3) (1967) 201–217.
measurement of time-resolved temperature and velocity fields in the [30] G.W. Swift, Analysis and performance of a large thermoacoustic engine, J.
thermoacoustic system, Int. J. Therm. Sci. 49 (9) (2010) 1688–1701. Acoust. Soc. Am. 92 (3) (1992) 1551–1563.