Convective heat transfer characteristics of laminar pulsating pipe air flow

 Heat transfer characteristics to laminar pulsating pipe flow under different conditions of Reynolds number and pulsation frequency were experimentally investigated. The tube wall of uniform heat flux condition was considered. Reynolds number was varied from 780 to 1987 while the frequency of pulsation ranged from 1 to 29.5 Hz. The results showed that the relative mean Nusselt number is strongly affected by pulsation frequency while it is slightly affected by Reynolds number. The results showed enhancements in the relative mean Nusselt number. In the frequency range of 1–4 Hz, an enhancement up to 30% (at Reynolds number of 1366 and pulsation frequency of 1.4 Hz) was obtained. In the frequency range of 17–25 Hz, an enhancement up to 9% (at Reynolds number of 1366 and pulsation frequency of 17.5 Hz) was indicated. The rate of enhancement of the relative mean Nusselt number decreased as pulsation frequency increased or as Reynolds number increased. A reduction in relative mean Nusselt number occurred outside these ranges of pulsation frequencies. A reduction in relative mean Nusselt number up to 40% for pulsation frequency range of 4.1–17 Hz and a reduction up to 20% for pulsation frequency range of 25–29.5 Hz for Reynolds numbers range of 780–1987 were considered. This reduction is directly proportional to the pulsation frequency. Empirical dimensionless equations have been developed for the relative mean Nusselt number that related to Reynolds number (750 < Re < 2000) and the dimensionless frequency (3<Ω<18) with about 10% rms. Received on 16 May 2000 / Published online: 29 November 2001     

Mixed convection from upward flow of hot air to a cooled vertical pipe

An experimental study had been carried out to investigate the buoyancy-opposed mixed convection from an upward flow of hot air to a vertical pipe with a cooled surface. The investigation covered a wide range of flow regime, ranging from the “free convection significant” to the “forced convection significant” conditions. Reynolds number of the flow extended from 966 to 14780, whereas the Buoyancy parameter, Ω [=Gr_d/(Re_d)²], varied from 0.008 to 2.77. A steady stream of hot air at a moderate pressure and a Prandtl number of 0.707 was arranged to flow upward through a vertical steel pipe, whose external wall was cooled uniformly by ambient air at 20°C. Test section of the vertical pipe was 1625 mm long with an internal diameter of 156 mm and an external diameter of 166 mm. Air temperature at inlet of the test section was varied from 40°C to 70°C. Both radial temperature and velocity profiles of the airflow were measured at inlet and exit of the test section respectively. Temperatures along the pipe wall were also measured. Non-dimensional expression for the prediction of the average heat transfer coefficient of the mixed convection from an upward flow of hot air to a vertical pipe with a cooled surface was developed from the experimental results. Convection heat transfer was found to impair when the flow is laminar and was enhanced for turbulent flow condition. Received on 20 July 1998     

Heat transfer by laminar flow in a vertical pipe with twisted-tape inserts

Heat transfer for laminar flow of water in an air-cooled vertical copper pipe with four twisted-tape inserts was determined experimentally. The tests were executed for laminar flow within 110 ≤ Re ≤ 1500, 8.1 ≤ Gz ≤  82.0 and 1.62 ≤ y ≤ 5.29. The correlation equation for heat transfer was defined for the tested range. The obtained results were compared to the results of other authors. Received on 28 April 1998     

An experimental investigation on performance of rectangular fins on a horizontal base in free convection heat transfer

This paper reports an experimental study of free convection heat transfer from rectangular fin-arrays on a horizontal base. An experimental set-up was constructed and calibrated, 15 sets of fin-arrays and a base plate without fins were tested in atmosphere. Fin height was varied from 6 mm to 26 mm, fin spacing was varied from 6.2 mm to 83 mm. The base-to-ambient temperature difference was also varied independently and systematically with the power supply to heater ranging from 8 W to 50 W. Fin length and fin thicknesses were fixed at 100 mm and 3 mm, respectively. The experimental program was conducted so as to clearly delineate the separate roles of fin height, fin spacing and base-to-ambient temperature difference. It was found that for a given base-to-ambient temperature difference the convection heat transfer rate from fin-arrays takes on a maximum value as a function of fin spacing and fin height. For a given base-to-ambient temperature difference the enhancement of the convection heat transfer rate of fin-arrays relative to that for base plate without fins is strongly dependent on the fin spacing to fin height ratio and number of fins. A correlation was also presented relating the convection heat transfer rate of fin-arrays relative to that for base plate without fins with the relevant non-dimensional parameters. Received on 7 August 1997     

Critical heat flux in forced convective boiling with a plane jet (Revised correlation for saturated condition)

 Critical heat flux (CHF) has been measured in saturated forced convective boiling with a wall jet on a rectangular heated surface of 40 and 80 mm in length and 20 mm in width. The jet velocity is varied from 3 to 15  m/s, and the system pressure is 0.1, 0.2, and 0.4 MPa for R113. It is found that the existing correlation for saturation condition can be applied to the CHF at high and low ρ_l/ρ_g values (e.g. water and R22), but hardly to the CHF at medium ρ_l/ρ_g values (e.g. R113 at 0.2 and 0.4 MPa). A revised correlation is proposed to predict most of the CHF data within an accuracy of ±25%. Received on 26 April 1999     

Experimental visualization of temperature fields and study of heat transfer enhancement in oscillatory flow in a grooved channel

An experimental study was conducted of incompressible, moderate Reynolds number flow of air over heated rectangular blocks in a two-dimensional, horizontal channel. Holographic interferometry combined with high-speed cinematography was used to visualize the unsteady temperature fields in self- sustained oscillatory flow. Experiments were conducted in the laminar, transitional and turbulent flow regimes for Reynolds numbers in the range from Re = 520 to Re = 6600. Interferometric measurements were obtained in the thermally and fluiddynamically periodically fully developed flow region on the ninth heated block. Flow oscillations were first observed between Re = 1054 and Re = 1318. The period of oscillations, wavelength and propagation speed of the Tollmien–Schlichting waves in the main channel were measured at two characteristic flow velocities, Re = 1580 and Re = 2370. For these Reynolds numbers it was observed that two to three waves span one geometric periodicity length. At Re = 1580 the dominant oscillation frequency was found to be around 26 Hz and at Re = 2370 the frequency distribution formed a band around 125 Hz. Results regarding heat transfer and pressure drop are presented as a function of the Reynolds number, in terms of the block-average Nusselt number and the local Nusselt number as well as the friction factor. Measurements of the local Nusselt number together with visual observations indicate that the lateral mixing caused by flow instabilities is most pronounced along the upstream vertical wall of the heated block in the groove region, and it is accompanied by high heat transfer coefficients. At Reynolds numbers beyond the onset of oscillations the heat transfer in the grooved channel exceeds the performance of the reference geometry, the asymmetrically heated parallel plate channel. Received on 26 April 2000     

An experimental investigation on performance of fins on a horizontal base in free convection heat transfer

 Natural convection heat transfer in rectangular fin-arrays mounted on a vertical base was investigated experimentally. An experimental set-up was constructed and calibrated to test 15 different fin configurations. Fin length and fin thickness were kept fixed at 100 and 3 mm respectively, while fin spacing was varied from 4.5 to 58.75 mm and fin height was varied from 5 to 25 mm. Base-to-ambient temperature difference was also varied through a calibrated wattmeter ranging from 10 to 50 W. The results showed that fin spacing is the most significant parameter in the performance of fin arrays; and for every fin height, for a given base-to-ambient temperature difference, there exists an optimum value for the fin spacing for which the heat transfer rate from the fin array is maximized. It was seen that higher heat transfer enhancement are obtained with vertically oriented bases than with horizontally oriented bases for fin arrays of the same geometry. Received on 16 February 2000     

An experimental study of a water droplet impinging on a liquid surface

An experimental study is presented for water droplet impingement on a liquid surface. The impaction process was recorded using a high-speed digital camera at 1,000 frames/s. The initial droplet diameter was fixed at 3.1 mm ± 0.1 mm, and all experiments were performed in atmospheric air. The impact velocity was varied from 0.36 m/s to 2.2 m/s thus varying the impact Weber number from 5.5 to 206. The impacted liquid surface consisted of two fluids, namely water and methoxy-nonafluorobutane, C₄F₉OCH₃ (HFE7100). The depth of the water and HFE7100 pool was varied from 2 mm to 25 mm. The collision dynamics of water in the HFE7100 pool was observed to be drastically different from that observed for the water droplet impingement on a water pool. The critical impact Weber number for jet breakup was found to be independent of liquid depth. Water–HFE7100 impact resulted in no jet breakup over the range of velocities studied. Therefore, no critical impact Weber number can be defined for water–HFE7100 impact. Received: 27 June 2001/Accepted: 29 November 2001     

Void-fraction measurements in a steady-state mercury-nitrogen flow loop

 A steady-state vertical mercury-nitrogen flow system has been investigated using three tomographic algorithms. Void-fraction profiles have been reconstructed by the data collected by a 60 mCi gamma-ray source and a single Na(Tl) detector. The results indicate excellent agreement (within ±1%) between the least-squares-solution and the chord-segment-inversion algorithms. The data-collection time for each data-ray was varied so that in each case 3000 counts could be collected to reduce the detrimental effect of Poisson statistics in the reconstructed profiles. Eleven sets of data have been collected for nitrogen flow-rates between 0.00123 and 0.00884  kg/s. The mercury flow-rates were between 17.6 and 34.1 kg/s. The resulting cross-sectional average void varied between 0.10 and 0.38. Received: 13 March 1997/Accepted: 12 September 1997     

Experimental study of massive wall systems with fins attached on the heated wall and with glazing

 Natural convection, radiation and conduction heat transfer in passive solar massive wall systems with fins attached to the heated surface and with glazing is experimentally studied. The system was 0.78 m high, 0.40 m wide and 0.10 m thick concrete wall with a glazing placed at 0.0265 m from the surface. It had 0.025 m long, 0.004 m thick horizontal fins made as an integral part of the massive wall and placed at 0.01 m intervals. A heat source was used to impose a constant heat flux which could be varied from about 200–800 W/m². Temperatures at various points and heat flux by convection at the back were measured. Using various assumptions, the systems was also analyzed theoretically. The results show that about 40% of the heat flux imposed on the finned surface goes through the system and is dissipated at the back. Received on 7 September 2000     

Heat transfer characteristics of pulsated turbulent pipe flow

Heat Transfer characteristics of pulsated turbulent pipe flow under different conditions of pulsation frequency, amplitude and Reynolds number were experimentally investigated. The pipe wall was kept at uniform heat flux. Reynolds number was varied from 5000 to 29 000 while frequency of pulsation ranged from 1 to 8 Hz. The results show an enhancement in the local Nusselt number at the entrance region. The rate of enhancement decreased as Re increased. Reduction of heat transfer coefficient was observed at higher frequencies and the effect of pulsation is found to be significant at high Reynolds number. It can be concluded that the effect of pulsation on the mean Nusselt numbers is insignificant at low values of Reynolds number. Received on 29 June 1998     

Optical probe visualization of air–water flow structure through a sudden area contraction

 Single-fiber optical probes were used to investigate the time-averaged structure of gas–liquid horizontal flow through a sharp-edged sudden area contraction. The probes allowed to measure the local void fraction distribution over several cross sections of a pipe having an inner diameter of 0.08 m upstream and 0.06 m downstream of the sudden contraction. The water mass flow rate was 3 kg/s, while the gas fraction of the volume flow ranged from 0.2 to 0.8. The local void fraction was plotted as a function of its two spatial coordinates, so that a representation of the time-averaged gas distribution over the cross section could be obtained. The contraction was shown to considerably alter the distribution of the phases, so that the correlations for straight pipes appear no longer suitable. Received: 27 August 2001 / Accepted: 19 November 2001     

Thermoreversible gels in oil/EVA systems

In this work we investigate the gel formation of EVA/recycled motor oil systems as a route to obtain synthetic binders which could be used instead of natural bitumen, as well as mixed with natural bitumen to modify adequately its viscoelastic response. The EVA copolymer studied in this work has a vinyl acetate content of 18 wt% and solutions of EVA/motor oil have been prepared up to a concentration of 50 wt%. Dynamic viscoelastic frequency sweeps between 10⁻² Hz and 10² Hz have shown that above 3 wt%, at 25 °C, EVA/motor oil systems form gels. It has been possible to define an elastic equilibrium modulus, G_e, for each gel. The dependence of G_e on concentration has been compared with that of PVC/DOP and SBS/oil gels on the basis of the De Gennes model. Thermal stability of EVA/motor oil gels has also been analyzed, indicating that for the highest polymer concentrations gel point is above 80 °C. Received: 23 December 1999 Accepted: 13 June 2000     

Full Field Strain Measurement in Compression and Tensile Split Hopkinson Bar Experiments

The 3D image correlation technique is used for full field measurement of strain (and strain rate) in compression and tensile split Hopkinson bar experiments using commercial image correlation software and two digital high-speed cameras that provide a synchronized stereo view of the specimen. Using an array of 128 × 80 (compression tests) and 258 × 48 (tensile tests) pixels, the cameras record about 110,000 frames per second. A random dot pattern is applied to the surface of the specimens. The image correlation algorithm uses the dot pattern to define a field of overlapping virtual gage boxes, and the 3-D coordinates of the center of each gage box are determined at each frame. The coordinates are then used for calculating the strains throughout the surface of the specimen. The strains determined with the image correlation method are compared with those determined from analyzing the elastic waves in the bars, and with strains measured with strain gages placed on the specimens. The system is used to study the response of OFE C10100 copper. In compression tests, the image correlation shows a nearly uniform deformation which agrees with the average strain that is determined from the waves in the bars and the strains measured with strain gages that are placed directly on the specimen. In tensile tests, the specimen geometry and properties affect the outcome from the experiment. The full field strain measurement provides means for examining the validity and accuracy of the tests. In tests where the deforming section of the specimen is well defined and the deformation is uniform, the strains measured with the image correlation technique agree with the average strain that is determined from the split Hopkinson bar wave records. If significant deformation is taking place outside the gage section, and when necking develops, the strains determined from the waves are not valid, but the image correlation method provides the accurate full field strain history.     

Evaluation of transient turbulent flow fields using digital cinematographic particle image velocimetry

This paper describes an experimental development for temporal and spatial reconstruction of continuously varying flow fields by means of digital cinematographic particle image velocimetry (PIV). The system uses a copper-vapor laser illumination synchronized with a high-speed camera, and continuously samples at 250 fps to measure transient and non-periodic turbulent flows with relatively low frequencies, i.e., the surf zone turbulence produced by depth-limited wave break in a long laboratory flume. The use of the developed PIV system comprehensively records the temporal development of both phase-averaged and instantaneous turbulent vortex flows descended from the breaking waves to the bottom. Also, the measured power spectra show harmonic frequencies, ranging from the orbital frequency of 0.5 Hz up to the order of 5 Hz, and the well-known −5/3 dependence upon the turbulence fluctuation frequencies thereafter. Received: 2 December 1999/Accepted: 6 September 2000     

Critical heat flux in forced convective subcooled boiling with a plane wall jet (effect of subcooling on CHF)

Critical heat flux (CHF) during forced convective subcooled boiling with a plane jet has been yet made insufficient except for saturation condition when comparing CHF with impinging jet system including multiple jets. The present experiment has measured the CHF with plane jet on a rectangular heated surface of 40, 60, and 80 mm in length and 10 and 20 mm in width. Subcooled liquid being supplied through the plane jet with a different thickness of 1 and 2 mm, covers the heated surface where rigorous boiling takes place. The experiment varies a jet velocity from 3 to 15 m/s, a subcooling from 0 to 60 K, and system pressure at 0.1 MPa for water and at 1.5 to 3.0 MPa for R22. It is found that the existing correlation for saturation can be applied to the CHF of water, but hardly to the CHF of R22 in spite of saturation condition. After the effects of jet velocity and subcooling on the CHF can be elucidated, a new correlation including the effect of subcooling is proposed to predict most of the CHF data within an accuracy of ±20 percent. This correlation for saturated condition is found to interestingly agree with that theoretically derived by applying the Katto and Haramura criterion to this system. Received on 8 January 1997     

Quantitative visualization of the near-wall structures in a turbulent pipe flow by image correlation velocimetry

Laser-induced fluorescent dye visualization and image correlation velocimetry were employed to delineate near-wall turbulent structures in a pipe flow. The sweeping and ejection events near the wall and the downstream evolution of a large-scale eddy structure rotating in a counter-clockwise direction were clearly reflected in the instantaneous fluctuating velocity fields. This eddy structure was found to form mostly in the logarithmic region and to dominate the flow structures there, while the ejection and sweeping events in the log layer were greatly influenced by the existence of the large-scale eddy structure. Received: 29 January 2001 / Accepted: 22 October 2001     

An assessment of pressure drop and void fraction correlations with data from two-phase natural circulation loops

 Void fraction and pressure drop correlations play an important role in predicting the performance of natural circulation loops. Hence an assessment of the commonly used and often cited correlations for pressure drop and void fraction has been carried out with data from natural circulation loops. This assessment considered 33 void fraction correlations and 14 pressure drop correlations. The void fraction correlations were initially tested against the various limiting conditions. Only 14 correlations were found to satisfy at least two limiting conditions (i.e., at x = 0; α = 0 and at x = 1; α = 1) and were assessed against the data. This assessment showed that the Chexal et al. (1996) correlation is better than all the others considered. The assessment of pressure drop correlations were carried out with the Chexal et al. (1996) correlation for void fraction and Saha-Zuber model for the onset of subcooled boiling. This assessment showed that most correlations give predictions close to each other. Received on 17 January 2000     

Flow measurement on an oscillating pipe flow near the entrance using the UVP method

 The authors have carried out a study to investigate and clarify the characteristics of purely oscillating pipe flows over the developing region. The main objective of this study is to establish the method of time-dependent velocity profiles obtained by the ultrasonic velocity profile (UVP) measurement method. First, the relationship between the test fluids and the microparticles, as reflectors of ultrasonic pulses, was investigated. In addition, the relationship between the sound speeds of the test fluids and the wall materials was studied. Second, the UVP was used to obtain the instantaneous velocity profiles in oscillating pipe flows, and the developing characteristics of the flows were analyzed. Finally, the “entrance length” (by analogy with a unidirectional pipe flow) required for oscillating pipe flows was analyzed by examining the amplitude of the harmonic spectral components of the oscillating frequency. A fast Fourier transform (FFT) is proposed as the applicable method to estimate the “entrance length”. From the Fourier transform of the velocity on the centerline, nonlinear oscillation of fluid occurs in the “entrance length” of the oscillating flows, and the viscous dissipation of the higher-order velocity harmoncis determines the entrance region. The “entrance length” can be obtained from the dissipation length of the third-order harmonic. These results prove that the UVP method is highly applicable to carry out the flow measurement in the “entrance length” of oscillating pipe flow. Received: 20 March 2000 / Accepted: 10 August 2001     

Evaluation Using Digital Image Correlation of Stress Intensity Factors in an Aerospace Panel

The fracture behavior of a crack propagating in a large (4.8 m × 1.4 m) aircraft panel was investigated quantitatively by experiment for the first time using digital image correlation. Mixed mode (I+II) stress intensity factors were evaluated using a methodology, which combined digital image correlation with the multi-point over-deterministic method to fit displacement field equations to the experimental data from around a crack tip. More than 800 images were taken during a 10-minute time period as the fracture of the panel occurred under monotonic loading. It was observed that the crack propagated through the skin of the panel at a relatively low speed, with an average crack tip velocity of 0.014 mm/s, and changed its propagation direction at particular points due to the reinforcement of the structure. In the later stages of the test, substantial shear lips were observed indicating a state of plane stress as would be expected in a thin, wide panel and the size of the plastic zone increased substantially. The value of the mode I stress intensity factor obtained from the measured displacement fields initially increased linearly to around 50 MPa√m (K_Ic = 37 MPa√m) and afterwards non-linearly reaching 300 to 400 MPa√m for crack extensions of the order of 100 mm. It is proposed that these high values of stress intensity factor do not represent an unrealistically high material fracture toughness but rather are indicative of the high resistance to crack growth of the structural assemblage of ribs, stringers and hole reinforcements in the panel which allow the skin to sustain a strain level that would otherwise cause unstable crack growth. Digital image correlation is demonstrated to be particularly powerful in elucidating this structural behavior.