Pool boiling experiments are performed to investigate the effects of inclination angle, θ, and average roughness, Ra, of uniformly heated 10 × 10 mm copper surfaces on saturation boiling of HFE-7000 dielectric liquid. Ra varied from 0.039 to 0.1.44 μm and θ from 0° (upward facing) to 180° (downward facing). In addition, high speed images and still photographs of nucleate boiling in the various regions and near CHF are captured to assist the interpretation of experimental results. The values of CHF and the maximum nucleate boiling heat transfer coefficient, h MNB , which occurs near the fully developed nucleate boiling region, increase with increasing Ra and/or decreasing θ. The corresponding surface superheats, however, decrease with increasing θ and/or Ra. In the upward facing orientation, CHF increases from ∼20.7 to ∼30.9 W/cm 2 with increasing Ra from 0.039 to 0.144 μm, and the corresponding CHF values in the downward facing orientation are 7.1 and 8.9 W/cm 2 , respectively. Nucleate boiling heat transfer coefficient, h NB , in the discrete bubbles region at low surface superheats increases, while in the fully developed nucleate boiling region at high superheats it increases with increasing Ra and/or decreasing θ. The developed correlations for CHF and h MNB and corresponding superheats, as functions of Ra and θ, are in good agreement with the experimental data.

Recent studies have indicated that ZnO superhydrophilic surfaces on copper and aluminum substrates enhance surface wettability and evaporation processes in droplet evaporation and pool boiling experiments. At slightly superheated surface temperatures, droplet evaporation on a nanoporous superhydrophilic surface exhibits onset of nucleation and nucleate boiling effects found in pool boiling processes. This study explores water droplet evaporation and quenching experiments conducted on nanostructured surfaces of a 45° downward facing pyramid copper, aluminum, and 304 stainless steel substrate. The nanostructured surfaces were used to conduct adiabatic droplet deposition, droplet evaporation, and quenching experiments. Through the three experiments and through surface characterization, by means of scanning electron microscopy, the underlying heat transfer performance and mechanism of droplet evaporation and pool boiling on various host metal substrates can be understood. The six surfaces tested were the following: bare copper, aluminum, and 304 stainless steel surfaces; and ZnO nanostructured surfaces on copper, aluminum, and 304 stainless steel. The experimental variables in this study were temperature and time. Through temperature and time measurements, the mean heat flux at varying superheats could be determined. Experimental results indicate the ZnO nanostructure enhances wettability and water evaporation on a variety of metal substrates. The presence of the ZnO nanostructure increases the mean heat flux and critical heat flux (CHF) in comparison to the bare metal surfaces. These results provide insight into the potential of enhancing spray cooling systems by growing a ZnO nanostructure on a heat exchanger fabricated of various metals.

In this research, the transient heat transfer due to exponentially increasing heat input was experimentally measured for upward water flowing in a vertical small tube. The heat generation rate was increased exponentially with a function of Qoexp ( t /τ), where, Qo is an initial heat generation rate, t represents time and τ is e-folding time. The heat generation rate was controlled by high speed computer system. The test tube was heated with exponentially increasing heat input by direct current. The average temperature of test tube was measured by resistance thermometry using a double bridge circuit. The experimental apparatus consists of a test section, a cooler, a heater, a pump, a tank and a pressurizer. The working fluid was distilled and deionized water. The inlet fluid temperature of test tube was controlled by the cooler and the heater. The system pressure was up to 800 kPa. The test tube was 0.7 mm in inner diameter and 12.0 mm in heated length respectively. The ratio of heated length to inner diameter was 17.1. The test tube was electrically isolated from experimental loop by Bakelite plates. The experimental data were compared with previous correlations of nucleate boiling. It was obtained that the experimented data agree well with full-developed flow boiling correlation by Rohsenow. Moreover, the transient critical heat flux (CHF) and nucleate boiling with onset of nucleate boiling (ONB) values increased with the increase in flow velocity. The transient CHFs and ONBs increased with a decrease in e-folding time at τ < 1 s, and they approached steady-state value at τ > 1 s. It was understood that the heat transfer is in steady-state at τ > 1 s, and it is in transient state at τ < 1 s.

In this study, a combination of synchronized high-speed video (HSV) and infrared (IR) thermography was used to characterize the nucleation, growth and detachment of bubbles generated during nucleate boiling inside the nanoemulsion fluid. The Ethanol/Polyalphaolefin nanoemulsion fluid was formed by dispersing ethanol nanodroplets into base fluid Polyalphaolefin, in which these nanodroplets can serve as the pre-seed boiling nuclei. With this unique combination, it allows controlled nucleation, time-resolved temperature distribution data for the boiling surface and direct visualization of the bubble cycle to track bubble nucleation and growth. Data gathered included measurements of bubble growth versus time, as well as 2D temperature history of the heater surface underneath the bubbles. Our findings demonstrate a significant difference of bubble dynamics between the nanoemulsion fluid and pure ethanol, which may also account for the substantial increase in heat transfer coefficient and critical heat flux of nanoemulsion fluid. It is also observed here that the bubbles occurred inside the nanoemulsion fluid appear to be more uniform and two orders-of-magnitude larger in size. While the growth rate of the bubbles inside pure ethanol was found to be heat diffusion controlled at a coefficient around ½, which however, dropped to be around 0.3 for nanoemulsion fluid. Further study on this unique system will help reveal its heat transfer mechanisms.

Three dimensional numerical simulations for pool boiling of saturated water at atmospheric pressure conditions are performed on a horizontal surface using finite difference method under the framework of parallel computing. Since heat conduction in the solid phase is not considered, in order to simulate realistic heating surface, dependence of bubble nucleation frequency and nucleation site density on wall superheat and contact angle are obtained from the correlations reported in the literature. Steady state boiling curve for all the three regimes viz. nucleate, transition and film boiling has been obtained with a unified numerical model by incrementing the wall superheat for a static contact angle of 38°. Evaporative heat flux from the microlayer is separately accounted for in the present study by sub grid modeling. Both the phases are considered as incompressible while the interface separating the phases is solved using level set method. The governing equations of mass, momentum and energy for both the liquid and the vapor phase are solved coupled with the jump conditions at the interface employing ghost fluid and cut cell method. Diffusion terms are treated implicitly while convection terms are treated using second order ENO scheme. Spatial and temporal averaged wall heat flux and wall void fraction are plotted and compared against correlations and experimental values previously reported. The nucleate boiling heat flux obtained from the present numerical model is under predicted in comparison to the Stephan and Abdelsalam correlation. Comparison of the computed wall void fraction against experimental values is done for the transition boiling region. At critical heat flux formation of long vapor column was seen while intermittent liquid surface contacts were seen in the transition boiling regime. The computed critical heat flux value is lower than that obtained from the correlation of Maracy and Winterton.

Critical heat flux (CHF) of convective boiling in a mini-tube due to power transient was measured. A platinum tube with an inner diameter of 1.0 mm was heated exponentially by a direct current power supply as Joule heating. The heated length of the platinum tube was 40.9 mm. The platinum tube was mounted vertically in the water-loop apparatus which consisted of a circulating pump, a pre-heater, a flow mater, a pressurizer, a cooler and a test section. The deionized water was pressurized by the pressurizer up to approximately 800 kPa to measure CHFs at the high subcooling. The upward flow velocity in the platinum tube was ranged from 5 to 11 m/s. The inlet subcooling was ranged from 92 to 117 K. The heat generation rate was controlled with exponential functions. The e-folding time of the heat generation rate was ranged from 30 ms to 18 s. As an experimental result, it was found that the CHFs increased with increasing the flow velocity and the inlet subcooling. The CHF also increased with decreasing the e-folding time of the heat generation rate. Since the heat generation rate of the platinum tube increased rapidly under the power transient condition, it was considered that the heat flux of the platinum tube increased until the vapor blanket covered the heated surface of the platinum tube.

Quenching experiments were performed with hot stainless steel spheres in a pool of water-based nanofluids in the presence of carbon nanotubes (CNTs) of various sizes. In order to explore the size effect, a test matrix was developed by choosing multi-walled CNTs with lengths from 1 μm to 5 μm and outer diameters from 30 nm to 60 nm. The concentration was fixed at 0.5% by mass for all types of CNTs. The initial temperature was 400 °C and the transient temperature variations at the center of the sphere were recorded as quenching curves. By establishing a lumped capacitance model, the transient surface heat flux variations were obtained as boiling curves. The original and boiled surfaces were both subjected to a series of characterizations to determine the changes in morphology, roughness, and wettability to identify the effects of CNT size on the surface properties of the formed deposition layers as well as to elucidate the mechanisms for regulation of the boiling and quenching behaviors. The results suggested that the critical heat flux (CHF) and the Leidenfrost point (LFP) are enhanced to various degrees due to the discrepancy in the size of the CNTs in nanofluids. It was shown that the CNTs deposited on the surfaces create various morphologies depending on their size. The CNTs with a length of 5 μm and a diameter of 60 nm exhibited the most significant effect on the boiling behaviors. In comparison to CNTs with a shorter length of 1 μm, the 5 μm long CNTs were much easier to form porous layers. The results of the contact angle and roughness tests showed that the porous layers tend to affect the surface roughness instead of surface wettability. The increases of the nucleation site density and surface roughness due to the presence of porous layers were identified as the primary cause for the modified boiling behaviors during quenching.

The present study is motivated by interest in understanding of physical mechanisms that govern the effect of material and micro-structural characteristics of heat surface on boiling heat transfer and burnout at high heat fluxes. The effect was reported and investigated experimentally and analytically over several past decades. Only recently, with the advent of nanotechnology including microscale manufacturing, it becomes possible to perform high heat-flux boiling experiments with control of surface conditions. Of particular importance for practice is the potential for significant enhancement of boiling heat transfer (BHT) and critical heat flux (CHF) in pool and flow boiling on heaters with specially manufactured and controlled micro-structured surfaces. This enhancement is very important to a very wide range of engineering applications, like heat exchanger and cooling system, where maximum flux is needed. Currently, there are many controlled experiments that investigate such effect and they lend themselves a subject for detailed computational analysis. The focus of this study is micro-hydrodynamics of the evaporating thin liquid film at the receding triple contact line, corresponding to formation of dry spot in the footprint of a growing bubble. Parametric investigations are performed to assess the hypotheses that micro-structured surfaces enhance resilience to burnout due to residual liquid in the dry patch after contact line receding. Towards the study objective, a particle-based (mesh-less) method of computational fluid dynamics called Smoothed Particle Hydrodynamics (SPH) is adopted. The SPH method is selected for its capability to handle fluid dynamics in complex geometries and free surface problems without mass loss (characteristic of alternative interface capturing schemes used in mesh-based methods). Both surface tension and surface adhesion (hydrophilicity) are implemented and tested. The solid (heater) surface and manufactured micro-structures are represented by solid-type particles. Heat transfer, phase change (evaporation) and vapor dynamics are not included in the present simulation. The bouncing drop case measures the contact time of water droplet with solid surface. This case is used for “mesh” sensitivity (particle size) study and calibration of boundary conditions and surface tension coefficient. Subsequently, case studies are formulated and performed for contact line dynamics on heater surfaces with the fabricated Micro Pillar Arrays surfaces (MPA) and smooth surface. Variable characteristics include surface tension and pillar density on structured surface (modified by changing distance between pillars). First of all, residual fluid are found in all simulations with structured surface, while fluid are drained for smooth cases. For structured surface, it’s found that after the contact line recedes, fluid with higher surface tension resides in the dry patch more than fluid with lower coefficient, and the relation tends to be non-linear. While for smooth surface, all fluid will be drained after certain time and the relations are non-monotonic; it’s also found that the amount of residual fluid increase as the distance between pillars decreases until a limit. The fluid then starts to decrease with pillars being set further apart. The increase starts from 30 μm and the limit is around 10 μm .

Jet impingement cooling with phase change has shown the potential to meet the increased cooling capacity demands of high-power-density (of order 100 W/cm 2 ) automotive electronics components. In addition to improved heat transfer, phase change cooling has the potential benefit of providing a relatively isothermal cooling surface. In the present study, two-phase jet impingement cooling of multiple electronic devices is investigated, where the fluorinated dielectric fluid HFE-7100 is used as the working fluid. Four different types of jet arrays, namely, a single round jet with orifice diameter of 3.75 mm, and three different 5 × 5 arrays of round jets with orifice diameters of 0.5 mm, 0.6 mm and 0.75 mm, were tested and compared for both heat transfer and pressure drop. The experimental Reynolds number at the orifice ranged from 1860 to 9300. The results show that for the same orifice pressure drop, the single jet reached CHF at approximately 60 W/cm 2 , while the 5 × 5 array ( d = 0.75 mm) safely reached heat fluxes exceeding 65 W/cm 2 without reaching CHF. Additionally, the experimental results show that the multi-device cooler design causes an unintended rise in pressure inside the test section and a subsequent increase in sub-cooling from 10 K to 23.3 K.

Flow oscillation is a crucial issue for the development of flow boiling heat transfer in the applications. Inlet orifice has been proven be an option to eliminate the oscillation. However, the effects of inlet orifice on critical heat flux and flow boiling heat transfer coefficient are lack of study. In this work, the effects of inlet restriction on critical heat flux and heat transfer coefficient in single horizontal microtube under uniform heating condition is experimentally investigated using FC-72 as working fluid. A stainless steel microtube with an inner diameter of 889 μm is selected as main microtube. Two smaller microtubes are assembled at the inlet of main microtube to achieve the restriction configurations of 50% and 20% area ratios. The experimental measurement is carried out at mass fluxes ranging from 160–870 kg/m 2 ·s and heat fluxes varying from 6–170 kW/m 2 . Two saturation pressures, 10 and 45 kPa, are tested. The experimental results of critical heat flux and two phase heat transfer coefficient obtained in the microtube without orifice are compared with the existing correlations. The addition of an orifice does not enhance the normal critical heat flux but increases the premature critical heat flux. In aspect of heat transfer, the orifice shows improvement on heat transfer coefficient at low mass flux and high saturation pressure.

A study was conducted to experimentally characterize the pool boiling performance of hydrofluorocarbon (HFC)-245fa. The motivation for this research is to characterize the performance of candidate refrigerants for potential use in automotive power electronics two-phase cooling systems. The HFC-245fa pool boiling experiments were conducted using horizontally oriented 1-cm 2 heated surfaces to quantify the effects of pressure and a microporous-enhanced coating on heat transfer coefficients and critical heat flux (CHF) values. Experiments were carried out at pressures ranging from 0.15 MPa to 1.1 MPa (reduced pressure range: 0.04–0.31). To enhance boiling heat transfer, a copper microporous coating was applied to the test surfaces. The coating was found to enhance heat transfer coefficients by as much as 430% and CHF by approximately 50%. Increasing pressure decreased the magnitude of the heat transfer coefficient enhancements but had minimal effect on CHF enhancements. The experimental data were then used to generate correlations for the boiling heat transfer coefficients and CHF values. Finally, the performance of HFC-245fa was compared to the performance of hydrofluoroolefin (HFO)-1234yf and HFC-134a at conditions of equivalent saturation temperatures and reduced pressures.

Boiling can provide orders of magnitude higher cooling performance than a traditional air cooled system especially related to electronics cooling application. It can dissipate large quantities of heat while maintaining a low surface temperature difference. Flow boiling with microchannels has shown a lot of potential due to its high surface area to volume ratio and latent heat removal. Flow instabilities and early critical heat flux have however prevented its successful implementation. A novel flow boiling design is experimentally investigated to overcome the above mentioned disadvantages while presenting a very low pressure drop. The design uses open microchannels with a tapered manifold (OMM) to provide stable and efficient operation. Distilled, degassed water at atmospheric pressure is used as the fluid medium. Effect of tapered block with varied dimension is investigated. Heat transfer coefficient and pressure drop data for uniform and tapered manifolds for plain and microchannel chips are presented. A maximum heat flux of 281.2 W/cm 2 at 10.1 °C wall superheat is obtained with microchannel chips using a tapered manifold. The CHF was not reached as the performance exceeded the heater capacity. The maximum pressure drop obtained for the above mentioned configuration was only 3.3 kPa.

The benefits of eliminating instabilities in two-phase microchannel flow with inlet orifices come with costs. This study describes the tradeoffs between microchannels with and without inlet orifices, focusing on results from critical heat flux data obtained for various orifice sizes and mass fluxes. An adjustable inlet orifice controlled with a micrometer was placed in front of an array of 31 parallel microchannels each with a hydraulic diameter of 0.235 mm and a length of 1.33 cm. For mass fluxes ranging from 186 kg m −2 s −1 to 847 kg m −2 s −1 , critical heat flux (CHF) data were obtained for 7 different orifice sizes. For low flow rates that provided a low quality saturated inlet condition, the difference in CHF values was found to be minimal between open and almost closed orifice conditions. The smallest orifice achieved a CHF value of 5 W cm −2 less than the largest orifice size for a mass flux of 186 kg m −2 s −1 , and 7 W cm −2 less for a mass flux of 433 kg m −2 s −1 . For mass fluxes higher than 433 kg m −2 s −1 , subcooled conditions were present at the orifice inlet, and the highest CHF values occurred with an orifice hydraulic diameter of 35 percent of fully open. For the higher mass flux cases, orifice sizes in the range of 1.8 percent to 28 percent of fully open caused CHF to occur at lower values than less restrictive orifice sizes. This was due to loss of cooling capacity from rapid pressure drop through the orifice. Slightly higher average channel pressures also decrease the refrigerant’s latent heat of vaporization. For the orifice sizes from 35 to 70 percent of unrestricted flow, a very minimal increase in pressure drop over fully open inlet conditions occurred and the general trend was higher CHF values. Very small inlet orifices are beneficial for steady state conditions that do not approach CHF; however, overly restricting the flow at the inlet to microchannels reduces cooling capacity significantly and will cause early onset of CHF. A slightly restrictive inlet orifice will increase CHF.

Data from an extensive experimental study of submerged jet impingement boiling performed with water at sub-atmospheric pressures and with FC-72 at atmospheric pressure are used to develop a predictive critical heat flux (CHF) correlation for use in thermal management of electronic components. The configuration was that of a circular submerged jet impinging on a high-thermal-capacity copper surface with a standoff distance of 6 nozzle diameters. Varied parameters included the Reynolds numbers (Re) from 0 (pool boiling) to 14000, surface-to-nozzle diameter ratios (by varying the nozzle diameter) from 25 to 6, system pressures (0.2, 0.3, 0.5, 1 bar), surface roughness (123 nm, 33 nm), and system subcooling. CHF is found to increase with Re, system pressure, subcooling, and roughness and decreases with increase in nozzle diameter for a fixed Re. Comparison with correlations in literature indicated that data of sub-atmospheric jet impingement of water were poorly predicted by existing correlations while the Monde and Katto correlation [1] was found to predict the atmospheric jet impingement data with FC-72 within 10 percent at Re >4000. Data from the experiments were fitted to a submerged forced convective CHF model proposed by Haramura and Katto [2] to develop a correlation for submerged jet impingement boiling over a wide range of density ratios. Using this model, the entire CHF dataset from experiments can be predicted with a maximum error of less than 11 percent and an average error of less than 2.6 percent.

The effect of a perturbation in the diameter of a circular microchannel on critical heat flux (CHF) is explored across a variety of predictive CHF correlations, and an optimum rate of diameter expansion is sought. It is demonstrated that a nonzero diameter expansion necessarily improves performance under several criteria for critical heat flux, and an optimum expansion parameter exists for many of these criteria. CHF relations are seen to follow a few distinct types, and those relations which contemplate effects which may directly influence CHF, such as pressure and phase velocity, tend to better reflect the experimentally demonstrated effect of the perturbation on the diameter of the microchannel.

A closed-loop two-phase microchannels cooling system using a micro-gear pump was built in this paper. The microchannels heat sink was made of oxygen-free copper, and 14 parallel microchannels with the dimension of 0.8mm(W)×1.5mm(D)×20mm(L) were formed by electric spark drilling followed by linear cutting which separated the channels from each other. The heat transfer performance was evaluated by the fluid temperature, the pressure drop across the micro-channels and the volumetric flow rate. Experiments were performed with refrigerant FC-72 which spanned the following conditions: initial pressure of P in = 73 kPa, mass velocity of G = 94 – 333 kg/m 2 s, outlet quality of x e,out = 0 – superheat and heat flux of q ″ = 25–140 W/cm 2 . The result showed that, the maximum heat flux achieved 96 W/cm 2 , as the heating surface temperature was kept below 85 °C and critical heat flux occurred in the condition of low flow rate. Average two-phase heat transfer coefficients increased with the heat flux at low mass flux (G = 94 and 180 kg/m 2 s) and all heat fluxes, high mass flux (G = 333 kg/m 2 s) and all heat fluxes, and moderate mass fluxes (G = 224kg/m 2 s) under low and moderate heat fluxes (q ″ <110 W/cm 2 for G = 224 kg/m 2 s), which was a feature of nucleate boiling mechanism. Pressure drop through microchannels heat sink was found to be below 4kPa.

Numerous methods, such as sintered or machined surface, have been proposed to enhance the boiling performance. On the other hand, nanofluids are also reported to increase the convective heat transfer and become attractive in many engineering applications. In this study, anodizing was tried to produce porous structure on the heater surface to enhance the boiling performance. Two electrolyes, phosphoric acid and oxalic acid, were used. It was found the anodized pores prepared by oxalic acid were smaller. However, the phosphoric acid anodized surface exhibited a CHF increment by 40% with almost no superheat increment. A nanofluid pre-boiling process was also proposed. An aluminum wire was used as the heater to boil nanofluid for a certain time at a certain heat flux. 0.1 wt% TiO 2 nanofluid was used as the working fluid during the pre-boiling process. The wire then went on being the heater for pure water. It is found that aluminum wires exhibit boiling enhancement after preboiling in nanofluid. The most effective process heat flux and duration were found experimentally. The experimental results showed an 11° decrease of the surface superheat. According to SEM photos, two layers were deposited on the aluminum surface. The top layer is more like clusters of deposited nanoparticles, and the layer could be dissolved into water when perform subsequent pure water boiling. The other layer located beneath is more condensed and reliable. It is believed that the second layer could be the main mechanism to exhibit the boiling enhancement.

Spherical glass and copper beads have been used to create bead packed porous structures for an investigation of two-phase heat transfer bubble dynamics under geometric constraints. The results demonstrated a variety of bubble dynamics characteristics under a range of heating conditions. At low heat flux of 18.9 kW/m 2 , a single spherical bubble formed at nucleation sites of a heating surface and departed to the interstitial spaces of porous structure. When heat flux increased to 47 kW/m 2 , a single bubble grew into a Y shape between beads layers and connected with others to generate a horizontal vapor column. As heat flux reached 76.3 kW/m 2 , vertical vapor columns obtained strong momentum to form several major vapor escaping arteries, and glass beads were pushed upward by the vapor in the escaping arteries. According to Zuber’s hydrodynamics theory, choking will take place when the size of vapor columns reaches a certain value that is comparable to the critical hydrodynamic wavelength of the vapor column in plain surface pool boiling. The experimental and simulation results of this investigation illustrated that, under the geometric constrains of bead packed porous structures, similar characteristics had been induced to trigger the earlier occurrence of vapor column chocking inside porous structures. The bubble generation, growth, and detachment during the nucleate pool boiling heat transfer have been filmed, the heating surface temperatures and heat flux were recorded, and theoretical models have been employed to study bubble dynamic characteristics. Computer simulation results were combined with experimental observations to clarify the details of the vapor bubble growth process and the liquid water replenishing the inside of the porous structures. This investigation has clearly shown, with both experimental and computer simulation evidence, that the millimeter scale bead packed porous structures could greatly influence pool boiling heat transfer by forcing a single bubble to depart at a smaller size as compared to that in a plain surface situation at low heat flux situations, and could trigger the earlier occurrence of critical heat flux (CHF) by trapping the vapor into interstitial space and forming a vapor column net. The results also proved data for further development of theoretical models of pool boiling heat transfer in bead packed porous structures.

This work investigates the subcooled flow boiling of aqueous based nanofluids in a 510 μm single microchannel with a focus on the effect of nanoparticles on the critical heat flux (CHF). The surface temperature distribution along the pipe, the inlet and outlet pressures and temperatures are measured simultaneously for different concentrations of alumina nanofluids and dionized water. The experiment shows a remarkable increase ∼ 31% in the CHF under very low nanoparticle concentrations (∼0.1v%) and a nonlinear influence of nanoparticles on the subcooled boiling heat transfer.

In nucleate boiling as the heat flux from the wall to the fluid is increased the heat transfer coefficient initially increases. At a sufficiently high heat flux called the critical heat flux (CHF) the heat transfer mechanism suddenly becomes less effective resulting in a rapid jump in wall temperature. In bubbly subcooled (or near-subcooled) conditions the CHF mechanism is referred to as departure from nucleate boiling. Departure from nucleate boiling (DNB) refers to the transition from nucleate boiling where liquid contacts the wall to film boiling in which a vapor layer contacts the wall. Various hypotheses have been used in modeling and predicting CHF. High speed video images of boiling water flows taken at Bettis Laboratory at the critical heat flux visually captured sufficient evidence of the DNB mechanism that improved insight into DNB modeling may be possible. This paper summarizes high speed video image analysis and the development of a new DNB critical heat flux model based on the image analysis findings. Using short window averages of image data, a significant increase in transmitted light intensity is seen near the wall just prior to CHF. The increase suggests that at CHF there is a transient reduction in the interfacial area concentration, a i , or bubble number density near the wall. This is believed to be the result of a sudden increase in bubble coalescence rates near the wall. The increase in coalescence rates results in a reduction in the interfacial area concentration causing it to reach a maximum at CHF. This near-wall maximum in a i at CHF under flow boiling conditions is consistent with recent pool boiling data in the literature. The image based observations motivated development of an interfacial area based CHF model to predict the maximum in the interfacial area concentration at CHF. The model predicts that a critical nucleation site density or a near-wall critical void fraction can be used as a DNB CHF criterion. This is a valuable simplification that can be directly implemented in three-dimensional thermal hydraulic codes. The critical nucleation site density result was used as an input to a simple wall heat transfer partition model to predict the critical heat flux. The model relies on correlation based estimates for the superheat temperature, bubble departure diameter, and bubble departure frequency. Model predictions are compared to CHF values taken from Groeneveld’s 2006 CHF look-up table.