Foil Technique Research Articles

Local heat transfer distribution in channel flow through a periodic octet foam with separation of fin and wall heat transfer

This study investigates local heat transfer distribution and pressure drop in octet-structured aluminium foam (AlSi10Mg) within a rectangular channel. The local heat transfer distribution is analysed using a thin metal foil technique and an IR camera. The foam has a thickness of 12.5 mm and 70.6 % porosity. Experiments were performed to study the effect of Reynolds numbers on heat transfer performance and assess the contribution of conductive versus convective heat transfer. A range of Reynolds numbers were tested from 1000 to 25000. The heat transfer coefficient asymptotes to a constant value beyond the Reynolds number of 20000. To segregate the effect of fin and wall heat transfer, separate experiments are conducted using resin foam. The effect of conduction heat transfer (fin) dominates convection heat transfer (wall), with a contribution ratio of 81–19 %. Metal foam exhibits a 3.5–5 times higher heat transfer coefficient than resin foam. The performance enhancement over smooth channels is 5 times higher for resin foam and 18 to 29 times higher for metal foam. The Performance Evaluation Criterion for metal foam ranges from 2.02 to 3.11, while for resin foam, Performance Evaluation Criterion is between 0.56 and 0.63. The Constant Pumping Power Criterion shows a 32 % higher thermal performance than the Performance Evaluation Criterion.

The effect of porous non-metallic balls on detonation propagation in hydrogen–oxygen mixtures

In this study, the influence of porous non-metallic balls on the dynamics of detonation propagation in hydrogen–oxygen mixtures is studied in stainless steel tubes with square cross-sections. The effect of the length and thickness of the balls is considered in detail. Four pressure transducers are used to record the detonation time-of-arrival, and the average velocity of detonation propagation is calculate. The smoked foil technique is performed to register the detonation cellular structures. The results indicate that increasing the filling length and thickness of the balls leads to a significant increase in velocity loss, critical pressure, and re-initiation distance during detonation propagation. Two propagation modes are observed near the critical pressure. In sub-critical mode, the detonation wave is attenuated and failed eventually, propagating to the end of the tube in the form of a low-speed deflagration wave. In super-critical mode, the detonation propagation can be promoted by inducing a vertical transverse wave generated by the detonation wave passing through a group of small balls. It is worth noting that the increase in filling thickness of the small balls leads to a greater velocity loss compared to the increase in filling length. However, an increase in the filling length will cause the detonation wave to undergo a longer and more sustained weakening process as it passes through the small balls, thereby resulting in a longer re-initiation distance and a lower average velocity over a short distance after re-initiation. By increasing the filling length of the small balls to 312 mm, the transition of the detonation wave from multi-head detonation to double-head detonation can be observed, and then successfully turns into stable detonation. However, when filling double-layer small balls, even in super-critical conditions, the attenuated detonation wave cannot achieve re-initiation over a short distance.

Measurement of local Nusselt number and local recovery factor for impinging multiple compressible jets

In the present study the appropriate reference temperature is identified for the compressible impinging jet. Using the measured reference temperature, local recovery factor is calculated. The steady state thin metal foil technique is used for the measurement of target plate temperature. In this study, the effect of Mach number (Ma) at a constant Reynolds number (Re) and the combined effect of Mach number and Reynolds number on the heat transfer rate are investigated. For both the cases, jet-to-plate distance is varied from z/d = 5 to 12. For the first case (effect of Mach number at a constant Re = 20,000), Ma is varied from 0.15 to 0.85. In the second case (combined effect of Mach number and Reynolds number), Ma is varied from 0.2 to 0.78 and the corresponding Re variation is 7200 to 29,000. At a constant Reynolds number, the heat transfer coefficient increases with the increase in the Mach number. For a given Mach number and Reynolds number, the heat transfer rate decreases with the increase in the jet-to-plate distance. The recovery factor remains unaffected by the Mach number and jet-to-plate distance in the case of the concurrent variation of the Mach number and Reynolds number.

The failure and re-initiation of quasi-detonations in stoichiometric acetylene-oxygen mixtures diluted by nitrogen

In this study, the failure and re-initiation of quasi-detonations are investigated systematically in stoichiometric acetylene-oxygen mixtures diluted by nitrogen, and the nitrogen dilution range of 10–45 % is considered. Photodiodes are introduced to obtain the average velocity by recording the detonation time-of-arrival. The smoked foil technique is performed to register the detonation cellular structures, and a high-speed schlieren system is used to observe the propagation processes. Three propagation modes are defined based on nine repeated experiments, i.e., Go, No-go and Critical regimes. For the “go” regime, the exiting detonation all successfully evolves into a spherical detonation in nine repetitions of the experiments. For the “no-go” case, the detonation all fails upon exit and becomes a deflagration. The case between the Go and No-go is called as the critical mode. Near the limits, the critical pressure of successful evolution into a spherical detonation is obtained, and the data indicates that the critical pressure is approximately independent of the wall roughness. But the critical pressure can increase obviously by introducing the nitrogen dilution. The cell sizes for quasi-detonations are measured, and the ratio of the critical tube diameter (dc) to the cell size (λ) is considered to quantify the critical criterion of detonation re-initiation. Note that the previous criterion of dc/λ ≥ 13 about normal detonation re-initiation cannot be applied to the quasi-detonation, and the re-initiation condition for quasi-detonations can be nearly quantified as dc/λ ≥ 8. The results are consistent with those obtained in our previous study (Xuxu Sun et al., Combustion and Flame, 112280). Corresponding to the normal detonation in a smooth tube, the critical value of dc/λ is always lower for quasi-detonations, which indicates a favorable effect on the detonation transmission is verified.

Production of 203Pb from enriched 205Tl using deuteron beams

Lead-203 is a SPECT emitter that can be used in theranostic applications as an imaging counterpart of lead-212 which is intended to be used for alpha therapy as lead-212/bismuth-212 in-vivo generator. In our study, we explore the production of lead-203 using enriched thallium-205 target irradiated by a deuteron beam. Excitation functions of deuteron induced reactions leading to the formation of 204m,203m2+m1+g,202m,201m+gPb, 202Tl and 203m+gHg isotopes were determined experimentally in the energy range from 21 MeV to 34 MeV. Cross sections were measured using the stacked foils technique and a set of two monitor foils, natNi and natTi for beam intensity evaluation. The experimental excitation functions of the investigated reactions were compared with the published data and also with the TENDL-2021 nuclear database. From our experimental data, we calculated lead-203 thick target yield in the energy range between 30 MeV and 32.5 MeV to be 56.7 MBq/μAh ±6.1 MBq/μAh. This value is compatible with large batch production showing that deuteron beams can be used for a routine production process. However, special attention must be paid to 203Hg and other lead contaminants.

Ultrathin Feldspathic Ceramic Veneers: A Pilot SEM Evaluation of Etched Intaglio Surfaces.

Ultrathin ceramic veneers are a viable therapeutic option to manage esthetic challenges in the anterior zone. Proper conditioning of the intaglio surface of porcelain veneers is essential to achieve an adequate bonding. In clinical practice, this is typically done with chemical etching using an acid-containing agent, such as hydrofluoric acid. While it is well established that the etching effect is dependent on etching time and the acid concentration, little is known about the impact of etching time and the veneer fabrication method. The purpose of this pilot study was to evaluate, using scanning electron microscopy (SEM), the effect that different etching-time protocols have on the intaglio surface characteristics of ultrathin ceramic veneers fabricated with either the platinum foil technique or the refractory die technique. Several replicas of an ultrathin feldspathic ceramic veneer for a maxillary central incisor were fabricated. Individual specimens were processed according to different intaglio surface-etching protocols: no etching, etching for 90 seconds, etching for 120 seconds, and etching for 150 seconds (9.6% hydrofluoric acid used for all etching groups). It was observed that the 120-second etching protocol resulted in a favorable microroughness surface pattern in the platinum foil group. This pattern was comparable to that obtained by etching for 90 seconds with hydrofluoric acid the intaglio surface of veneers fabricated with the refractory die technique. Increasing the etching time to 150 seconds did not result in a more favorable roughness pattern.

Co-production of 155Tb and 152Tb irradiating 155Gd / 151Eu tandem target with a medium energy α-particle beam

IntroductionFour terbium isotopes 149,152,155,161Tb emitting various types of radiation can be used for both diagnostics and therapy. 152Tb emits positrons and is ideal for PET. 155Tb is considered a promising Auger emitter and a diagnostic pair for other terbium therapeutic isotopes. Several methods for the production of 155Tb using charged particle accelerators have been proposed, but they all have significant limitations. The restricted availability of this isotope hinders its medical applications. We have proposed a new method for production of 155Tb, irradiating enriched 155Gd by alpha particles. The possibility of simultaneous production of two isotopes of terbium, 152,155Tb, was also studied for more efficient cyclotron beam use. MethodsIrradiation of 155Gd enriched targets and 155Gd / 151Eu tandem target with alpha-particles with an energy of 54 MeV was carried out at the U-150 cyclotron at the NRC "Kurchatov Institute". The cross sections of nuclear reactions on enr-155Gd were measured by the stack foil technique, detecting the gamma-radiation of the activation products. The separation of rare earth elements was performed by extraction chromatography with the LN Resin. 155Tb was produced via 155Dy decay. ResultsThe cross sections for the 155,156Tb and 155,157Dy production were measured by the irradiation of a gadolinium target enriched with the 155Gd isotope with alpha-particles in an energy range of 54 → 33 MeV. The yield of 155Dy on a thick target at 54 MeV was 130 MBq/μAh, which makes it possible to obtain 1 GBq of 155Tb in 11 hour-irradiation with 20 μA beam current. The possibility of simultaneous production of 152,155Tb by irradiation of 155Gd and 151Eu tandem target with medium-energy alpha-particles is implemented. Optimal irradiation energy ranges of alpha -particles as 54 → 42 MeV for 155Tb and 42 → 34 MeV for 152Tb were suggested. Product activity and radionuclidic purity were calculated.

Influence of thickness of metal foam on the conduction and convection heat transfer for a flat plate with metal foam impinged by a single circular air jet

ABSTRACT The effect of the metal foam thickness on the conduction and convection heat transfer for a metal foam flat plate impinged by a circular air jet is investigated. The IR thermography and thin-metal foil technique are used for the measurement of local heat transfer. An open-cell aluminum metal foam is used for the metal foam flat plate. A 3D-printed resin foam and detached metal foam flat plate are used for the appreciation of the conduction and convection heat transfer. The varying parameters are the thickness of the foam, Reynolds number, and the nozzle exit to plate distance. The presence of the metal foam offers a conduction effect. This predominates over the attenuation in the convective heat transfer by foam due to additional hydraulic resistance. The additional hydraulic resistance offered by the porous foam increases with the increase in the foam thickness. The heat transfer of a porous foamed flat plate decreases with the increase in the foam thickness. The local Nusselt number of the resin foam and detached foam flat plate is almost the same. The conduction effect and attenuation in the convection heat transfer of a metal foam flat plate are quantified by attenuation and enhancement factors. The overall augmentation offered by 4, 8, and 12 mm thick metal foam flat plates is 1.71, 1.42, and 1.43 times compared to the smooth flat plate case, respectively. Hence, it is advisable to use a metal foam flat plate with 4-mm-thick metal foam under circular air jet impingement.

Full Electromagnetic Integration of Impedance-Balanced EMI Filters for Single-Phase Power Converters

Utilizing passive electromagnetic interference (EMI) filters is the most widely used approach to restrain the conducted emissions from power converters. However, the introduction of these filters will significantly raise the weight and size of the systems. Using the emerging flexible multi-layer foil (FMLF) technique, this paper investigates a complete electromagnetically integrated design concept of single-phase EMI filters based on EE-type cores, which can be applicable to engineering application scenarios with different EMI noise attenuation requirements. Two types of integrated structures are constructed with different well-designed FMLF-windings and terminal configurations. One is a common-mode (CM) choke with controllable leakage inductances, which is able to realize the integration design of CM inductor, CM capacitors and differential-mode (DM) inductors. The other is an EMI choke with integrating the CM and DM inductances & capacitances, which achieves electromagnetic integration of all the CM and DM filtering components. Compared to the existing UU-type core based integration methods, the proposed methods can ameliorate the noise suppression performance while reducing filter's weight and volume. Theoretical analyses for modeling and design of the EMI chokes is presented. Then prototypes of different FMLFs based EMI filters have been built for a 500W 100kHz SiC-MOSFET inverter, experiments demonstrate the feasibility and validity of the proposed structures.

Effect of jet plate thickness on the local heat transfer coefficient with multiple air jet impingement

For designing a multiple jet impingement cooling system, the thickness of jet plate (a passage through which jet comes out) is one of the important parameters. Considering the thermal point of view, the thickness of the plate should be chosen so that it would enhance the heat transfer rate. In this study, the effect of jet plate thickness on the local heat transfer coefficient on a flat plate with multiple impinging air jets is investigated. The thickness ratios (t/d) of 0.5, 1.0 and 2.67 are studied. Reynolds number is varied from 5000 to 20,000 and jet-to-plate distance (z/d) is varied from 1 to 5. Jet diameter (d) is chosen to be 6 mm. The jet-to-jet distance (p/d) is kept constant at 3.Steady-state thin metal foil technique, along with infrared thermography, is used to obtain the temperature distribution and thereby the heat transfer coefficient. Coefficient of discharge for each jet plate, central line velocity decay and pressure distribution on the target surface are measured. These measurements provided new insights to better understand the heat transfer results in this study. For lower jet-to-plate distances (t/d = 0.5 to 1), heat transfer decreases with the increase in the jet plate thickness ratio. However, heat transfer is not affected with the further increase in the jet plate thickness ratio i.e., t/d = 1 to 2.67. Therefore, smaller jet plate thickness ratio is suitable to obtained high heat transfer rate. This study presents a new correlation for the local Nusselt number distribution or different jet plate thickness.

Modeling and Design of Full Electromagnetic Integration of a Symmetrical EMI Filtering Circuit With Flexible Multi-Layer Foil Technique

In a high-frequency power electronic converter with wide-bandgap device, electromagnetic interference (EMI) filter is an essential module for mitigating the conducted EMI issues. However, traditional discrete EMI filter will significantly increase total size and weight of the system. In this article, an UU-core based fully integrated symmetrical EMI filter is proposed using flexible multilayer foil technique, all the common-mode (CM) and differential-mode (DM) filtering components are integrated in the same unit for a more compact design. Moreover, a function-decoupling state is realized between the CM and DM elements, which contributes to predigesting the lumped equivalent CM and DM circuits. Then the parameter design procedure can be simplified. Prototypes of the proposed and other typical types of EMI filters are built for a 100 kHz SiC- <sc xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">mosfet</small> voltage source inverter, contrastive experiments have verified the feasibility and validity of the full electromagnetic integration approach.

Detonation in ammonia-oxygen and ammonia-nitrous oxide mixtures

The sensitivity to detonation of ammonia-oxygen (NH3-O2) and ammonia-nitrous oxide (NH3-N2O) mixtures has been investigated experimentally and numerically. Detonation were studied in a stainless steel tube with a length of 4.6 m and an inner diameter of 78 mm. The initiation of the detonation wave was achieved using a weak electric spark and a Shchelkin spiral to trigger flame acceleration and transition to detonation. The soot foil technique was employed to determine the detonation sensitivity. For the numerical simulations, the Shock and Detonation Toolbox in Cantera was employed. The pressure in experiment was below 100 kPa and was extended to 4.5 MPa in the numerical study using a real gas model based on the Peng–Robinson equation of state. Overall, detonation in ammonia-based mixtures have an irregular structure and do not demonstrate a high sensitivity. At ambient temperature, the experimental cell width ranges between 14 and 54 mm for NH3-O2 mixtures and between 7 and 23 mm for NH3-N2O mixtures in the equivalence ratio and pressure ranges ϕ=0.6–1.5, and P1=43–100 kPa, and ϕ=0.3–1.25, and P1=41–80 kPa, respectively. These cell widths are larger than for CH4-O2 mixtures under similar conditions. The mixture with N2O is more sensitive to detonation than the mixture with O2 at low pressure, but becomes less sensitive at elevated pressure. Increasing pressure also tends to stabilize the detonation by raising the isentropic coefficient. Through detailed thermo-chemical analysis, it was shown that detonation in ammonia-based mixtures show pathological detonation behavior at low pressure. The major exothermic reaction is NH3 + OH = NH2 + H2O in NH3-O2 but it is outweighed by H + N2O = N2 + OH in NH3-N2O mixture. One of the dominant radicals is OH, which is supplied by H + O2 = O + OH at low pressure and by H2O2 (+M) = 2OH (+M) at elevated pressure in NH3-O2 mixture; and by H + N2O = N2 + OH in NH3-N2O mixture. The characteristic length scale, i.e., the induction distance, is sensitive to reactions responsible for supplying OH radical in both mixtures. In NH3-N2O mixture, the induction distance is also sensitive to reactions involving the oxidizer, N2O.

Mach reflection of detonation wave on porous wall

This study reports the Mach reflection of gaseous detonation on porous wedges experimentally, in which the porous wall is consisted of equidistant inline square columns. The smoked foil technique was utilized to monitor the evolution of the triple-point trajectory and detonation cells in the Mach stem region. In addition to the wedge angle and initial pressure of gaseous mixture, this paper also focuses on the effect of porosity and pore size on the Mach reflection and its mechanism. The results show that the strength of the Mach stem is significantly weakened by the porous media compared with that on the smooth wedge, which is ascribed to the diffraction and reflection waves generated by the interaction of the Mach stem front with the pore. Furthermore, the onset of the triple-point trajectory is delayed, the angle of which is decreased. The porosity and pore size present distinct influence on the Mach stem height. With the increase in the porosity and the decrease in the pore size, the Mach stem region is attenuated more dramatically and the Mach stem is harder to be formed or even could not be observed. Furthermore, the triple-point trajectory on the porous wall exhibits local self-similarity and satisfies the frozen limit in the near field and the equilibrium limit in the far field. However, the lengths of the existence of the frozen limit and the transition to the equilibrium limit on the porous wall are found to be much shorter than the hydrodynamic thickness, and the recovery of the self-similarity depends largely on the porosity and pore size.

141Pr(α,x): New Cross-Section Data With Special Reference to 140Nd Production for Medicine

Cross sections for the 141Pr(α,x) 144,143,141Pm, 138mPr, 141cum,140cumNd, and 139cumCe reactions were measured by the stacked foil technique at an alpha-particle range of 60 to 20 MeV. Most of the data were obtained for the first time. Experimental data were compared with earlier research and theoretical values from the TENDL-2021 database. Thick target yields of activation products were calculated. The reaction 141Pr(α,5n)140Pm (9.2 s)→140Nd was compared with other 140Nd production routes.

Terminal Configuration Principles for Achieving Various Electromagnetic Integration Units with Flexible Multilayer Foil Technique

Terminal Configuration Principles for Achieving Various Electromagnetic Integration Units with Flexible Multilayer Foil Technique

Excitation functions of alpha particle induced nuclear reactions on natBa up to 60 MeV

The excitation functions of nuclear reactions natBa(α,xn)132, 133m, 134, 135, 137m,g, 139, 141Ce, natBa(α,x)135, 140La and natBa(α,x)131, 133m, 135mBa were measured for the first time in the energy range of 60 → 20 MeV using stack foil technique combined with γ-ray spectrometry. A comparative analysis of the experimental results with TENDL-2021 data was carried out. The thick target yields of radioisotopes Ba, La, and Ce were calculated by integrating the values of the cross sections. The feasibility of practical use of the studied reactions for the production of medical radioisotopes 134Ce and 135La is discussed.

Spectrometry of neutron field based on p(20)+Be source reaction

The research deals with a spectrometry of neutron field based on the p(20)+Be source reaction. The p(20)+Be interaction stands for a nuclear reaction where protons accelerated to energy of 20 MeV bombard Be target. Produced neutron field has broad energy distribution up to 18 MeV. The research was carried out at the Nuclear Physics Institute of the Czech Academy of Sciences and was motivated by extension of experimental possibilities of the NG-2 target station. To determine the neutron spectrum, the activation foil technique was used. During the experiment, ten activation foils were irradiated. Activated foils were measured using the HPGe detector to determine the reaction rates of observed reactions. The neutron field was simulated in the MCNPX code, and the simulation served as a priori information for the unfolding of the neutron spectrum in the SAND-II code. A determined neutron spectrum is essential for various experiments, e.g., material research, measurements of nuclear data, etc.

Effect of metal foam thickness on the conduction and convection heat transfer for a flat plate with metal foam impinged by multiple jets

ABSTRACT The influence of metal foam thickness on the convection and conduction heat transfer is studied by measuring the local heat transfer of a smooth flat plate, attached and detached metal foamed flat plate under multiple jet impingement conditions. 27 jets of 3 mm diameter are introduced from a jet plate. These jets impinge on a thin stainless steel foil that serves as a targeted plate. The pitch of the jets is 4d in both the spanwise and streamwise directions. The spent fluid is exiting in all directions. The influence of metal foam thickness is investigated for an open-cell aluminum foam with 4, 8, and 12 mm thickness. The porosity of aluminum metal foam is 0.92. The pore density (pores per inch) is 20 PPI. The local heat transfer is measured using infrared thermography and a thin metal foil technique. A smooth flat plate impinged by multiple jets serves as a reference case. The evaluation of the convection and conduction heat transfer offered by the metal foamed flat plate is obtained with the help of the detached foam configuration. The jet diameter-based Reynolds number ranges from 2500 to 15000. The range of the jet to plate spacing covered is 3d to 6d. Depending on the metal foam thickness, the metal foamed flat plate offers additional hydraulic resistance to the jet flow and conduction effect. The additional hydraulic resistance attenuates the convective heat transfer and is represented by attenuation factor (α). For metal foam having 4, 8, and 12 mm thickness, the attenuation factor is around 1.11, 0.99, and 0.64, respectively. The conduction effect offered by metal foam is responsible for the enhancement of the heat transfer. The conduction effect is represented as an enhancement factor (E). For metal foam having 4, 8, and 12 mm thickness, the enhancement factor is around 1.51, 1.58, and 2.32, respectively. The conduction effect presented in the metal foamed flat plate dominates over the attenuation in the convective heat transfer. The overall enhancement offered by 4, 8, and 12 mm metal foamed flat plate is around 1.67, 1.56, and 1.48, respectively.

Heat Transfer of a Metal Foamed Flat Plate Impinged by Multiple Jets

The present study focuses on the measurement of the local heat transfer distribution of a smooth flat plate, a flat plate with attached metal foam, and detached metal foam under an inline array of multiple jets using thermal imaging and the thin metal foil technique. After impingement, jet flow exits from all directions. The range of the Reynolds numbers covered in the study is 2500 to 15,000. The impinging distance of to is studied. A 4-mm-thick aluminum metal foam having 92% porosity and 20-pore-per-inch pore density is used. The contribution of the convection and conduction heat for a flat plate with metal foam is studied using a flat plate with a detached foam configuration. The effect of the impinging distance and Reynolds number on the local, spanwise-averaged, and overall averaged Nusselt numbers is studied. The presence of the 4 mm metal foam enhances the local heat transfer in comparison with a smooth flat plate due to conduction heat transfer without the addition of extra hydraulic resistance to the jet flow. The enhancement of the heat transfer due to the presence of the metal foam is quantified by the enhancement factor , and it is found to be around 1.6.

Influence of ratio of nozzle length to diameter on local heat transfer study of an unconfined circular air jet impingement

The influence of ratio of nozzle length to diameter on local heat transfer characteristics of an unconfined circular air jet impingement on flat plate is studied experimentally. The ratio of nozzle length to diameter (L/D = 0.5 to 83), nozzle to plate surface (z/D = 0.2 to 2) and Reynolds number (Re = 12,000 to 28,000) is investigated. The local heat transfer characteristics are reported at stagnation and wall jet region by using IR thermal imager and thin foil technique. Peak Nu occurs at wall jet area for z/D = 0.2 rather than at stagnation point for all L/D ratios and is substantially dependent on z/D and Re. It is caused by flow acceleration due to venacontraction effect between the nozzle and target surfaces which results in higher Nu in wall jet area than in stagnation area. For all Reynolds number, long pipe nozzle gives higher stagnation Nu value at z/D < 0.75 due to fully developed flow at nozzle outlet and L/D = 0.5 gives higher stagnation Nu value at z/D > 0.75 due to higher turbulence motion at nozzle outlet. At z/D < 0.4, L/D = 0.5 has a lower Nu at all L/D ratios in wall jet area (R/D > 2) and it decreases as Re increases. The influence of L/D ratio in wall jet zone (R/D > 1.8) is nearly comparable at all Re for z/D > 0.75. The L/D ratios are least influenced at lower Re and higher z/D value. The correlation of stagnation Nu was obtained with maximum deviation of ±16% to the maximum heat transfer rate for optimum conditions.