Seppo Syrjälä

Laminar flow of viscoelastic fluids in rectangular ducts with heat transfer: A finite element analysis

Abstract A numerical study on the laminar flow and heat transfer behavior of viscoelastic fluids in rectangular ducts is conducted using the finite element approach. A Criminale-Ericksen-Fibley relation is applied to describe the viscoelastic character of the fluid, and a hydrodynamically and thermally fully developed flow with the H1 thermal boundary condition is considered. The finite element procedure employed yields essentially mesh-independent predictions with a fairly moderate computational effort. Computed results are presented and discussed in terms of the secondary flow field, the temperature field, the friction factor and the Nusselt number. In particular it is shown that the presence of a secondary flow markedly alters the temperature field and results in a substantial heat transfer enhancement with all duct aspect ratios considered.

Finite-element analysis of fully developed laminar flow of power-law non-Newtonian fluid in a rectangular duct

The finite-element method has been applied to solve numerically equations governing fully developed laminar flow of power-law non-Newtonian fluid in a rectangular duct. The resulting finite-element solutions for the velocity profile have been further utilized to obtain the numerical values of the friction factor-Reynolds number product for duct aspect ratios ranging from 0.05 to 1.0 and for power-law indices ranging from 0.2 to 2.0, i.e., both shear-thinning and shear-thickening cases are considered.

Higher order penalty-Galerkin finite element approach to laminar natural convection in a square cavity

The performance of a penalty-Galerkin finite element method with quartic triangular elements when applied to laminar natural convection in a square cavity with differentially heated vertical walk and adiabatic horizontal walls is studied. The numerical calculations were carried out for Rayleigh numbers of 104 105, 106, and 107 on several successively refined meshes. The computational results are presented in terms of velocity components, temperature, stream function, and Nusselt number. The numerical procedure adopted in the present study yields consistent performance with respect to mesh refinement, and comparisons with previously published results indicate fairly close agreements.

Further finite element analyses of fully developed laminar flow of power-law non-Newtonian fluid in rectangular ducts: Heat transfer predictions

Forced convection heat transfer to hydrodynamically and thermally fully developed laminar flow of power-law non-Newtonian fluid in rectangular ducts has been studied for the H1 and T thermal boundary conditions. The solutions for the velocity and temperature fields were obtained numerically using the finite element method with quartic triangular elements. From these solutions, very accurate Nusselt number values were determined. Computations were performed over a range of power-law indices and duct aspect ratios.

NUMERICAL SIMULATION OF NONISOTHERMAL FLOW OF POLYMER MELT IN A SINGLE-SCREW EXTRUDER: A VALIDATION STUDY

A recently developed approach to simulate the nonisothermal polymer melt flow in a single-screw extruder is validated against experimental data taken from the literature. Comparisons are made with eight sets of experimental results, and for each case several numerical runs are performed to provide some insight into the sensitivity of simulation outcomes to input material parameters (i.e., viscosity, density, thermal conductivity, and specific heat). In general, predictions for the pressure distribution along the extruder and for the mean melt temperature at the exit of the extruder compare favorably with pertinent experimental results. On the other hand, calculated estimates for the power consumption of an extruder are systematically lower than corresponding measured values.A recently developed approach to simulate the nonisothermal polymer melt flow in a single-screw extruder is validated against experimental data taken from the literature. Comparisons are made with eight sets of experimental results, and for each case several numerical runs are performed to provide some insight into the sensitivity of simulation outcomes to input material parameters (i.e., viscosity, density, thermal conductivity, and specific heat). In general, predictions for the pressure distribution along the extruder and for the mean melt temperature at the exit of the extruder compare favorably with pertinent experimental results. On the other hand, calculated estimates for the power consumption of an extruder are systematically lower than corresponding measured values.

On the analysis of fluid flow and heat transfer in the melt conveying section of a single-screw extruder

An approach is presented to efficiently compute the three-dimensional fluid flow and heat transfer in the melt conveying section of a single-screw extruder. The analysis is based on a moving barrel formulation with an unwound screw channel of rectangular cross section. The viscosity of the polymer melt is described by the power law model together with the exponential temperature dependence. The solution technique consists of marching in the down-channel direction using a fully implicit scheme and solving the resulting equations in each cross-sectional plane by the Galerkin finite element method. Unlike most previous analyses, the present approach takes full account of the recirculatory nature of the cross-channel flow, and therefore also, the cross-stream convection effects are properly accounted for. The present numerical scheme is capable of producing converged wellbehaved solutions without the traditional need for upwinding even at very high values of the Peclet number. The numerical results obtained e...

Effect of the plate surface characteristics and gap height on yield stresses of a magnetorheological fluid

Effects of the plate material, surface roughness and measuring gap height on static and dynamic yield stresses of a magnetorheological (MR) fluid were investigated with a commercial plate?plate magnetorheometer. Magnetic and non-magnetic plates with smooth (Ra???0.3??m) and rough (Ra???10??m) surface finishes were used. It was shown by Hall probe measurements and finite element simulations that the use of magnetic plates or higher gap heights increases the level of magnetic flux density and changes the shape of the radial flux density profile. The yield stress increase caused by these factors was determined and subtracted from the measured values in order to examine only the effect of the wall characteristics or the gap height. Roughening of the surfaces offered a significant increase in the yield stresses for non-magnetic plates. With magnetic plates the yield stresses were higher to start with, but roughening did not increase them further. A significant part of the difference in measured stresses between rough non-magnetic and magnetic plates was caused by changes in magnetic flux density rather than by better contact of the particles to the plate surfaces. In a similar manner, an increase in gap height from 0.25 to 1.00?mm can lead to over 20% increase in measured stresses due to changes in the flux density profile. When these changes were compensated the dynamic yield stresses generally remained independent of the gap height, even in the cases where it was obvious that the wall slip was present. This suggests that with MR fluids the wall slip cannot be reliably detected by comparison of flow curves measured at different gap heights.

ACCURATE PREDICTION OF FRICTION FACTOR AND NUSSELT NUMBER FOR SOME DUCT FLOWS OF POWER-LAW NON-NEWTONIAN FLUIDS

Using the high-order finite element method, we provide accurate predictions of the friction factor and the Nusselt number for the fully developed laminar flows of power-law-type non-Newtonian fluids in rhombic, isosceles-triangular, elliptical, and semielliptical ducts. Two kinds of thermal boundary conditions, commonly referred to as the H1 and T conditions, are considered. For each type of duct, computed results are presented for several aspect ratios with the power-law index ranging from 0.2 to 1.0. Mesh refinement tests with four different meshes are carried out for two specific duct configurations, indicating that the present predictions are remarkably insensitive to mesh density.

Sedimentation stability and rheological properties of ionic liquid–based bidisperse magnetorheological fluids

The sedimentation stability and rheological properties of ionic liquid–based magnetorheological fluids comprising a mixture of micron- and nano-sized particles were experimentally studied. Three different fluids with the same total particle concentration of 15 vol% were prepared for testing: one containing only microparticles and two others in which 5 or 10 wt% of the microparticles were replaced by nanoparticles. The nanoparticles were surface stabilized against oxidation. For comparison purposes, silicon oil–based magnetorheological fluids with similar solid fractions were also prepared and tested. The results indicate that, with ionic liquid as a carrier fluid, the addition of nanoparticles at 10 wt% reduces the sedimentation rate almost by an order of magnitude from that without nanoparticles, while the reduction in the dynamic yield stress is only marginal. The ionic liquid–based fluids also had a better dispersion of particles.

Inmould integration of a microscope add-on system to a 1.3 Mpix camera phone

A microscope add-on device to a 1.3 Mpix camera phone was selected as a demonstrator system for testing inmould integration of electronic substrates and plastic optics. Optical design of the device was quite challenging due to the fact that illumination system needed to be integrated with a double aspheric singlet lens structure as a single optical piece. The designed imaging lens resolution was adequate to resolve 10 &mgr;m features with a mobile phone camera. In the illumination optics the light from LEDs embedded into the plastic structure was collected and guided to the surface that was imaged. Illumination was designed to be uniform and adequately bright to achieve high resolution images with the camera phone. Lens mould design was tested by using injection moulding simulation software. The critical mould optical surfaces were designed as separate insert parts. Final shapes producing lens surfaces were tooled by diamond turning on nickel coatings. Electronic circuit board inserts with bonded bare LED chips and packaged SMD LEDs were assembled to the mould and then overmoulded with optical grade PMMA. Experiences proved that inmould integration of electronic substrates, bare LED chips and high resolution imaging optics in injection-compression moulding process is feasible. The yield of embedded packaged and also bare chip components was close to 100% after the right injection moulding process parameters were found. Prototype add-on system was characterized by testing the imaging properties of the device with a camera phone.