Jet diameters and velocity profiles in continuous ink-jets

Theoretical predictions of the diameters of continuous ink-jets downstream of long nozzles are generalized to include the important cases of ink-jet fluids and shorter nozzles where the velocity profile at the nozzle exit is undeveloped (non-parabolic). Comparisons of the new predictions with experiments and simulations are made for fairly long nozzles with tapered profiles and short nozzles with conical profiles; experimental and simulated profiles are also compared downstream of the nozzle exit for both industrial and large scale ink-jet print heads. Precise measurements of the un-modulated jet diameters downstream of the nozzle exit can set really useful limits to the possible shapes of the flow profile right at the nozzle exit, and in particular allow some assessment of the axial velocity gradients and fluid shear rates at the nozzle exit where direct speed measurement is usually impractical. Simulations allow further study of the relaxation of the velocity profile downstream of the nozzle exit, and are reported for both un-modulated and modulated CIJ jetting. Implications of this work include speeding up CIJ simulations, absolute calibration of the applied CIJ system modulation, and the likely magnitude of dynamic surface tension effects on observed CIJ satellite speeds.

Rigorous finite-time guarantees for optimization on continuous domains by simulated annealing

Simulated annealing is a popular method for approaching the solution of a global optimization problem. Existing results on its performance apply to discrete combinatorial optimization where the optimization variables can assume only a finite set of possible values. We introduce a new general formulation of simulated annealing which allows one to guarantee finite-time performance in the optimization of functions of continuous variables. The results hold universally for any optimization problem on a bounded domain and establish a connection between simulated annealing and up-to-date theory of convergence of Markov chain Monte Carlo methods on continuous domains. This work is inspired by the concept of finite-time learning with known accuracy and confidence developed in statistical learning theory.