IceCube CFD Drilling Model

A numerical model of a hot water drill used to produce deep holes in clear ice at the South Pole for the IceCube neutrino observatory program scheduled for completion in 2010 has been developed. The model was built using the ANSYS commercial computational fluid dynamics (CFD) code, ANSYS CFX. This drill model is helping us to understand the water/ice melting process near the bottom of the drill hole, and to evaluate the influence of nozzle size, spray angle, water flow rate, and water temperature on the drill hole shape, and on drilling speed. The basis for the model is ANSYS CFX, which has multi-phase, conjugate heat transfer capabilities. The model utilizes a multi-phase approach, and simulates motion of the drill with respect to the ice. The sensitivity of model predictions to mesh resolution, turbulence model, and interfacial heat transfer coefficients, area, and drag coefficient was studied, and the results were used to determine preferred values in each case. This multi-phase model was selected after evaluating an equilibrium model and obtaining results showing not completely satisfactory comparisons to experimental data from the South Pole. Computations at a drill depth of 1292 m allowed validation of code results using actual field data obtained during the 2004–2005 IceCube drilling season at Antarctica. A series of steady-state runs using two drill sizes, two drill speeds, and one spray angle were performed for conditions at 1292 m to determine if a smaller nozzle orifice would enable faster drilling, preferably by a factor of two. The model predicted a drill hole diameter of from 18 in to 25 in and an up-flow water temperature of from 20°C to 28°C. The drill hole diameter is consistent with values measured at the IceCube site, but the water temperature is about 10°C low. No evidence of the nozzle tip impacting the bottom of the hole was found in the drill speed range 3.5 ft/min to 7 ft/min. A nozzle spray angle of 25 degrees was found to make little difference in hole depth or diameter. Reducing the nozzle diameter from 1 in to 0.75 in at the same water volumetric flow rate resulted in an increase in the drill hole depth by from 16% to 20%. The latter result implies that faster drilling is expected when using a smaller size orifice with zero degree spray angle. The IceCube drill model is now available to determine the effects of key variables, to evaluate the performance of new nozzle designs, and to specify drill speed versus depth. Recommendations specific to faster drilling speeds resulted in a near doubling of actual speed during the 2005–2006 season at the South Pole.