heated long tubes

Richard

I am investigating heat transfer from laminar flow fluids in air chillers. My models, for comparison with experimental results, consist of two tubes 9.14mm in diameter, under constant surface heat flux conditions , one entering and one exiting a u-bend. Surface temperature results both in the inlet and exit tubes compare excellently with experimental results up to a distance of approximately 0.5m down the tube. At this point the surface temperature of the pipe begins to decrease despite the constant surface heat flux. I have also investiagted longer tubes and an oscillating suface temperature can be found to occur which unsurprisingly does not occur in reality.

[1]Does anybody know of any such issues encountered in modeling flow through long thin pipes or how best to approach this problem?

[2]Which solver is best for these type of problems, coupled, segregated, implicit, expilicit etc.?