thermal entrance length in pipe network of varying diameters

[trich122]

Has anyone come across some research regarding the thermal entrance length for a pipe that has many steps / changes in diameters? I suppose the same question follows for hydrodynamic entrance length

I am wondering how to calculate the Graetz number (Dh/L)*Pr*Re and wondering how "long" my pipe should be (i.e. should it be the entirety of the pipe network, or should it "start" again from each change in cross sectional shape in the network)

any direction would be greatly appreciated!

Thanks!