I used the E.P.A.'s EPAnet program to run a network analysis. I noticed that if I change the roughness of the main pipe connecting the reservoir with the first junction, the pressures do change on the network but all the velocities are same! Also if I do change the roughness on an internal different pipe, pressures but also velocities change. Why does it have that behavior? I heard that it's related to the supply remaining the same and I will look into it but I was also wondering for more sources of information. I'm particularly surprised of the fact the main pipe's roughness change doesn't change any velocity on the network (even though it does change pressures) but an internal pipe's roughness affects both. Formula for head losses used: Darcy-W. I was instructed it's the most suitable.