[pachek]

Hi. I'm simulating water through pipes while adapting the design for cooling performance while keeping the pressure drops below certain limit. Water flows through a DN32 pipe with an average speed of 5 m/sec. Does the roughness height has any significant influence on the results? Mainly interested in the delta P. I left the fluent default parameters but one colleague suggest me that the simulation shall be done taking into account the corrosion over time and subsequent roughness increase. Main problem is that I have no specific data for stainless steel pipes corrosion aging and secondly this question I'm putting here on which is the real impact.

Alternative thinking that having access to the formulas used to resolve the wall function of k-epsilon model or the Omega I could roughly calculate the impact and if it is worth to be considered

Thanks

[LuckyTran]

Well. In short if the effect of roughness was significant, you wouldn't be able to model it in Fluent anyway.

If your interest is the deltaP you can calculate quite easily the effect of surface roughness from the Colebrook-White equation (aka the Moody chart). This is of course in a straight pipe, but it quickly lets you determine what is the impact.

Fluent models roughness using the sand-grain roughness approach. It is debatable whether this at all represents the roughness you find in practical pipes or corrosion. But if it did, the effect is marginal. The effect of the sand-grain roughness is to modify the law of the wall. As far as I know this should affect only the momentum equation and the turbulence equations are unaffected.

These equations are available in the Fluent User's Guide, chapter 6/7 depending on your version. Look for Wall Boundary Conditions. But I googled it on the internet here. Notice that for very small roughness in the hydrodynamically smooth regime there is no impact. Scale build-up (being only a few microns thick) often falls into this category.

[pachek]

Thanks a lot! this answer is great hep