i want to simulate heat transfer enhancement using twisted tape insert for turbulent flow but want to know which viscous model should i use and how i input heat energy in the tube surface (using heat flux or temperature source). if insert always rise the outlet fluid temperature then plain tube? should this cfd study be a steady case?
Depending on the problem lenght scale, Reynolds number and computational scale you could use a DNS using only the physical value of the viscosity of the fluid. If you are not able to perform a DNS you could think to use LES with the dynamic eddy viscosity where you do not have to specify a value for the turbulent viscosity. If also LES is too computational expensive for you, then you have only RANS.
https://www.sciencedirect.com/search?qs=viscous+model+for+CFD+study+of+twisted+tape+insert+for+turbulent+flow+region&authors=&pub=&volume=&issue=&page=&origin=home&zone=qSearch
I've mostly seen k-epsilon RNG is used for Twisted Tape inserts in some articles for reynolds varying from 10'000 to 20'000.
If you are using RANS, the RNG and Realizable k-epsilon turbulence models with enhanced wall functions probably work well for you.
- Badges
- Science topic