Question about using turbulence model to laminar flow
Would using turbulent flow model(like K-Omega or Reynolds stress model) to laminar flow bring large inaccuracy?
You mean what happens if we use a turbulent model in a laminar flow?
Turbulent flows have different physics, so we use different turbulent equations because the Navier-Stokes equations do not give a correct answer when the flow enters into the turbulent territory (for example Newton laws cannot be applied in quantum physics). On the other hand, solving a laminar flow with a turbulent model does not give a correct answer either and we have to be sure about the flow regime, and then choose the models.
what about domain with different flow regimes?
For instance in some part of domain you can consider a laminar flow (Re < 2100) but in other flow is turbulent or transient
What turbulence model should be applied in this case?
Pure laminar flow:
In A perfect turb model, eddy viscosity should return to zero in the framework of EVM, however this is not always true. So, Most turb models are dissipative by nature, and if your flow does not have turbulent dissipation, and extracting energy from it would surely cause deviation. <Needs more input here .... >
Be aware that the models you listed are supposed to run in FT mode, they are not working well in transitional flow. Huge error would occur. There are models can handle this, but of cause not to predict natural transition.
Pay special attention, that in lots of industrial flow, regions of laminar exist, turb models you listed are still having great difficulties in these flows.
Thanks for your answer. I have to use Reynolds stress model to fluid transforming from turbulent to laminar.(turbulent flow go into a channel with complex wall geometry) So I have to use LRR model to turbulent, transitional, laminar fluid at the same time. The Re is about 1000.
So I'm interested to know: How inaccurate would Reynolds stress model be for transitional and laminar flow?
|All times are GMT -4. The time now is 03:07.|