Large room flow quantification

F.Andre · April 1, 2019, 16:16

I was wondering if anyone has any experience with large room flow modelling with good success.

I have been trying to model a room of 20m x 10m x 10m using Ansys CFX, it seems just as what I can find in the literature that the error between measured and simulated speed is on the same magnitude as the speed itself close to the floor level O(0.5 m/s). (Assuming that the flow inlet is at the ceiling with inlet flow O(10 m/s)).

- I have not obtained mesh independent solution with meshes on the order of 10M nodes.

- I am considering SST, RNG and k-eps turbulence models and they all give similar errors.

- Steady-State and Isothermal assumption are made.

Am I using the wrong software or doing something wrong? Is CFD just not that accurate for such big domains?

ghorrocks · April 1, 2019, 18:28

CFD, including CFX is extensively used in HVAC applications. So used correctly it is accurate enough to be useful.

The first step is to look in the literature and see how they set up their simulations.

But here are some comments from me: HVAC flows commonly involve large scale flow movements so steady state simulations may not be appropriate. Also isothermal conditions might not be suitable, buoyancy effects could be important.

April 1, 2019, 16:16	Large room flow quantification	#1
F.Andre New Member Andre Join Date: Apr 2019 Posts: 1 Rep Power: 0	I was wondering if anyone has any experience with large room flow modelling with good success. I have been trying to model a room of 20m x 10m x 10m using Ansys CFX, it seems just as what I can find in the literature that the error between measured and simulated speed is on the same magnitude as the speed itself close to the floor level O(0.5 m/s). (Assuming that the flow inlet is at the ceiling with inlet flow O(10 m/s)). - I have not obtained mesh independent solution with meshes on the order of 10M nodes. - I am considering SST, RNG and k-eps turbulence models and they all give similar errors. - Steady-State and Isothermal assumption are made. Am I using the wrong software or doing something wrong? Is CFD just not that accurate for such big domains?

April 1, 2019, 18:28		#2
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,711 Rep Power: 143	CFD, including CFX is extensively used in HVAC applications. So used correctly it is accurate enough to be useful. The first step is to look in the literature and see how they set up their simulations. But here are some comments from me: HVAC flows commonly involve large scale flow movements so steady state simulations may not be appropriate. Also isothermal conditions might not be suitable, buoyancy effects could be important. __________________ Note: I do not answer CFD questions by PM. CFD questions should be posted on the forum.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Review: Reversed flow	CRT	FLUENT	1	May 7, 2018 05:36
Modelling room ventilation air flow	todda7	Main CFD Forum	0	April 12, 2017 14:51
What may lead to a large value of k on the wall in a channel flow?	Sunxing	OpenFOAM	0	November 17, 2012 07:42
buoyantBoussinesqSimpleFoam - Turbulent transient flow in a room with inlets and outl	Mojtaba.a	OpenFOAM Running, Solving & CFD	3	August 22, 2012 07:56
Sealed room air flow	dave_k	FLUENT	0	March 15, 2010 03:46