FloVENT and PCR method question

Martin Hegedus · February 22, 2012, 23:03

Sorry, I'm from the compressible CFD world and this may seem like an ignorant question to incompressible folks.

Anyway, I was checking out the data sheet for FloVENT and the solver section says "The conjugate nature of heat transfer is concurrently solved using a preconditioned conjugate residual solver together with a flexible cycle multi-grid solution technique."

Not really sure what that is. Is this a time accurate solver? FloVENT is used for HVAC of rooms so I assume it is. But, in my compressible world, a multi-grid solution method is not time accurate.

Also, what equations are begin solved and how are they coupled? I assume that the conservation of mass is incompressible. So is the energy equation coupled to the momentum with the equation of state or is the density in the momentum equation also incompressible (i.e. constant), thus uncoupling the energy equation? However, uncoupling the energy equation seems opposite of what "conjugate nature of heat transfer" means. Also, if the energy equation is uncoupled, I assume buoyancy is not modeled. Which seems odd. Is buoyancy modeled?

RobinB · March 2, 2012, 08:58

Density can either assumed to be constant (valid for low mach no. and, for HVAC, when you don't have the massive density variations you might encounter with fire modelling) or derived using the the ideal gas law ((molecular weigth x (p+ref p))/(RxTabs)). One thing FloVENT doesn't do is shock capture which, for HVAC, I think is OK.

Buoyancy is modelled either using the Boussinesq assumption when density is assumed to be constant (gravitational force per unit volume = g x expansivity x density x (T - Tref)) or if density is set to be variable gravitational force per unit volume = g x (density - reference density).

Multi grid solver is used to accelerate the solution of the temperature variable, that's all.

Going back to your question, the conservation of mass therefore need not assume constant density and thus can be considered compressible.

We normally say that if you get dTs in the model > 60degC leading to about >20% change in volumetric expansion then use the ideal gas law.

You can turn off buoyancy and thus decouple the hydrodynamic and temp equations but should be done with caution, only when forced convection dominates. If in doubt do two models, one with grav on, the other off, if they differ, use grav on.

If you want any more feel free to contact me at robin_bornoff@mentor.com. I'm the FloVENT product manager by the way.

Martin Hegedus · March 2, 2012, 11:03

Hi Robin, thank you for your response. In regards to both the content and posting it here at cfd-online. If I have any questions about FloVENT I'll email you directly. But, unfortunately and at this point, it will probably have to wait for the next project.

February 22, 2012, 23:03	FloVENT and PCR method question	#1
Martin Hegedus Senior Member Martin Hegedus Join Date: Feb 2011 Posts: 500 Rep Power: 19	Sorry, I'm from the compressible CFD world and this may seem like an ignorant question to incompressible folks. Anyway, I was checking out the data sheet for FloVENT and the solver section says "The conjugate nature of heat transfer is concurrently solved using a preconditioned conjugate residual solver together with a flexible cycle multi-grid solution technique." Not really sure what that is. Is this a time accurate solver? FloVENT is used for HVAC of rooms so I assume it is. But, in my compressible world, a multi-grid solution method is not time accurate. Also, what equations are begin solved and how are they coupled? I assume that the conservation of mass is incompressible. So is the energy equation coupled to the momentum with the equation of state or is the density in the momentum equation also incompressible (i.e. constant), thus uncoupling the energy equation? However, uncoupling the energy equation seems opposite of what "conjugate nature of heat transfer" means. Also, if the energy equation is uncoupled, I assume buoyancy is not modeled. Which seems odd. Is buoyancy modeled?

March 2, 2012, 08:58		#2
RobinB New Member Robin Bornoff Join Date: Mar 2009 Posts: 23 Rep Power: 17	Density can either assumed to be constant (valid for low mach no. and, for HVAC, when you don't have the massive density variations you might encounter with fire modelling) or derived using the the ideal gas law ((molecular weigth x (p+ref p))/(RxTabs)). One thing FloVENT doesn't do is shock capture which, for HVAC, I think is OK. Buoyancy is modelled either using the Boussinesq assumption when density is assumed to be constant (gravitational force per unit volume = g x expansivity x density x (T - Tref)) or if density is set to be variable gravitational force per unit volume = g x (density - reference density). Multi grid solver is used to accelerate the solution of the temperature variable, that's all. Going back to your question, the conservation of mass therefore need not assume constant density and thus can be considered compressible. We normally say that if you get dTs in the model > 60degC leading to about >20% change in volumetric expansion then use the ideal gas law. You can turn off buoyancy and thus decouple the hydrodynamic and temp equations but should be done with caution, only when forced convection dominates. If in doubt do two models, one with grav on, the other off, if they differ, use grav on. If you want any more feel free to contact me at robin_bornoff@mentor.com. I'm the FloVENT product manager by the way. Martin Hegedus likes this.

March 2, 2012, 11:03		#3
Martin Hegedus Senior Member Martin Hegedus Join Date: Feb 2011 Posts: 500 Rep Power: 19	Hi Robin, thank you for your response. In regards to both the content and posting it here at cfd-online. If I have any questions about FloVENT I'll email you directly. But, unfortunately and at this point, it will probably have to wait for the next project.