Learning Forum

[gartz89]

Hello, I'm reasearching the natural ventilation coupled with mechanical ventilation "FAN". The pressure difference between the external and internal pressures is the stack pressure available which can be created naturally by density differences to drive a ventilation flow through the building. Unfortunately, this stack effect in multi-storey buildings arises from the buoyancy forces which influence on the efficiency of the "small FAN" (imagine these in the bathroom). When I solve the model equations my solution gives information about the velocity distribution from the buoyancy forces, temperature distribution, pressure etc. 

Everything looks good, but when I coupled some FAN as boundary conditions I would like to check what is the difference between the flow rate of every single FAN at the floors. The floors are five, the fans are 5, I add intake fan boundary conditions and add pressure jump for characteristics of the FAN. But there isn't influence from the inlet velocity if I compare with only natural ventilation. 

Some boundary conditions I miss or something I have to repair in the cell zone condition I'll be glad to help me. Regards!

[Rob]

Did you use a unique boundary label for each fan?

[gartz89]

Thank you for your reply.
Yes, the boundary conditions for each fan are the same, but I don't know what did you mean about this unique boundary label. Could you give me more details? Thank you.

[Rob]

If each fan is the same label you can't get the individual mass flow, you'll get the sum/mean from the five surfaces. If they're the same condition but different labels you can report the results for each surface separately.

[gartz89]

Yes, but unfortunately the task is different, it is looking for the velocity inlet for each FAN. When the natural convection starts everything looks good, but in real life, these five inlets have some loss coefficient and pressure jump for each FAN. I think if have 5 more domains at the inlet and if I specified the BC for inlet vent and interface zone as a FAN (interior) and add the pressure jump there. It could give some interesting results approximately.
That's the task that I'm looking for... If you have one more idea you could say it. Thank you for your reply, Sir.

[Rob]

If you have 5 distinct and uniquely labelled inlets with a fixed velocity you can look at the pressure to see if there's a difference. Intake-fan allows fan curve data to be used.

[gartz89]

Dear friend I solved the problem! As I said I made 5 more domains to specified the loss coefficient of the vent. After that, I add 5 interface zone Fan boundary conditions between vent domains and the main computational domain. Finally, I specified the fan curve data as a pressure polynomial function of limits of the velocity.

So thank you for your reply. The simulation of the natural ventilation coupled with mechanical ventilation was already done. Regards!