How do you create your own boundary conditions?

greg_2666 · March 11, 2015, 13:11

Hi all, for a project I'm modelling a data centre in Star CCM+.

I need a velocity inlet (to a server) to have a zero heat flux boundary condition. Is there any way of programming this/ have I overlooked it?

The other issue is the outlet temperature of the servers depend on the inlet temperature. Is there any way of saying at the start of each iteration check the corresponding inlet for the temperature of the flow?

Thanks

Greg

fluid23 · March 11, 2015, 19:37

Perhaps I misunderstand your issue, but as far as I know the velocity inlet does not have a heat flux component to the definition. Are we talking constant temperature? You can define heat flux at a wall BC by going in to Regions > Boundary > walls and define a heat flux under physics conditions. That option does not appear for velocity inlet.

As far as the outlet temperature goes, you can define a field function to set the outlet temp as a function of inlet temp. Those are pretty easy, just refer to the help documentation on field functions.

greg_2666 · March 12, 2015, 06:09

Quote:

Originally Posted by MBdonCFD

Perhaps I misunderstand your issue, but as far as I know the velocity inlet does not have a heat flux component to the definition. Are we talking constant temperature? You can define heat flux at a wall BC by going in to Regions > Boundary > walls and define a heat flux under physics conditions. That option does not appear for velocity inlet.

As far as the outlet temperature goes, you can define a field function to set the outlet temp as a function of inlet temp. Those are pretty easy, just refer to the help documentation on field functions.

Thanks Ill give that a go.

G

lcarasik · March 12, 2015, 13:01

Where is your heat being input into your control volume? Are your server racks producing heat and heating up the air flowing through the data center? Also, are you just looking for the flow profile through the data center?

greg_2666 · March 13, 2015, 06:16

Quote:

Originally Posted by lcarasik

Where is your heat being input into your control volume? Are your server racks producing heat and heating up the air flowing through the data center? Also, are you just looking for the flow profile through the data center?

The heat should be coming from the servers as the air flows through. I'm looking at how to minimise recirculation and bypass through varying flow rate from the CRAC unit.

lcarasik · March 13, 2015, 12:40

I would start your analysis using an isothermal flow simulation. This should give you an idea of flow physics before you start attempting to model the heat transfer occurring. You should not attempt to set your outlet boundary condition as a function of your inlet.

After running an isothermal set of test runs, I would recommend engaging the segregated fluid temperature model. Set your inlet boundary to a velocity boundary and define your inlet temperature (I'm assuming a set inlet temperature that does not vary). On your servers, I would set their boundaries (or an approximation it) using a set temperature or heat flux that the air is exposed to. Set your outlet boundary to a pressure outlet and you shouldn't have to set the temperature for it. (Also, I recommend looking into invoking the buoyancy source terms if there is a large enough density change of the air after getting heated by the servers.)

This should get you started, but I'd recommend you read over meshing sensitivity studies, basic turbulence modeling, and heat transfer modeling. Also, I recommend you contact CD-Adapco (if you are able to) with questions like this. They should be able to provide more guidance.

March 11, 2015, 13:11	How do you create your own boundary conditions?	#1
greg_2666 New Member Gregory Hannam Join Date: Mar 2015 Posts: 7 Rep Power: 11	Hi all, for a project I'm modelling a data centre in Star CCM+. I need a velocity inlet (to a server) to have a zero heat flux boundary condition. Is there any way of programming this/ have I overlooked it? The other issue is the outlet temperature of the servers depend on the inlet temperature. Is there any way of saying at the start of each iteration check the corresponding inlet for the temperature of the flow? Thanks Greg

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Domain Imbalance	HMR	CFX	5	October 10, 2016 05:57
Setting rotating frame of referece.	RPFigueiredo	CFX	3	October 28, 2014 04:59
fluent add additional zones for the mesh file	SSL	FLUENT	2	January 26, 2008 11:55
[Commercial meshers] Trimmed cell and embedded refinement mesh conversion issues	michele	OpenFOAM Meshing & Mesh Conversion	2	July 15, 2005 04:15
Convective Heat Transfer - Heat Exchanger	Mark	CFX	6	November 15, 2004 15:55

March 11, 2015, 19:37		#2
fluid23 Senior Member Matt Join Date: Aug 2014 Posts: 947 Rep Power: 17	Perhaps I misunderstand your issue, but as far as I know the velocity inlet does not have a heat flux component to the definition. Are we talking constant temperature? You can define heat flux at a wall BC by going in to Regions > Boundary > walls and define a heat flux under physics conditions. That option does not appear for velocity inlet. As far as the outlet temperature goes, you can define a field function to set the outlet temp as a function of inlet temp. Those are pretty easy, just refer to the help documentation on field functions.

March 12, 2015, 13:01		#4
lcarasik Senior Member Lane Carasik Join Date: Aug 2014 Posts: 692 Rep Power: 14	Where is your heat being input into your control volume? Are your server racks producing heat and heating up the air flowing through the data center? Also, are you just looking for the flow profile through the data center?

March 13, 2015, 12:40		#6
lcarasik Senior Member Lane Carasik Join Date: Aug 2014 Posts: 692 Rep Power: 14	I would start your analysis using an isothermal flow simulation. This should give you an idea of flow physics before you start attempting to model the heat transfer occurring. You should not attempt to set your outlet boundary condition as a function of your inlet. After running an isothermal set of test runs, I would recommend engaging the segregated fluid temperature model. Set your inlet boundary to a velocity boundary and define your inlet temperature (I'm assuming a set inlet temperature that does not vary). On your servers, I would set their boundaries (or an approximation it) using a set temperature or heat flux that the air is exposed to. Set your outlet boundary to a pressure outlet and you shouldn't have to set the temperature for it. (Also, I recommend looking into invoking the buoyancy source terms if there is a large enough density change of the air after getting heated by the servers.) This should get you started, but I'd recommend you read over meshing sensitivity studies, basic turbulence modeling, and heat transfer modeling. Also, I recommend you contact CD-Adapco (if you are able to) with questions like this. They should be able to provide more guidance.