Register Blogs Members List Search Today's Posts Mark Forums Read

 July 29, 2013, 03:19 About periodic boundary condition. #1 New Member   Ilias Join Date: Nov 2012 Location: QLD, Australia Posts: 2 Rep Power: 0 Could any one please give me the answer of the follwing question? What is mean by upstream bulk temperature at periodic condistions? If temperature is changing with respect to time in my periodic boundary, can I use periodic boundary condition? And what will be the value of upstream Bulk Temperature when temperature vary along the periodic boundary with respec to time? Adavance thanks

 July 29, 2013, 05:08 #2 Senior Member     Paolo Lampitella Join Date: Mar 2009 Location: Italy Posts: 2,120 Blog Entries: 29 Rep Power: 38 There are several methods known in literature to handle periodic heat transfer calculations, the main fact being that it is probably a not really well posed problem. However, the basics in Fluent is that, for a periodic heat transfer calculation, unless nothing interesting happens in your domain, some temperature difference will arise between your inlet and outlet periodic faces. So the problem is, how can Fluent manage the fact that from one time step (iteration) to the next the temperature will keep changing (for fixed heat flux boundary conditions) or converging toward a constant temperature (for fixed wall temperature)? The method adopted in Fluent is based on the fact that, for the two conditions above, the overall jump (between the periodic faces) in the bulk temperature can be somehow computed; as a consequence it is just subtracted before mapping the temperature from the outlet to the inlet. The way it is actually done is by modifying the coefficients coupling the cells adjacent to the periodic faces, so that it is not anymore a straight periodic condition but a scaled version. In order for this method to work, the reference bulk temperature at inlet is required. This is just the mean temperature (or something similar) and will not actually affect your computation besides setting a reference value for the temperature in the domain (indeed you are not allowed, for periodic heat transfer computations, to have temperature dependent coefficients). Hence the periodic condition will give you the same spatial pattern for the temperature on the two faces, just that one is scaled with respect to the other. Is this method time accurate? I don't know but i don't think so. Is it possible to have the bulk temperature specified at each time step: I don't know but i don't think so.

 July 29, 2013, 21:18 #3 New Member   Ilias Join Date: Nov 2012 Location: QLD, Australia Posts: 2 Rep Power: 0 Thanks sbaffini. I am sloving for transient case and also mean temperature or free stream bulk temperature is changing with respect to time. I dont have any idea how to put bulk temp at every time step. I dont think so it is practicable. My question is : Since its changing with respect to time, so that my assumption never be accurate. Now if my assumption is wrong, how they are going to effect my result or can I neglect this error? Advance thanks...

 July 30, 2013, 07:37 #4 Senior Member     Paolo Lampitella Join Date: Mar 2009 Location: Italy Posts: 2,120 Blog Entries: 29 Rep Power: 38 At this point, i would like to ask you what are you exactly trying to model as there might be more options to handle your specific case. Can you give some additional details?