# Transient heat flow

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

 June 1, 2011, 12:17 Transient heat flow #1 Senior Member   --------- Join Date: Oct 2010 Posts: 303 Rep Power: 10 I'm totally new to electronics cooling and transient simulations. I had a newbie question. I was simulating the air flow inside a power amplifier cabinet which had Fans, Power sources and power amplifiers inside. The power emanation from the amplifiers is of a transient nature and is described by the following details: Output Power = 200 W Duty cycle: Transient attribute:Gaussian, Total pulse duration: 6400 ms So I see that I need to consider 6400 ms as the total time with a time step of 400 ms. But this is something relevant to a single duty cycle so will it be fine to analyze the heat flow inside the cabinet just for a single duty cycle on these lines. Can some one please suggest me regarding this. __________________ Best regards, Santhosh.

 June 7, 2011, 05:22 #2 Senior Member   Boris Marovic Join Date: Jul 2009 Posts: 468 Rep Power: 15 Hi Santhosh, First I wouldn't define 400ms time steps as your calculation would only have 16 Iterations which is way too short to get a flow or temperature distribution in your model. The time step should be the size to resolve what is happening in the flow. So I would suggest at least around 10ms. That way you have 640 iterations for one cycle. You can monitor the parts heat flux in a volume goal and should see how rough the steps between the iterations are. That way you would understand that 400ms is far to coarse. Then if you run only one cycle your whole model is most certainly not reached any developed state as it would be as if the model would run for several hours. One thing you could do is run a steady state analysis to get the general flow field converged and use these results for a following transient calculation. You should therfore use an effective averaged value of your heat source as the steady state would be not for the full power of 200W. When you've done that steady state project, simply clone the project and make the changes for the transient project. If you then go into the general settings you can select "transfered" instead "user defined" in the top row of the initial condition settings and link it to the staedy state project. That way it uses the staedy state flow as the initial flow for the transient and the amount of overall cylces would be much less than if you would calculate just transient. So instead of running 200 cycles you could try just 10-20 and see if the flow field in general changes any further in it's peak values. It should look like starting from a constant value (the steady state value) and then start to oszillate as you run through your min and max values of your heat source cycles and at some point these min and max values shouldn't change any more, somthing similar to the image attached. It is best to deactivate the maximum time and just let it run so you can abort it when you think it's fine. I hope this helps, Boris

 June 7, 2011, 08:45 #3 Senior Member   --------- Join Date: Oct 2010 Posts: 303 Rep Power: 10 Many thanks for the suggestions Boris. Could you please tell me how to calculate the effective averaged value of the heat source for the steady state. I really apologize for asking such newbie question but as I mentioned it i'm really new to this so your suggestions can certainly help me learn better. __________________ Best regards, Santhosh.

 June 7, 2011, 09:13 #4 Senior Member   Boris Marovic Join Date: Jul 2009 Posts: 468 Rep Power: 15 Ups, sorry bad translation. It is the RMS (Root Mean Square) value. Boris

 Thread Tools Display Modes Linear Mode

 Posting Rules You may not post new threads You may not post replies You may not post attachments You may not edit your posts BB code is On Smilies are On [IMG] code is On HTML code is OffTrackbacks are On Pingbacks are On Refbacks are On Forum Rules

 Similar Threads Thread Thread Starter Forum Replies Last Post glenn creten Main CFD Forum 1 August 14, 2012 05:10 Sas CFX 15 July 13, 2010 08:56 rajuks Main CFD Forum 0 February 15, 2010 10:04 doodek Main CFD Forum 2 November 23, 2009 09:48 Matthieu Ubas Main CFD Forum 2 November 5, 1999 15:20

All times are GMT -4. The time now is 18:40.