Temperature Analysis

[Clay1969]

For transient heat transfer analysis one should define the initial BCs first. Initial BCs are difficult to predict within the system. Hence it is advised that one first runs Steady state analysis (Boundary Value Problem) with assigning BC temperature at Measured points . THe STep of Steady State temperature should be followed by Transient Heat Transfer Step. DO you agree with me?

 

[corus]

It all depends. If you assign measured temperatures at discrete points then you might see these as hot spots in your results. It's better to assign proper boundary conditions that you think will apply and then fine tune these to give predicted values that will correspond with your measured values, as best they can, at those discrete positions.

corus

 

[Clay1969]

corus
But still I need to run two analyses
Corus : What do you mean boundary conditions that you think will apply
A clearer answer is highly appreciated
Clay1969

 

[Clay1969]

Sorry corus Still I need two steps not two analyses

 

[corus]

It's difficult to put a clearer answer when the question is non-specific. In most cases I'll start with a structure that is at room temperature and then build up temperatures through transient changes. In some cases I might know that steady state temperatures have been achieved and then wish to investigate some kind of transient. In all cases I'll try and verify the answers by common sense/analytical means or preferably validate the answers by measurements. If the model results differ from measurements then I might 'tweak' the model to give better results in line with measured values. For transients you can start with initial temperatures or a steady state run before running the transient analysis in one analysis.
For 'boundary conditions I think that may apply' in a thermal analysis I refer to assumptions regarding fluid flow, or whether fixed temperatures apply, or whether radiation is significant or not, or whatever. It all depends.

corus