LM1085: Thermal Information difference between Datasheet and Model

Hello Mark:

 Would you give me more information on what model you refer to that gives you theta jc of 3 C/W and theta jb of 3.043 C/W?

For the LM1085, KTT package has significant higher thermal resistances than the NDE package. In high power dissipation condition( like 4W), NDE could be a better option. 

Regards,

Min

Hi Min, I would be linking the customer to give you additionals details. Thank you very much

Hi,

The model that used theta jc of 3 C/W and theta jb of 3.043 C/W was referring to was the historical values that Raytheon used in the past to model this component. I am trying to verify what we should use when modeling junction to board.  Should NDE and KTT have that different of resistant values?  Lots of component manufactures will only give a theta jc value junction to case (bottom).  When they give a theta jb value then they assume how the component should be mounted to the board.  I am just looking for some clarification on this part.

Then the theta jb value of 23.3 for KTT had been verified and is accurate? It makes sense that it is higher, but that number doesn't seem right, unless my power dissipation assumption of 4W is incorrect. Thank you.

Hi Aaron:

  I have forward your concern to the engineer who is in charge of the package thermal modelling. I'll let you know by the end of this week. 

Regards,

Min

Hi Aaron:

One possibility for the theta jb difference between the 2 packages is that the NDE package did not have the pins connected to the board in the thermal analysis since it is not a typical SMT package.  Because of this, Theta-JB would be lower.  The pins of the KTT package likely did contact the board, which would make Theta-JB higher.

Regards,

Min

I still have not received the answer that I am looking for.  I am NOT interested in the NDE package.  I just want to know if the R_theta_j_b value for the KTT package is correct and has been validated?  Is 23.1 degC/W correct? Where did it come from?