I want to also top cool the HMC440QS16G; do you have a theta-JC number I can use?
Per the datasheet the Junction to Case (ground paddle) thermal resistance is 17.6°C / W on the HMC440QS16G for roughly 22°C temp rise; at 85°C this results in a 107°C for the junction temperature. Not quite sure why you need additional cooling at such a low Tj. We don't have data to the top of the plastic package if that's what you're looking for.
I'm assuming the 5V at 250 mA is for the part, NOT counting the load.
Trying to increase the system reliability and lower all the TJ's on all the parts.
I'm in Central Apps, and this came in from a mechanical/thermal engineer from Raytheon,
a strategic customer for us in McKinney, TX. He replied:
"To answer your question "Not quite sure why you need additional cooling at such a low Tj",Current CCA thermal model with NO top cooling on this component predicts a board temperature of 106 C, giving a Tj of 128 C (106 + 1.25W x 17.6 C/W).This is for a chassis rail temp of 43 C, which is very low for our types of design in mil spec environments.I have been asked by the project to determine how high the rail temperature can go before we exceed a Tj max, which for this component is 135 C. So we can only increase the rail temperature by 7 degrees (135 - 128) if we don't add top cooling for this component.FYI there is an aluminum cover above this component that we can connect by thermal gap filler to the top of this HMC440QS16G.But to determine how effective this approach will be I need to have the Theta (j to case-top) thermal resistance. If I try and "guesstimate" what it might be, I would calculate the thermal resistance from an internal die to the top surface of the plastic package, orR = L / k A = 0.03" / 0.0267 W/in C .091" .069" = 178.9 C/W (neglecting any heat spreading in the plastic)where L is the estimated thickness of the plastic above the die, k is the thermal conductivity of the plastic mold compound: 1.05 W/mK = .0267 W/in C, A is the estimated area of the die, which I guessed as the size of the ground paddle.Can you tell me if I need to adjust any of these assumptions to better reflect the actual internal details?
Thanks for your help."
While I won't pretend to understand your system or the thermal loads that may be proximal to this part, the primary path for heat flow of this device is from the backside of the die to the lead frame which has a paddle on the backside to maximize grounding. We have a recommended land that recommends maximizing the quantity of vias from this land to ground to improve the grounding and thermal path (link attached). Just to clarify, our thermal resistance of 17.6°C/W is "junction to case" NOT "junction to ambient". With all due respect, I'm not sure what a "CCA thermal model" is or how this model arrives at 106°C board temperature with a 43°C chassis temperature." It seems that there may be components nearby (*PA's or ?) that are generating significant heat, a very thick board made of a material that may not have the best thermal properties (FR4) or other factors contributing the elevated board temperatures. I recommend that you contact the factory directly if further assistance is needed.
Retrieving data ...