About the TLV3801 in the WSON (DSG0008B) package,
Could you tell me if there is a thermal problem with the thermal pad being connected to the VEE but not to the inner layer?
This thread has been locked.
If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.
These packages were originally designed for switching power supplies; for logic devices, the pad usually does not need to be connected.
Comparing RqJB and RqJA, and assuming an infinitely large board, the quiescent supply current will result in a temperature increase of 20 mA × 5 V × 36.2 °C/W = 3.62 °C (with complete pad connection) vs. 20 mA × 5 V × 69.4 °C/W = 6.94 °C (without connection) above ambient. I have not been able to find information about the power dissipation capacitance, but this appears to be far away from having thermal problems.
We do not recommend floating the thermal pad.
The pad is not bonded-out specifically to any pin, but it is in contact with the bottom of the die through conductive mount compound - so it is not really "floating".
Because of this, the thermal pad should be tied to the most negative voltage (usually directly to VEE) to minimize any possible leakage paths.
Also floating the pad can allow noise into, or out of, the device (the internal junctions can capacitively couple to the bottom "plate" - think of a bottom ground plane of a PCB.)
As Clemens stated, there are not any real thermal problems - just a few degrees of self heating. The power dissipation may rise slightly at high toggle rates (LVDS output supply current is fairly constant).