On the LM95235 datasheet it states the following with respect to decoupling capacitors:
"VDD should be bypassed with a 0.1 µF capacitor in parallel with 100 pF. The 100 pF capacitor should be placed as close as possible to the power supply pin. A bulk capacitance of approximately 10 µF needs to be in the near vicinity of the LM95235."
There is no other information about the type of the capacitors (e.g., MLCC, Tantalum) or their characteristics (e.g., ESL or ESR).
0.1 uF and 100 pF MLCC capacitors are readily available in 0402 package size. They have similar ESLs and therefore self-resonant frequencies (SRFs) that are approx. 30 times different. Connecting two such capacitors in parallel would lead to a large spike in power supply impedance between the SRFs of the two capacitors.
I have performed a SPICE simulation with 100 pF 0402 (ESLmounted=1n; ESR=100m-ohms), 100 nF 0402 (ESLmounted=1n; ESR=20m-ohms) and 10 uF 1206 (ESLmounted=2n; ESR=5m-ohms) MLCC capacitors in parallel. This leads to peaks of approx. 500 m-ohms @ 9MHz and 24 ohms at 390MHz. Removing the 100 pF capacitor reduces the peak at 390MHz to 1.6 ohms. I.e. the impedance of the supply is LOWER AT ALL FREQUENCIES up to 1GHz WITHOUT the 100pF capacitor.
Given the emphasis placed in the datasheet on the need for the 100pF decoupling capacitor I would like an explanation as to its function, or if (as I suspect) the section on decoupling is gibberish, it to be corrected.
