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.

TPS563209: TPS563209 additional Cout 100uF after bead

Part Number: TPS563209

We are planning to use TPS563209 from 12V to 0.95V with a WEBENCH designed output filter of 1.5uH and 3x22uF ceramic capcitors. This will power two branches each starting with a bead ( Murata BLM21PG221, 0.045 DC ohm, 220 ohm at 100MHz) and then a 100uF ceramic capacitor to GND. This is typical requirement for Arria 10 FPGA tranceiver power. One branch is 900mA, and the other is 400mA.This violate the 22-68uF recommendation in the datasheet. But since there is a bead filter between this and the next 100uF I would think this still might be stable? Any advice? Should we reduce Cout from 66uF to 22uF?

  • I am still hoping to get som feedback on this one. I may add that my conclusion so far is that I assume that a bead, acting as a coil ( approx 1uH) below about 25Mhz, will introduce enough filter between the Cout of 66uF and the additional load of 100uF to avoid any stability problems. Comments on this? Any way this can be simulated using tools?

  • Hi Svein,

    Sorry for the late response. For my opinion, there could be some concern for the stability. I used to see cases where big output cap( like 1000uF) is placed at output. The resonant double pole is placed at a very low frequency( several kHz). There is an internal zero for DCAP2 parts which is located at around 10k-20kHz. Usually this zero will boost the phase up right before crossover frequency and system will be stable with enough margin. However, if the double pole is at low frequency, the crossover frequency is too low and the zero is at higher frequency. There is no phase boost and the phase margin could be small. Since you placed 100uF at output, It may not have a worse margin like this but the effect will be the same. 

    My suggestion for you is to do a load transient test with a fast load changing slew rate. If you did not see too much overshoot and no ringing happened during the transient The system is stable.