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.

TPS544C20: Inquiry about output abnormality

Part Number: TPS544C20

Hello,

My customer is using the TPS544C20 under the following conditions:

  - VIN=12V, VOUT=0.85V, IOUTmax=25A, FSW=1MHz(Rrt=196kohm), without PMBus

They have an issue that when about 10 seconds have elapsed after Power On, the output was blocked  as the ripple increases as shown below and the PG alarm occurred.

When they lowered the FSW to 500kHz (Rrt=39kohm), this output error did not occur.

Could you please explain what is the cause of this symptom?

Please suggest the best solution to solve the issue.

In Addition, please review the circuit diagram below to see iff there are any problems with the device design.

Thank you.

JH

  •  

    It looks like an instability in the output voltage loop triggered by a load transient causing the TPS544C20 to detect either OV or UV and shutdown.

    The TPS544C20 auto-scales the internal ramp time-constant used for ESR emulation, so checking the switching frequency would also change the time-constant, and thus the emulated ESR, so it can result in a chance in the loop stability.

    I see 8x 100μF capacitors, plus a remote sense line going off-page in the schematic.  Is there additional off-page schematic on the VDD_0V85 rail before the remote sense?  Could there also be parasitic trace inductance adding phase delay to the sensed output voltage?

    Unfortunately, the existing schematic does not appear to have locations for the components that would normally help.

    1) A resistor in series with COMP to increase the loop gain and improve stability.  If you can add a resistor in series with C19, start with about 20kΩ

    2) A capacitor from the output voltage close to the inductor to the differential feed-back capacitor (C26) to feed the dynamic output voltage back to the sense lines without the remote sense delay.  Your sense line impedance is 50Ω.  On the ringing, I count 5 cycles on about 1/4 of a μs division for about 25μ per cycle or 40kHz.

    40kHz on 50-Ω is about 80nF.  You can try a 100nF output feed-forward capacitor.