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.

About the countermeasure for undesired oscillation of TPS5430

Other Parts Discussed in Thread: TPS5430

Dear All,

Please let me talk about the countermeasures for undesired oscillation of TPS5430 on the customer board.

Our customer confirmed an undesired oscillation of Vout on their board.

We attach the document which we summarized board layout and the oscillation waveform of our customer in.

Our customer is going to change the input capacitance. (from ceramic to POSCAP)

With the data sheet, it recommend the using the ceramic capacitor as an input capacitor.

■Question1

Is not there the problem even if customer use POSCAP for input? (What kind of problem is thought about?)

■Question2

Do you think the cause of undesirable oscillation to be what?

(We think whether it is a cause that ESR of the parallel capacitance of the output side is too low. Is there the Min spec of the ESR? )

--------------------------------------------------------------------

■Problem and its features

①An undesired oscillation of Vout was confirmed with a constant current.

(Especially, undesired oscillation is remarkable with the high load current.)

②In this case of ①, undesired oscillation might not happen.

Under the situation, this problem occurred when a customer cooled these parallel capacitances of the output side with a cold wind.

③Although the low load current, this problem occurred when its part was cooled.

■Their countermeasure

Customer tried a change of these input capacitances. (value/kind)

These points are C1235 and C1236 on customer’s block diagram.

Before:Ceramic capasitance 4.7μF

 After:POSCAP 33μF

TPS5430_UndesiredOscillation2013-3-27.pdf
  • Yes, I think the root cause is the low ESR of the output capacitors.  The TPS5430 is designed for a POSCAP type for most output applications.   I am somewhat surprised that changing the input capacitance to POSCAP fixes the issue.  To fix the stability, you may want to consider some external circuitry around the feedback voltage divider to modify the compensation.  I believe there is an app note on the product page for that.

    I will say that for C1236, I would recommend some additional capacitance in parallel with the ceramic cap such as the POSCAP.  That is the main input capacitor for the IC.  I think the C1235 should be ok with just a ceramic. 

  • 4744.Oscilloscope captures of 1kHz large oscillation in Vout and noise on Vout and PN nodes.docx

    5582.n_11V_schematic.TSC - TINA.pdf

    I am using a similar circuit (TPS5430 in inverting buck-boost configuration with enable logic level shifted with BJT controlled voltage dividers).

    My application is Vin=15.7 V,  Vout= -11V  Iload= 40mA to 2A

    I have significant high frequency noise on the PH node and Vout (which is unexpected since I have substantial low ESR filter capacitors), but the really strange behavior is a 1V, 1kHz triangular oscillation in the output voltage when powering a 5W resistive load.

    This phenomenon is shown in the attached pdf.

    I also included a schematic of my power supply circuit and screen captures of the PH node voltage and Vout high frequency noise.

    All filter caps (330 uF and 10uF are aluminum electrolytic low ESR).

    The compensation network values were computed using the procedure outlined in TI  app note SLVA237C 

    (http://www.ti.com/lit/an/slva237c/slva237c.pdf).

    I have no idea why the large 1kHz oscillation in Vout occurs.  Any advice would be appreciated.

    Thank you.