• State TI Thinks Resolved
  • Locked Locked
  • Replies 4 replies
  • Answers 1 answer
  • Subscribers 64 subscribers
  • Views 141 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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: abnormal operation

David Park
Expert 6810 points
Part Number: TPS544C20


Tool/software:

Hi,

There is abnormal operation intermittently.

Abnormal operation means no switching and no output.

It's occurred in 3 samples of 600 samples,

This symptom is duplicated at once of 10 times trying.

I checked BP3 and BP6. it shows 3.2V, 6.5V at normal and abnormal case.

Could you guide what I have to check as next step?

I attached the schematic.

TPS544C20 abnormal output.pdf

  • 0
    TI__Genius 12035 points

    Hello,

    A response will be posted this week

    Thanks,

    Calan

  • 0
    TI__Guru 56985 points

     

    The default fault response for the TPS544C20 is shutdown and do not restart (Latch-Off).  An oscilloscope waveform of the start-up attempt would help debug the issue.

    There is a lot of output capacitance  2x 560μF + 4x 47μF + 8x 47μF + 4x 10μF = 1,724μF  which is generating an LC resonance of 4.5kHz

    The loop is generating an output impedance of 37mΩ, which is creating a loop crossover frequency of 2.5kHz.  I suspect that the loop is crossing near or below the LC resonance and might not be stable.

    Can we reduce the output capacitance?

    Can we increase the switching frequency?

    If we can't reduce the output capacitance, can we update the COMP (pin 35) to AGND from a single capacitor to a R-C + C network so we can add some gain to the loop with an Rcomp resistor or a Feed-forward capacitor across R10?

  • 0
    Prodigy 10 points

    Switching Frequency

    Test Results

    400KHz

    Failed all 10 times

    500KHz

    Failed all 10 times

    650KHz

    Failed all 10 times

    Switching Frequency

    Output capacitance

    Test Result

    250KHz

    1724 μF -> 604μF

    All tests passed  980 times

    Q1) How is the DC/DC loop crossover frequency verified?

    Q2) What are the recommendations regarding the difference between the loop crossover frequency and the LC resonance frequency?

    Q3) What were the reasons the circuit did not operate when the switching frequency was increased?

  • 0 in reply to chae heeyong
    TI__Guru 56985 points
    chae heeyong said:
    Q1) How is the DC/DC loop crossover frequency verified?

    The D-CAP2 control loop creates a forward transconductance from FB to IL equal to:

    (tau / L) x (1+ Rcomp x gM)

    tau is adjusted based on switching frequency in table 11 on page 55 of the datasheet (https://www.ti.com/lit/ds/symlink/tps544c20.pdf#page=55) 

    250kHz : 104μs

    400kHz: 87μs

    500kHz: 76μs 

    650kHz: 60μs 

    gM: 130μS 

    The DIFFO to FB divider is reducing the sensed VOUT by VOUT / VREF (0.6V) or 5.5 for Vout = 3.3V so we are getting a Vsense to IL trans conductance with switching frequency of:

    250kHz 148.6 / 5.5 = 27.0S (1/37mΩ)

    400kHz 124/3 / 5.5 = 22.6S (1/44mΩ)

    500kHz 108.6 / 5.5 = 19.7S (1/50mΩ)

    650kHz 85.7 / 5.5 = 15.6 (1/64mΩ) 

    The Crossover frequency (Unity Gain Frequency) is where the product of the Output Capacitor Impedance (Zout) and the Vsense to IL forward transconductance (GM) is 1.  Or Zout(f) = 1 / GM

    chae heeyong said:
    Q2) What are the recommendations regarding the difference between the loop crossover frequency and the LC resonance frequency?

    We typically recommend the Unity Gain frequency is at least 2x the LC resonance frequency and greater than 10kHz, but no more than 1/4 of the switching frequency.

    chae heeyong said:
    Q3) What were the reasons the circuit did not operate when the switching frequency was increased?

    The reduction of the internal ramp generator time-constant (tau) without a corresponding reduction in the inductor value (L) causes the forward transconductance to increase with increased switching frequency.  The higher forward transconductance resulted in a lower unity gain bandwidth and reduced stability.