• State
  • Locked Locked
  • Replies 3 replies
  • Subscribers 64 subscribers
  • Views 1400 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.

TPS54320: Non-Monotonic Startup

Casey Weltzin93
Prodigy 30 points
Part Number: TPS54320

I am using the TPS54320 for several supplies. Some show a non-monotonic startup, and I'm not certain what is different about the affected supplies.

Some details on one of the supplies:

Vin=12V, Vout = 1.9V

Fsw = 750kHz

Tss = 2ms

Lout=6.8uH, Cout=66uF, Imax=1.25A

Rcomp=1.87k, Ccomp=22nF

Measured bode plot looks good (Fc = 40kHz, PM = 70 Deg), as does the transient response. All indications are that the supply is well compensated.

Can anyone provide some troubleshooting suggestions here? The startup waveform is non-monotonic with any load, from about 0V to 800mV. The non-monotonicity looks like a damped sine wave riding on top of the desired ramp -- the frequency of oscillation is 16kHz.

I have probed the soft-start pin, enable pin, frequency set pin, and Vin and none show any oscillation.

  • 0
    Prodigy 30 points
    Update: I suspect this is due to the minimum on time behavior of the IC. In my case a minimum on time of ~100ns with a 750kHz switching frequency and 12Vin translates to about 900mV. This is very close to the point in the ramp-up waveform where the oscillation stops. So I think the IC is overshooting due to minimum on time, then pulse skipping, and repeating. Probing the switch node confirms this. Any advice on mitigation is still appreciated.
  • 0
    Prodigy 30 points
    This issue is solved, but I will post for future reference. I found that the minimum on-time behavior of the TPS54320 can cause non-monotonic startup if the soft-start time is too slow. Basically, before the reference has ramped up to the voltage dictated by the minimum on-time, the output is effectively open loop and the LC output filter will resonate. You want that resonance to be slower than the soft-start time.
  • 0 in reply to Casey Weltzin93
    TI__Mastermind 35185 points

    Hi Casey,

    Thanks for the updates. I just released an app note last week on this exact topic. Link is below. I however did not look too closely at how different soft-start times affects the startup ramp. Faster soft-start times can definitely help though.

    Best Regards,
    Anthony