• State TI Thinks Resolved
  • Locked Locked
  • Replies 5 replies
  • Answers 2 answers
  • Subscribers 63 subscribers
  • Views 118 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

LM25148-Q1: Output ripple waveform question

Victor Levandovsky
Prodigy 20 points
Part Number: LM25148-Q1
Other Parts Discussed in Thread: LM25148

Hello

I design LM25148 (LM25148QRGYRQ1) based DC/DC

Vin = 9-30V
Vout = 5V (6, 7.2 - options)
Iout(max) = 10A

Fsw=500 kHz, FPWM mode
Primary, SSC On, Dual Phase Enabled (SYNCOUT Enabled)
External Compensation

Components was calculated using LM5149-LM25149 Quickstart Calculator_rev3.xlsm
Only one thing I missed. I mistaken with ext. compensation values. Recommended values for ~ 80 uF total Cout (soldered) are (6.34K+5n6)||68pF
(my values on schematic for 100uF, not 80uF). But I think it`s not critical for this issue.

See schematic bellow.

Now I testing my design in lab. All work fine (Vout stability, efficiency) except one puzzle with Vout ripple waveform.

See screen dumps from my oscilloscope bellow, please.

PIC_1. ~60 us (16 kHz) ripple

PIC_2. Zoomed x25 (500 kHz, Fsw)

My main question about "16kHz" ripple from pic.1

This ripple NOT changes with Vin (from 12 to 30V), Iout (from no load to 8A). Added 1000uF capacitance on Vout, Vin (after input LC- filter) is not change this ripple also.
Replacing laboratory power supply with battery, replacing DC electronic load with resistor and replacing probes in not help also.
Also I think this is not sub-harmonic oscillation (D always less then 50%)

In realty waveforms from screen dumps "breathe" and not stationary, modulation observed.

I suppose that "16kHz" ripple is a result of DRSS. Am I correct?
( also see red arrow on PIC_2. I think it`s phase change of LF part of DRSS)

Could anyone comment all described above, please?

-- Best Regards,

Victor

P.S.

Test setup

1. Laboratory power supply, battery

2. LeCroy DDA-125 (aka LC-684)

3. LeCroy AP020 active probes.

4. -3dB Custom probe: 50R ---> RG174---> DC block --> DDA125 (50 Ohm input)

5. TTi LD300 (DC Electronic load)

  • 0
    TI__Guru 50120 points

    Hello,

    Yes the dual random spread spectrum feature is modulating the change of frequency and would naturally introduce a low kHz tone onto the output voltage. 

    If you connect a 29.9kOhm resistor on the CONFIG pin, the undulation of Vout should go away leaving only the sinusoidal output voltage ripple which is a result of the inductor current across the low ESR MLCC capacitor.

    Regards,

    Jimmy

  • 0 in reply to Jimmy Hua
    Prodigy 20 points

    Hello Jimmy,

    Unfortunately your advice is not helped. I placed 51K1 in parallel with 71K5 already soldered on my board.
    Resulting resistance measured as 29.77 kOhm (fit into 1% tolerance for recommended 29.9 kOhm)

    Oscilloscope waveforms are different comparing DRSS mode but I not understand what is a source of LF part (now is 100 kHz) of output voltage ripple.
    Waveforms the same for no load, 2A and 8A, 100kHz ripple. But different for different input voltage.
    (Also I checked my power supply, no 100kHz ripple)

    See screens bellow.

    1. Vin=11.1V

    2. Vin = 14.8V

    3. Vin = 22.3V

    4. Vin = 29.5V

    Could you comment please this results?
    Worst case ripple is relatively low (15 mVrms), but I can not understand source of this effect.
    Is it my designs (schematic or PCB) problem or LM25148 normal behavior?

    --

    Kind regards,
    Victor

  • 0 in reply to Victor Levandovsky
    TI__Guru 50120 points

    Hello Victor,

    How are you measuring this output ripple? Right across the MLCC output capacitor and making sure the loop does not pick up noise from probing method?

    Also can you provide me the filled out excel quickstart calculator tool you mentioned? I'd like to use it for review of the design.

    Also can you provide the inductor current and SW voltage waveform as well? If what you are showing me is Vout ripple, then i suspect both inductor current and switching waveform to look unstable. 

    Regards,

    Jimmy

  • 0 in reply to Jimmy Hua
    Prodigy 20 points

    Hello Jimmy

    Sorry... Yes, waveforms in my previous message are ripples.

    Thank you. I sent you PM.

    -- Regards,

    Victor

  • 0 in reply to Victor Levandovsky
    TI__Guru 50120 points

    Hi Victor,

    Please contact me at j-hua@ti.com and let's continue the conversation from there.

    Regards,

    Jimmy