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.

LM5002: LM5002 SEPIC with non-coupled inductors - Ripple /Switching Noise issue

Part Number: LM5002

Hi E2E forum!

We've used LM5002 in SEPIC configuration as per example below:

We have just changed the Feedback resistors to have 15V as output.
Our specification is Vin=14V to 20V, Vout=15V at 50mA

From the operation side the SEPIC is working quite good: sweeping from 14V to 20 the output is always stable at 15V. However something strange on the output is happening, it seems ripple but in fact seems to be more a switching noise issue.
At the beginning we was using as switching Diode (D2) a Vishay V8PAL45 but seems to have a high junction capacitance(1400pF)  so we changed with a SS16 (less than 100pF).

Changing diode and output capacitor (from electrolitic to ceramic) have improved the ripple as below image to 200mV.

Before:

After:

Zoom on the peak-pulse:

All the test has been performed with a 50mA load.

We're basically looking for suggestion on hot to improve the AC-Ripple/Switching noise at the output: We tried to add a RC snubber in parallel to D2 but unfortunately this have not helped.
Anyone has got similar issues?

Thanks in advance

Davide

5 Replies

  • I done some further investigation, and yes seems really to be more a Switching noise instead of AC Ripple:

    Measurement has been made with the tip-ring to shorten as soon as possible the ground path.

    Trace #1 (Yellow): Ouput , coupling in AC

    Trace#2 (Green): Pin N° 1 LM5002 (coupled in DC).

  • In reply to Davide Dal Fra:

    Hi David,

    I think the behavior is mainly caused by the probe setup as show in the image

    Best Regards

    Jasper Li
    Boost Converter Solution 

  • In reply to Jasper Li:

    Hi Jasper,

    Thankyou for your reply.
    However, the measurement was made correctly by using the tip-spring tool on the oscilloscope probe instead of using the normal gnd wire+clamp.

    Regards

    Davide

  • In reply to Davide Dal Fra:

    Hi Davide

    Is the waveform measured right on the two terminal of the output capacitor?
    the solutions would be add high frequency capacitor, such as 0402 package 0.1uF ceramic capacitor to the output.
    or add small LC filter to the output if the spike impacts the loading.

    Best Regards

    Jasper Li
    Boost Converter Solution 

  • In reply to Jasper Li:

    Hi Jasper,

    Yes exactly the waveform has been measured directly on the output capacitor.
    Adding a 0603 100nF capacitor (we have finished 0402) have reduced about 5/7% the spikes.


    We'll try add a smaller LC filter, and i'll let you know.

    Thanks in advance for your support

    Davide

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.