• State Resolved
  • Locked Locked
  • Replies 5 replies
  • Subscribers 65 subscribers
  • Views 1936 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.

LM22672 output oscillations during load

Christian Engstrand
Prodigy 90 points
Other Parts Discussed in Thread: LM22672

I've just received our prototype boards with the LM22672 5 V DC/DC.
Problem is that the output voltage has oscillations when the load is increased. These oscillations are in the same frequency range as our RFID reading (125 kHz) so it causes major reduction in tag reading distance.

Attached below is schematic of the DC/DC and oscilloscope screenshots for different load conditions. Channel 1 is at 5V node, channel 2 is SW node. Electronic load used for loading.


Do you have recommendations on how to avoid these oscillations?

~50 mA load:

~150 mA load:

~250 mA load:

~350 mA load:

~450 mA load:

  • 0
    Prodigy 90 points
    Replaced C12 (100 µF electrolytic) with 2 x 10 µF ceramic and it now behaves perfectly stable as far as I can see (0-800 mA load). Only a little ripple on the output at every switch, but no oscillations!

    But, datasheet says:
    "Typically, a total value of 100 µF or greater is recommended for output capacitance."
    "If the switching frequency is set higher than 500 kHz, the capacitance value may not be reduced proportionally
    due to stability requirements."

    I'm a little too stressed for the moment to dig deeper into stability analysis. Anyone has a recommendation on how to proceed? 20 µF ceramic should be good or are there any further risks?
  • 0 in reply to Christian Engstrand
    TI__Genius 16025 points
    Hi Christian,

    Can we see your layout? Ceramic capacitors are usually used in small values such as 0.1uF and 1uF to provide high frequency bypass. Try adding a .1uF and a 4.7uF along with the 100uF electrolytic and let us know if your performance improves. The ceramics should keep unwanted parasitics from affecting the stability of the IC.

    Thanks,
    Anston
  • 0 in reply to Anston Lobo
    Prodigy 90 points

    Hi Anston,

    Thanks for your reply! C13, 10µF, was already in parallell with the electrolytic, however not placed directly next to it. And perhaps it was too large.

    I'll try to test with a smaller ceramic placed close to C12 today!

    DC/DC part of the layout below:

    Comments on the layout are welcome as I'll do some small updates soon anyway :)

  • 0 in reply to Anston Lobo
    Prodigy 90 points
    Now I tested with .1µF and 4.7µF in parallel with C12 (place right next to it). Oscillations might have slighly lower amplitude but are still present.

    So currently seems performance is best with 2x10µF ceramic instead of C12. Is there any risks with this I'm not seeing?
  • 0 in reply to Christian Engstrand
    TI__Genius 15810 points
    Hello Christian,
    The 22 uF for the 800 kHz operation should be fine. To make sure everything is fine, I would do a load transient test of 80mA to 800mA and check what happens to VOUT.

    Stability analysis of the LM22672 is a bit harder than the other voltage regulators because of its voltage mode architecture. The higher output cap in your previous design would have caused a lower L-C double pole frequency and introduced instabilities which manifested as oscillations. You remedied that by reducing COUT and pushing that pole at higher frequencies. A load transient test should show a stable VOUT without oscillation.

    Regards,
    Akshay