• State Resolved
  • Locked Locked
  • Replies 3 replies
  • Subscribers 63 subscribers
  • Views 660 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.

Improved circuit for LM3485

Mahendra Patel1
Expert 3560 points
Other Parts Discussed in Thread: LM3485, LM5085

Hello team,

For LM3485, to get a stable switching frequency, we have referred two different docs:

 1. http://powerelectronics.com/site-files/powerelectronics.com/files/archive/powerelectronics.com/power_management/power_ics/803PET22.pdf

 figure 3.

2. LM5085 datasheet configuration in block diagram on page 9.

Kindly let me know, which configuration is stable or better for LM3485.

 Thank you.

 Regards,

Mahendra Patel,

Analog Applications Engineer

  • 0
    TI__Intellectual 2320 points

    Hello Mahendra,

    Stability-wise both the methods have the same impact but there purposes are different.

    While relying solely on ESR of the output capacitor for ripple generation, there will be phase lag between the inductor current and the ouptut capacitor voltage, leading to unstable operation.One has to use either an external ESR in series with the output capacitor or a ripple injection circuit as shown in LM5085 DS.

    When you use an external ESR, there will be an incerased output voltage ripple. In order to have  readuce the impact, one uses a feed forward cap like Cff. In other words, when the ripple at the output node matches the required hysteretic comp ripple and  is ac coupled to the FB pin through the Cff cap, a customer can use a smaller ESR in series with the output capacitor.

    However, while using a ripple injection circuit one does not need a Cff, as the ripple is ac coupled straight at the FB node with little or no addtional output ripple at the output. The only drawback would be the increased component count.

    Regards,

    Sourav

     

  • 0 in reply to Sourav.Sen
    Prodigy 10 points

    Hi Sourav,

    Can you tell the difference between ripple injection method given in the block diagram in page No.9 of LM5085 and Fig.3 of the document http://powerelectronics.com/site-files/powerelectronics.com/files/archive/powerelectronics.com/power_management/power_ics/803PET22.pdf

    Both methods are of emulated ripple injection. Which method can I use for LM3485 Hysteretic buck controller. I need stable operating frequency . I mean operating frequency has to stable with respect to component variation from board to board because of tolerances. Operating frequency should be load dependent.

    Regards,

    Dhananjai

  • 0 in reply to DHANANJAI CHINTALA
    TI__Intellectual 2320 points

    Hello Dhananjai,

    Please read my previous post. I have tried explaining there the difference between the two circuits.

    Hysteretic controllers are not the best examples for for stable frequency operation over input voltage. COTs like LM5085 (with one shot on-timers) are better at maintaing the switchinig frequency over the input voltage operation.

    However, with respect the injection circuit components, there is not much difference betwenn the two topologies. As far as the emulated ripple generation circuits,  the ripple injection circuit tends  to have (qualitative) a more stable operation (over the loads or input voltage range) as the ripple is coupled right at the FB pin/node unlike the ESR and Cff circuit where a bad, long trace between the FB pin and the FB node can make the FB node susceptible to noise  (and the SW node jittery).

    However, this is a very subjective suggestion, as in most cases you will not be able to differentiate between the two cases in case of SW node frequency stability  as long as you are injecting enough ripple at the FB node. If you are using LM3485, you will still need an external ESR in series with the Cout cap to inject enough ripple at the FB node.

    Regards,

    Sourav