• State TI Thinks Resolved
  • Locked Locked
  • Replies 7 replies
  • Answers 1 answer
  • Subscribers 64 subscribers
  • Views 358 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.

LMZM33604: Syncing 2x LMZM33604 modules

Saxon Liw
Prodigy 80 points
Part Number: LMZM33604
Other Parts Discussed in Thread: LMZ23610

Hi,

Is it possible to sync 2x LMZM33604 modules with one as a master and another as a slave using the SW pin as the output and the SYNC/MODE pin as the input? Datasheet says not to connect any components on the SW pins, but could this be a possible  solution if I use a resistor-zener combination with a high enough resistance?

If this is possible, what should be the possible range of resistance I can use in series with a 4.5V Zener (eg not too low as to interfere with the SW pin and not too high to be loaded by the SYNC/MODE pin)? I am not able to find information on the input impedance of the SYNC/MODE pin but measuring with a DMM gives me around 600kΩ at steady state.

Thank you.

Best regards,
Saxon

  • 0
    TI__Mastermind 34875 points

    Hi Saxon,

    The master module must works in FPWM mode, the sync clock can be generated from SW node waveform.

    What's the input voltage range and output voltage in your application?  what's your target switching freq?

    B R

    Andy

  • 0
    TI__Guru 50120 points

    Hi Saxon,

    Theoretically this can be done. However a scenario that might be problematic is when you are pulling large output current and the part is not evenly distributing the current load. In this case, one of the two power modules may hit current limit. Because of this, it is recommended to use a power module that is specifically designed for current sharing such as the LMZ23610.

    The LMZ23610 has a wide input voltage range and is equipped with a SH(share) pin to enable multiple LMZ23610 modules to share the load following the master.

    Regards,

    Jimmy 

  • 0 in reply to Jimmy Hua
    Prodigy 80 points

    Hi Jimmy,

    I am using them to drive simple resistive loads that are identical, therefore I would like to run them out of phase to see if I can reduce the switching noise at the input side. The choice for LMZM33604 is because of the integrated inductor that makes the design simple with as little part count as possible.

    Thank you.

    Best regards,
    Saxon

  • 0 in reply to Saxon Liw
    TI__Guru 50120 points

    Hi Saxon,

    I'd like to also provide a reference on this application (http://www.ti.com/lit/an/slyt748/slyt748.pdf). 

    Note that there may be some interleaving issue since no two power modules have the exact same internal clock frequency. "The switching frequency of each power module can be different when the temperature changes, and if the master and slave clock frequencies are significantly different, forcing them to synchronize may cause PWM pulse stretching resulting in damage to the part." 

    Again I would not recommend this, but you can always take two EVM and test this condition for your design verification. It can work, however this has not been tested on the device. 

    Regards,
    Jimmy 

  • 0 in reply to Andy Chen
    Prodigy 80 points

    Hi Andy,

    The input is a wide range 9-36V with a 7V 3.6A fixed output. I have not optimized the switching frequency yet but from Table 6 in the datasheet, it looks like I do have a wide range of possible frequencies that is available up to at least around 1.5MHz. Currently I have only tested a single module with fsw=500kHz on the EVM.

    Best regards,
    Saxon

  • 0 in reply to Jimmy Hua
    Prodigy 80 points

    Hi Jimmy,

    Thanks for the article. I had a quick look but I am thinking more of running my regulators with differential phase rather than interleaving them. This is more to reduce the EMI at the input than to improve ripple at the output. My loads are independent and therefore the outputs from each of the module would see each other but the loads they see are essentially identical. In any case, I will take a deeper look into the article which might give me a better insight.

    I will put the EVMs together once my diodes are in and scope them up to see if they function as intended.

    Best regards,

    Saxon

  • 0 in reply to Saxon Liw
    TI__Guru 50120 points

    Hi Saxon,

    Thanks for the update. Please keep me informed on your testing. 

    If you have any further questions, please let me know. 

    Regards,

    Jimmy