• State Resolved
  • Locked Locked
  • Replies 3 replies
  • Answers 1 answer
  • Subscribers 64 subscribers
  • Views 248 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.

TPS50601A-SP: Synch two devices, but separate Enable

Mark Guastaferro
Genius 11645 points
Part Number: TPS50601A-SP


Customer has a design with two TPS50601A-SP devices. The clocks are synchronized using the Master/Slave approach at 500 KHz, but each device outputs a different voltage. Can the devices be Enabled independently? 

I believe this is OK.  In the case where the Master is enabled, but the lave is not,  the Master will still send out an inverted clock on the SYNC pin.  The slave would have power (VIN/PVIN) but will not power up.  In the case where the master is disabled,  then the slave would operate with it's internal clock based on the RT resistor, which is set for 500 kHz.  If the master is enabled with the slave already Enabled,  the clock of the slave will eventually switch over to the inverted Master clock. 

Any concerns we are overlooking?

  • 0
    TI__Mastermind 28915 points

    Hey Mark,

    I just want to confirm that the COMP/VSENSE pins are not connected between the devices in order to make sure they don't interfere with each others operation.

    As long as each device is only connected at the SYNC connection there should be no issue with enabling them separately.
    They will be out of phase (180 degrees) with each other which may or may not matter for your application.

    Thanks,
    Daniel

  • 0 in reply to Daniel Hartung
    TI__Genius 11645 points

    Yes, that is correct. 

    Another related question:

    In the case where they are both enabled and the slave is operating from the master's clock output Then the master is disabled.  The slave will switch to its internal clock, but this takes 20-25 usec. In that switchover time, there will be no switching on the slave, is this correct? At 500 kHz, that is about 10-13 clock cycles with no switching and the output will sag or overshoot depending on the state of the FETs and the load.  

  • +1 in reply to Mark Guastaferro
    TI__Mastermind 28915 points

    Hey Mark,

    The device would stop switching for that time as not clock is detected.
    This should for the most part always cause the output to dip, and never go up as there would be no energy transfer without the switching.

    Thanks,
    Daniel