• State TI Thinks Resolved
  • Locked Locked
  • Replies 3 replies
  • Answers 2 answers
  • Subscribers 63 subscribers
  • Views 128 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.

UCC12050: Parallel Setup

Mark Philip Felipe
Genius 9705 points
Part Number: UCC12050

Hi Team, need your inputs on this concern.

Need to parallel 3 of these devices for the power level required. Initially attempted using or'ing diodes on the output of each DC-DC, but have found we have devices failing in the testing phase. Have seen the sync option, but struggled to find any documentation on how to properly use this. Please confirm that failure of the devices is expected if not using the sync option, and provide some guidance on the proper way to parallel 3 devices using the sync option. Thanks.

-Mark

  • 0
    TI__Mastermind 21765 points

    Mark,

    With or without using the SYNC function, you cannot parallel the outputs of two or more UCC12050. They do not share current. Is it possible to partition your load requirement in a way that each block is <500mW? Regarding the SYNC function, this feature is for EMI but also has some interesting details to be aware of. When external SYNC is applied, internal spread spectrum modulation (SSM) is disabled. The SYNC signal must be 16 MHz±10%which allows some wiggle room to synchronize to an external clock and move the fundamental to away from 8 MHz, ±10%. If the external clock edge happens to fall outside the limits of 16 MHz±10%, The UCC12050 will default back to it's internal clock frequency of 8 MHz and the internal SSM will be active. Also, since SSM is disabled when a valid SYNC signal is applied, the external clock should include some intelligent method of clock dithering in order to pass CISPR EMI standards. The dithering depth must be kept within the min/max limits. 

    Regards,

    Steve M

  • 0 in reply to Steven Mappus1
    TI__Genius 9705 points

    Hi Steve, just a follow-up.

    Will voltage droop control work, i.e. inserting a precision resistance at the output of each converter? As far as the sync option, I understand now that this is not intended for sharing purposes. Oddly enough though, in a test we did on a total of four boards (two having the 3 paralleled DC-DCs using our flawed or'ing diode method and two using our "flawed" sync option with 3 paralleled DC-DCs), and the two sync setup test worked fine for many hours while the other two boards both stopped working within a short time frame.

    Thank you.

    -Mark

  • 0 in reply to Mark Philip Felipe
    TI__Mastermind 21765 points

    Mark,

    Voltage droop control works when the output rectifier of each parallel converter is passive. Without giving away our rectification control technique, I can tell you that it is not pure passive and this is one of the reasons they do not work with droop control for load sharing. If you believe the outputs are sharing current correctly, have you measured the case temperature of each of the three DC/DC modules? Have you measured and compared the input current to each of the three DC/DC modules (If load share is done correctly, Iin1=Iin2=Iin3)? You may see parallel operation appear to be working under a specific steady state condition but when you consider dynamic loading, startup, line/load transients and the increase input voltage ripple resulting from parallel operation, TI's recommendation is not to parallel the UCC12050 outputs for load sharing. 

    Regards,

    Steve M