• State TI Thinks Resolved
  • Locked Locked
  • Replies 5 replies
  • Answers 1 answer
  • Subscribers 62 subscribers
  • Views 1327 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.

UCC256301: Is this part intended to replace UCC25600 - recommendation for new designs

Paul Hoffmann
Prodigy 110 points
Part Number: UCC256301
Other Parts Discussed in Thread: UCC25600, , TIDA-01494

There is a new series of LLC resonant controller ICs although the old one, the UCC25600, is still available.

Does the UCC25600 have some advantages over the new series. Which one is recommended for new designs?

Reference designs using the UCC25600 always use an additional gate driver IC (like UCC27714D, up to 4 A drive and sink current). None of the UCC25630X reference designs uses an additional gate drive. This is especially surprizing as the UCC25630X series have less gate drive and sink currents than the UCC25600 (0.6 to 0.8) and far less than an external gate driver.

Has the omission of the gate driver IC something to do with the automatic dead time feature of the UCC25630X? 

Is the lack of current and therefore less steep switching slopes compensated by longer dead times?

Would the automatic dead time feature work with an additonal gate driver?

Thank you for your effort.

Paul Hoffmann

  • 0
    TI__Mastermind 27145 points

    Hi Paul,

    Thanks for your interest in UCC256301. Here is a quick summary of UCC256301 features that are not included in UCC25600.

    1. Active X Cap Discharge Function

    2. Capacitive Region Avoidance

    3. Hybrid Hysteretic Control

    4. integrated High Side and Low Side gate driver (does not require gate drive transformer or external gate driver for low to medium power designs).

    UCC256301 is not intended to directly replace UCC25600 but given the improved feature set, TI does recommend using UCC25630x for new designs.

    The inclusion of the high side grate driver allows the UCC25630x controller to drive the high side mosfet directly. This is something UCC25600 is unable to support.

    The adaptive dead time feature relies primarily on the HS, Vcr sense and Isense signals to set the dead time. UCC256301 is compatible with external gate drivers if this is necessary for your design.

    Best Regards,

    Ben Lough

  • 0 in reply to Benjamin Lough
    Prodigy 110 points
    Dear Ben Lough,

    thank you for your fast reply.

    These are really nice features ...

    But I am still confused about gate drivers. You mention low and medium power designs (do you consider about 500 W of the reference designs, TIDA-01501 or TIDA-01494, medium power?). Even when playing around with the Webench Designer - no gate drivers even for designs >1kW.

    Can you please be more specific. Why does this device does not require a gate driver anymore. As far as I understood high current gates drivers are used to provide enought current during on- and off-phases of the switching process. If it is not available the steepness suffers seriously, high switching frequencies are not possible and, even worse, off- and onsets of the two MOSFETS may overlap.
    A solution would be to use very low gate charge MOSFETS. But with 600V MOSFETs gate charges are not really low.

    Therefore, what is the secret that UCC25630Xs does not require additional gate drivers while being only able to provide 0.6 A drive and sink current?

    Sorry to bother you.

    Thank you
    Paul Hoffmann
  • 0 in reply to Paul Hoffmann
    TI__Mastermind 27145 points

    Hi Paul,

    UCC25630x has an internal bootstrap circuit. When the low side mosfet is on, the switch node is connected to gnd through the low side fet. During this time, RVCC charges the bootstrap cap. When the low side switch is turned off, the high side gate driver uses the charge on the bootstrap cap to drive the high side fet. This is possible because the high side gate driver is referenced to the switch node. UCC25600 does not have a bootstrap circuit so in order to be able to drive the high side fet, an external gate drive transformer or external gate driver is necessary.

    You are correct in that a higher source/sink current allows for smaller turn-on/turn-off times. Whether an external gate driver is needed or not will depend on the total gate charge of the mosfets and the desired resonant frequency of the LLC design. The gate drive voltage of UCC25630x is 12V. From the fets you choose, there should be a graph in the datasheet showing Qg vs Vgs

    The graph will give you the total gate charge. The required gate drive current is dependent on how quickly you need the fet to turn on or turn off

    Ig = Qg/(T_transition)

    One item I should clarify is that the gate drivers integrated into UCC25630x is 0.6A source, 1.2A sink.This is specified on page 8 of the datasheet: http://www.ti.com/lit/ds/symlink/ucc256301.pdf

    Best Regards,

    Ben Lough

  • 0 in reply to Benjamin Lough
    Prodigy 110 points
    Thanks a lot for your answer.

    This sounds quiet reasonable (bootstrapping) although I cannot really weight it up numerically. Additionally a "realistic" T_transition of a design seems to be difficult to estimate. The faster the better would be my first approach - but this does not seem to be practical and may cause other problems. Additionally actual switching frequencies are not too hight with higher power PS.

    Currently I am designing a medium power PS using the UCC25600 and a gate driver. Although I already have a good working prototype I am thinking of a redesign using the new part, mainly to get rid of the additional housekeeping power supply. Not using a gate driver would be nice too. But the large differences in drive and sinks currents between older recommendations and the new reference designs makes me a little bit leery.

    However,
    thank you for your support.

    Paul Hoffmann
  • 0 in reply to Paul Hoffmann
    TI__Mastermind 27145 points

    Hi Paul,

    Page 2 of this application report provides some additional info on bootstrapping if you are interested: . The datasheet also provides a recommendation for the minimum bootstrap capacitance in section 9.2: .

    If you could share your deisgn spec I can make a recommendation if an external gate driver is necessary. 

    Best Regards,

    Ben Lough