• State TI Thinks Resolved
  • Locked Locked
  • Replies 14 replies
  • Answers 2 answers
  • Subscribers 65 subscribers
  • Views 785 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.

UCC28C42: UCC28C42

AKASH JAIN
Intellectual 620 points
Part Number: UCC28C42
Other Parts Discussed in Thread: UCC28630

Hello 

We have developed one AC/DC and DC/DC card using TI ICs .

Can you please review it once.

Regards

Akash Jain

  • 0
    Intellectual 930 points
    Hello,

    While reviewing, Please note that output voltage is set to 12.2V.

    Regards

    Vishal Kakade
  • 0 in reply to Vishal Kakade
    TI__Mastermind 31580 points
    Hi Vishal,

    We will review the schematics and reply early next week.

    Regards

    Peter
  • 0 in reply to Peter Meaney
    TI__Mastermind 31580 points
    Hi Vishal

    AC/DC Design

    No Input fuse?
    RT1 & 2 could be combined into one RT device
    Replace C3 to C6 with one 68uF cap? Webench selects parts from a limited database and sometimes picks multiple parts where one device is a more effective solution but that device may not be in the Webench database.
    C7, is this a film cap, film caps are more robust than high voltage ceramics
    Q2, R8 and D6 circuit can be removed as they add little value for 12V design
    C26 should be in parallel with series circuit for R22 and C27
    R11 to 10k

    DC/DC Design
    FS1 3A rating too low, at 12V 60W output input current will be greater than 5A, consider 8A fuse?
    Vin min is 12V therefore UCC28C42 will not startup as turn-on is 14.5V, consider the 43 or 40 variant
    R25 and R28 need to be reduced in value as at 12V input it could take a long time for C44 to charge to the Vcc UVLO turn-on value
    Please check T2 primary inductance and turns ratio value - what equations were used ?
    R30, Q3 and D13 circuit can be removed for 12V output, C50 needs to be in parallel with R43 and C51
    R34 to 10K
  • 0 in reply to Peter Meaney
    Intellectual 930 points

    Hello Peter,

    Please note my comments below.

    AC_DC

    RT1 & 2 could be combined into one RT device: We need L/N reversal protection so RT1 and RT2 are separate

    .
    Replace C3 to C6 with one 68uF cap? Webench selects parts from a limited database and sometimes picks multiple parts where one device is a more effective solution but that device may not be in the Webench database.: We will do it.


    C7, is this a film cap, film caps are more robust than high voltage ceramics: Its ceramic cap


    Q2, R8 and D6 circuit can be removed as they add little value for 12V design: We will see. 
    C26 should be in parallel with series circuit for R22 and C27: Yes as per datasheet it should be in parallel. but as per webench design, it is parallel only with the R22. Please clarify on this point.
    R11 to 10k

    DC-DC

    FS1 3A rating too low, at 12V 60W output input current will be greater than 5A, consider 8A fuse?: yes , we will update it.


    Vin min is 12V therefore UCC28C42 will not startup as turn-on is 14.5V, consider the 43 or 40 variant
    R25 and R28 need to be reduced in value as at 12V input it could take a long time for C44 to charge to the Vcc UVLO turn-on value: My input is not 12V, its ranges from 44 to 55.5V and output is 12V.

    R30, Q3 and D13 circuit can be removed for 12V output, C50 needs to be in parallel with R43 and C51
    R34 to 10K: the Same query as AC_DC, please clarify

    Please check T2 primary inductance and turns ratio value - what equations were used ?: Primary inductance is 63uH.

    I have updated my design on webench as per my calculations and shared a public link for same. Please go through it.

    https://webench.ti.com/appinfo/webench/scripts/SDP.cgi?ID=9314612C8D52DBDB

    https://webench.ti.com/appinfo/webench/scripts/SDP.cgi?ID=398595656EA35102

    Thanks and Regards

    Vishal Kakade

  • 0 in reply to Vishal Kakade
    TI__Mastermind 31580 points
    Hi Vishal,

    AC_DC

    RT1 & 2 could be combined into one RT device: We need L/N reversal protection so RT1 and RT2 are separate - Separating the devices will not make any difference, one RT device will protect from inrush irrespective of L/N orientation.

    C7, is this a film cap, film caps are more robust than high voltage ceramics: Its ceramic cap - ceramic caps are sensitive to mechanical stress and can fail short circuit, please following manufacturers guidelines.

    C26 should be in parallel with series circuit for R22 and C27: Yes as per datasheet it should be in parallel. but as per webench design, it is parallel only with the R22. Please clarify on this point. If C26 is much smaller in value than C27 then this arrangement is ok, the high frequency roll off is determined by C26 + C27, so C26 will dominate if it is much smaller value. If C26 is close in value to C27 then the high frequency roll of characteristic will be at a lower frequency.

    R11 to 10k - 10k pull down is good design practice to ensure the mosfet remains off during start-up while Vcc is at a low voltage.

    Both Webench designs look like good to prototype.

    Regards

    Peter
  • 0 in reply to Peter Meaney
    Intellectual 930 points

    Hello Peter,

    I have designed external circuits which give me input UVLO/OVLO alarm signal at 3.3V and 0V.

    My UVLO alarms is a transition from 3.3V to 0V, OVLO alarms is a transition from 0V to 3.3V.

    I want to use these alarm inputs to disable UCC28C42.

    I am connecting UVLO alarm to CS pin through an optocoupler. and UVLO alarm to COMP pin through an optocoupler. in order to disable the controller at Lower and Upper threshold.

    Please go through my design and let me know for any other alternative or reviews.

    Regards

    Vishal Kakade

  • 0 in reply to Vishal Kakade
    TI__Mastermind 31580 points
    Hi Vishal,

    I responded to your other posts on this topic.

    Regards

    Peter
  • 0 in reply to Peter Meaney
    Intellectual 930 points

    Hello Petrer,

    Yeah, I know but you suggested a different part for implementation. In the datasheet of Ucc28C42(page no: 19) it is mentioned that UCC28c42 can be disabled by pulling comp pin to low or CS to high. Using these I have designed my circuit. Can you verify it?

    I am not able to simulate the complete design in TINA also. I am attaching the circuit if you can make it work in tina then I can do the rest.

    DC_DC.TSC

  • 0 in reply to Peter Meaney
    Intellectual 620 points

    Hello Peter

    Thanks for all the review points.

    Can you please also check if need to add anything more the EMC/EMI also?

    Regards

    Akash Jain

  • 0 in reply to AKASH JAIN
    TI__Mastermind 31580 points
    Hi Akash Jain,

    I suggest you use the EMI filter design used in the UCC28630 EVM, www.ti.com/.../ucc28630evm-572

    This will act a s a starting point for your design.

    Regards

    Peter
  • 0 in reply to Peter Meaney
    Intellectual 620 points
    Hello Peter,
    Ok we will use that and any thing for the surge and ESD and voltage Dip test?

    Regards
    Akash Jain
  • 0 in reply to Peter Meaney
    Intellectual 930 points

    Hello peter,

    Currently, I am using EMI filter from "PMP40025 " Reference design.  UCC28630 also looks good. Can you suggest an EMI filter/Surge protection for our DC-DC power supply design?

    Regards

    Vishal

  • 0 in reply to Vishal Kakade
    TI__Mastermind 31580 points

    Hi Vishal,

    There is a Varistor V1 fitted to the UCC28630.

    This will help manage line surge, pleas note that EVM's may not have been submitted for AC line surge testing.

    Regards

    Peter

  • 0 in reply to Peter Meaney
    TI__Mastermind 31580 points
    Hi Vishal

    Sorry I missed your question on the DC/DC design. I would leave space for PI input filter, C-L-C, as well as a common mode choke. It is possible to make some calculation on the pulse currents drawn from the source and the level of differential mode filtering required but its almost impossible to make a meaningful calculation for the common mode noise.

    Regards

    Peter