• State Resolved
  • Locked Locked
  • Replies 18 replies
  • Answers 3 answers
  • Subscribers 63 subscribers
  • Views 682 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.

UCC28250: Input over voltage and under voltage protection in PMP8878

Aneesh TS
Expert 1735 points
Part Number: UCC28250
Other Parts Discussed in Thread: PMP8878, UCC25230, , PMP8877, UCD3138

Hello sir,

We are planning to develop a Brick module taking PMP8878 as a reference. But we have the requirement of Input over voltage and under voltage protection.

Is there any input over voltage and under voltage protection circuit in PMP8878?. I could not see any such circuit. How it is implemented?

Regards

Aneesh

  • 0
    TI__Expert 8210 points
    Aneesh

    PMP8878 features our PWM controller UCC25230. This controller includes input voltage UVLO (under voltage lockout) and OVLO (under voltage lockout) comparators with hysteresis. Please refer to it's datasheet for more information.

    Best Regards,
    Eric
  • 0 in reply to Eric Faraci
    Expert 1735 points

    Dear Eric,

    you mean to say UVLO (under voltage lockout) and OVLO (under voltage lockout) feature can be implemented by UCC25230 IC instead of the main controller UCC28250 correct?.

    But why UVLO and OVLO is not implemented in PMP8878?.I can see pin no 3(UV) of UCC25230  is open.

    Regards

    Aneesh

  • 0 in reply to Aneesh TS
    TI__Expert 8210 points
    Annesh

    You are correct about the part numbers, sorry about that. UCC25230 is used to control the bias power supply and UCC28250 is the main power stage controller. Disabling UCC25230 in this design will remove bias power to UCC28250, causing the converter to stop operation.

    PMP8878 is an example reference design that has it's own requirements. I cannot comment on the specifics of the design criteria, but it may not have had any input over voltage and under voltage requirement.

    Best Regards,
    Eric
  • 0 in reply to Eric Faraci
    Expert 1735 points

    Dear Eric,

    I got your points and cleared my doubts.

    I have one more query for you

    What is the use of capacitor C2(2.2uF,50V) in PMP8878 ?

    Regards

    Aneesh

  • 0 in reply to Aneesh TS
    TI__Expert 8210 points
    Aneesh

    C2 is for DC flux blocking in this full-bridge voltage-mode converter, basically to balance volt-seconds between each half switching cycle so to avoid the transformer saturation. This cap is needed because of the unsymmetrical reality of each half switching cycle due to the switching pulse width difference as well as external parameter differences such as from MOSFETs of its gate thresholds, etc.

    Best Regards,
    Eric
  • 0 in reply to Eric Faraci
    Expert 1735 points

    Dear Eric,

    Thanks for your valuable reply. I understood the functionality of C2 Capacitor in PMP8878. But I have still some more quires regarding this.

    1. C2 comes in series with the  primary and C2 needs to carry the entire primary current of around 4.5A (4.5=12*15/40). 1210 package capacitor is used as C2 such a small package can carry such a large current(around 4.5A)

    2. What is the design formula of C2 capacitor and how much derating has considered for C2 ?.

    3. Why such a series capacitor is not used in PMP8877 brick module design?. PMP8877 also a full bridge based converter.

    Regards

    Aneesh

  • 0 in reply to Aneesh TS
    TI__Mastermind 47195 points
    1. Yes, 1210 is ok to use with that current, You can review MLCC capacitor specs to know this. Our practices as well as industrial numerous practices have proved such for decades.

    2. It is simple, just calculate the C2 impedance based on your switching frequency so to make sure the impedance is small enough compared to the total other impedance when the loop is flowing the current, where mainly is the transformer primary magnetizing inductance impedance on the same switching frequency.

    3. PMP8877 is based on UCD3138 which can programmed to balance the volt-seconds. UCC28250 does not have such a feature.
  • 0 in reply to Hong Huang
    Expert 1735 points

    Dear Hong Huang,

    Thanks for your valuable reply. 

    1. Could you please tell me the rms current rating of C2(2.2u/50V 1210) capacitor since part number of C2 is not available in BOM.

    2. Also Could you please share me the core part number of Transformer T2 and inductor L2 ?

    Regards

    Aneesh

  • 0 in reply to Aneesh TS
    TI__Mastermind 47195 points
    1. GRM32ER72A225KA35, or similar. For MLCC, there is really no current specs but MLCC needs to be considered for their self-heating.
    2. T2, RM6, 3C97.
    L2, RM6, N87
  • 0 in reply to Hong Huang
    Expert 1735 points
    Dear Hong Huang,

    Thanks for sharing the part number.

    1. Do you have any design calculator for the transformer T2 and inductor L2 ? we are not familiar with the PCB winded planar transformer design.

    2. could you please share me the detail of the flux density and the total loss of the transformer?

    Regards
    Aneesh
  • 0 in reply to Aneesh TS
    TI__Mastermind 47195 points

    We do not have a calculator for planar transformer design since it is same for bobbin transformer design. The main difference is how to use the PCB traces to achieve winding so you need to calculate to know how many layers and how much copper weight the design needs. If you are not familiar with PCB winding for planar transformer, one way to learn is to review the gerber file of PMP8878 and PMP8877 - so overall you can just follow these with different turns you need.

    Flux density is overall < 300mT for initial design, but each design will need to fine tune to meet the specs. If you want to copy this design, the 42uH primary inductance is what needed. Total loss can be estimated based on the core specs. The material used in this design was the best at that time. We really did not particularly to estimate the power losses. A practical way is to test several material of state-of-art to see which is better as usually the core vendors only use sine wave to produce datasheet which is different from the PWM. 

  • 0 in reply to Hong Huang
    Expert 1735 points

    PMP8878 Transformer Flux density and temperature rise.docxDear Hong Huang,

    I have just reverse calculated the flux density and the temperature rise of T2 of PMP8878.

    I got Flux density as 0.405T and temperature rise of core is 67.2 degree .

    Is this calculated results are correct?

    Kindly find the attached document for the calculation, and correct me if there is any mistake in calculation.

    Regards

    Aneesh

  • 0 in reply to Aneesh TS
    TI__Mastermind 47195 points

    The flux density is 0.405T/2 = 0.2025T. The 0.405T is peak to peak flux but this 0.405T  varies in 1st to 3rd quadrant so flux density is half of this total delta, as this is a full bridge converter.

  • 0 in reply to Hong Huang
    Expert 1735 points
    Dear Hong Huang,

    Thanks for your valuable time and reply. I understood your points.

    What about the temperature rise calculation?. Is the temperature rise is 67.2 degree?

    Regards
    Aneesh
  • 0 in reply to Aneesh TS
    TI__Mastermind 47195 points

    Yes, that is correct. But notice that is for T=100C. Usually in this type of design, there will be thermal derating requirement to achieve reasonable operation including temperature rise.

  • 0 in reply to Hong Huang
    Expert 1735 points
    Dear Hong Huang,

    Thanks for your continuous support, i have a last and final question on this topic.
    Normally for transformers we have to take copper loss and core loss for calculating the temperature rise, but i have taken only core loss for calculating the temperature rise in PMP8878. Since our transformer is having winding on the PCB i think no need to take copper loss for calculation. Is this correct?.

    Regards
    Aneesh
  • 0 in reply to Aneesh TS
    TI__Mastermind 47195 points
    That temperature calculation you did is on the assumption of core power loss at T = 100C. PCB winding design will make different temperature T for power loss so you need to know T in order to estimate core loss and T includes PCB traces temperature rise.
  • 0 in reply to Hong Huang
    Expert 1735 points

    Dear sir,

    Thanks for your valuable feedback.

    Regards

    Aneesh