• State Resolved
  • Locked Locked
  • Replies 8 replies
  • Subscribers 63 subscribers
  • Views 298 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.

TPS54A20: Regarding the resistance around the FB pin of the TPS54A20

gomi.ryota
Prodigy 140 points
Part Number: TPS54A20

Hi TI Experts, 

I have two questions about the resistance around the FB pin of the TPS54A20.

1. P32 of the TPS54A20 specification states "To reduce high-frequency noise, connect a resistor in series with the FB pin as shown in Figure 62", but what does it mean?

2. P24 of the TPS54A20 specification says "SLVA289 describes the reason for connecting the capacitor (C8) and resistor (R8) in parallel with the FB pin", but SLVA289 does not describe the reason for selecting the resistor, How were they selected?

Clippings of the relevant parts are included in the attachments. Please refer to

6087.TI_question.pdf

  • +1
    TI__Mastermind 18370 points

    Hi,

    Regarding question 1; in some cases (For example when Cff(C8) value is high, noise entering FB pin may be excessive and cause regulation issues.
    This can be remedied by either adding a Rff resistor in series with Cff(C8) or adding a resistor just before FB pin.

    The value of Rff/R8 is usually 1/10 of the value of R9.
    C8 adds a zero to boost the phase margin and should be selected based on the phase margin.

    You can use the attached Excel calculator toll to determine all the values.

    Hope this helps.

    Yitzhak BolurianTPS54A20_excelCalculator_20200508.xlsx

  • 0 in reply to Yitzhak Bolurian
    Prodigy 140 points

    Sorry for the late reply.

    I received a lot of useful information, so it took me a while to scrutinize it.

    I would like you to tell me about the variation in resistance.

    ・　Is there any specification for the variation of R1?

     ・　R7 is ±1%, are there any limits on R8 and C8 as well?

  • +1 in reply to gomi.ryota
    TI__Mastermind 18370 points

    Hi,

    The tolerance of R1 (The feedback voltage divider) depends on the required output voltage accuracy.
    The accuracy of the internal reference in TPS54A20 is +/-0.5%.

    To keep the Vout accuracy to be about 2%, you should use R1 and R2 tolerances of 1% or better.

    R8 and C8 tolerances can be of 20% or better.
    If the above addresses your question, please set the status of this post to resolved.

    Regards,

    Yitzhak Bolurian

  • +1 in reply to Yitzhak Bolurian
    Prodigy 140 points

    Thank you for teaching me a lot.

    I will continue to consider TI's high-spec products, so please support me in that case.

  • 0 in reply to Yitzhak Bolurian
    Prodigy 140 points

    Hi,

    Long time no see.

    I have two questions about the EXCEL file you received, as it differs from the specifications.

    1. For Rbot (R7), the specifications say that 1k to 10kΩ is recommended, but EXCEL says that 10k to 20kΩ is recommended. Since R7 on the evaluation board is also 14kΩ, is EXCEL correct?

    2. The specifications for RPGOOD (R6) say that 10k to 100kΩ is recommended, but EXCEL says 1k to 100kΩ. which one is correct? (Because it's a pull-up, I think anything would be fine as long as it's high enough.)

    Best regards.

  • +1 in reply to gomi.ryota
    TI__Mastermind 29440 points

    Hello gomi.ryota,

    1-  R _bot  is recommended from 1 kΩ and 10 kΩ. please go with the datasheet recommendations. 

    2-  The recommended 10 kΩ and 100 kΩ to a voltage source that is 5.5 V or less.

    Thanks! 

    Tahar

  • 0 in reply to Tahar Allag
    Prodigy 140 points

    Hello Tahar Allag,

    1.　Why is 14.3kΩ used on the evaluation board?

    Best regards.

    Ryota

  • +1 in reply to gomi.ryota
    TI__Mastermind 29440 points

    Hi gomi.ryota,

    14.3K is not too far from the recommended range,  it will work just fine. The recommendation of 1K to 10K is not a hard requirement, you just need nee to be aware that too small FB resistance will increase the power loss and decrease the device efficiency at light loads. Too large resistance will be more susceptible to noise and voltage errors from VFB input current, which affects output voltage accuracy. So, if you can meet the 1K to 10K, that is optimum point, if you deviated from it, just make sure it not too large to create noise and error on the feedback loop.  

    Thanks!

    Tahar