• State Resolved
  • Locked Locked
  • Replies 13 replies
  • Subscribers 63 subscribers
  • Views 1972 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.

TPS568215: Could D-CAP3 is Loop response measurement possible?

Bob Lee62837
Mastermind 7230 points
Part Number: TPS568215
Other Parts Discussed in Thread: TPS548A20,

Dear support member,

My customer used TPS568215,TPS548A20.

TPS568215,TPS548A20 is D-CAP3 topology.

D-CAP3 advantages is adaptive ripple compensation and no-compensation.


There is Datasheet description.
TPS568215 Fig29 Loop Response graph.
TPS548A20 Table2 Recomended component values.

Could D-CAP3 is Loop response measurement possible?
Or impossible?

If Loop response measurement possible,
Could you teach calculate formula?

Best regard.
Bob Lee.

  • 0
    TI__Guru**** 191445 points
    Yes you can measure the loop response in the usual manner, the same as other topologies. If you look at the TPS568215 EVM users guide, www.ti.com/.../slvuaw5.pdf figure 24 schematic, you can replace the 0 ohm R6 with 50 ohm and use TP8 TP9 and TP10 as monitor points for your network analyzer. Details will depend on the model of your analyzer.
  • 0 in reply to JohnTucker
    Mastermind 7230 points

    Dear JohnTucker

    Thank you very much for reply.


    A calculation formula such as DCAP 2 is described
    Is there an application report（SLVA546)?

    www.ti.com/.../slva546.pdf

    Best regard.

    Bob Lee.

  • 0 in reply to Bob Lee62837
    TI__Guru**** 191445 points
    The concept is similar for TPS568215, but the internal circuitry is somewhat different. So far as I know there is no tool available for it. You can only measure the circuit directly.
  • 0 in reply to JohnTucker
    Mastermind 7230 points

    Dear JohnTucker,

    Thank you very much for reply.


    Althought it said, D-CAP3 is similiar D-CAP2,
    Loop bandwidth was different Transient Response.


    Datasheet description
    Loop bandwidth：170kHz (5.86μs）
    Transient Response：20μs


    Why different ?
    What internal control have?


    I have attached file.
    Could you check file?


    Best regard.
    Bob Lee.

    TPS568215_Question.pdf

  • 0 in reply to Bob Lee62837
    TI__Guru**** 191445 points

    That is the actual measured data.  Let me look into it.

  • 0 in reply to JohnTucker
    TI__Guru**** 191445 points
    Actually to correlate the time domain to frequency domain, you have to look at the time from the start of the voltage drop to the time where the voltage starts to recover. It is hard to see in that time scale but it is about 10 usec. That time tp is related to Fco by:

    tp = 1 / ( 4 * Fco) = 1 / (4 * 17000) = 1.4 usec

    This approximation is usually valid for current mode control but may be limited by other factors such as slew rate limiting by the inductor. See this blog post:

    e2e.ti.com/.../how-to-determine-bandwidth-from-the-transient-response-measurement
  • 0 in reply to JohnTucker
    Mastermind 7230 points

    Dear JohnTucker,

    Thank you very much for reply.

    I calculated tp from your send blog.
    Tp is 1.47μs from Loop response.
    Tp is 18μs from Transient Response.

    Why differnt?
    Do D-CAP3 have Duty-cycle limiting?

    I have attached file.
    Could you check attached file?

    TPS568215_Add Question_170123.pdf

    May I ask a once again question?

    Your send blog description is current-mode control and voltage-mode control.
    Is D-CAP3-control current-mode control or voltage-mode control?

    Is D-CAP3-control similiar D-CAP2 Ripple contorl?


    Best regard.
    Bob Lee.

  • 0 in reply to Bob Lee62837
    TI__Guru**** 191445 points

    DCAP and its variants are neither voltage mode or current mode control.  It is most similar to COT or constant on time control.  DCAP is adaptive on time / hysteretic off time control.  The on time is st at the beginning of the switching cycle based on Vin and Vout.  The off time is controlled by a hysteretic comparator.  So for the original DCAP control there must be significant ripple on the output for the hysteretic comparator to function.  This ripple is typically provided by ESR in the output capacitor.  For low ESR ceramic types,, the ripple must be injected externally.  The variants such as DCAP2 and DCAP3 have internal ripple injection.  The key feature of DCAP mode control id fast transient response.  The adaptive on time can respond instantaneously to voltage drops due to transient load changes.  The duty cycle tends towards maximum and the apparent switching frequency increases.  The adaptive on time control is essentially "outside the loop".  That is why the approximation in the blog post does not match well with DCAP.

  • 0 in reply to JohnTucker
    Mastermind 7230 points

    Dear JohnTucker,

    Thank you very much for kindly answer.

    I understood that
    TPS568215 is D-CAP3 control and D-CAP3 is adaptive on time control.


    You said that 『The adaptive on time control is essentially "outside the loop". 』.


    However,
    Tp is 18μs from Transient response.

    Tp is 1.47μs form Loop bandwith.


    It looking at time alone,
    inside the loop.


    Why Transient response delay?

    Best regard.

    Bob Lee.

  • 0 in reply to Bob Lee62837
    TI__Guru**** 191445 points
    Yes ther is both duty cycle limiting as well as frequency change. During the transient event, the on time is increased and the off time is at the minimum. The effective switching frequency is higher. Just out of curiosity, whty is this a big concern for you?
  • 0 in reply to JohnTucker
    Mastermind 7230 points

    Dear JohnTucker,

    Thank you very much for kindly reply.

    You said that 『Duty cycle limiting』.
    How many Duty cycle limiting?


    Althought Datasheet slew rate is 0.5A/usec,
    How many DCAP3 Load slew rate?


    My customer load slew rate is 1A/μsec.
    Is 1A/usec also possible?


    Best regard.
    Bob Lee.

  • 0 in reply to Bob Lee62837
    TI__Guru**** 191445 points

    On the initial transient load step, all the additional current will be provided by energy stored in the output capacitor.  This will cause the output voltage to drop.  When the output voltage is lower, the duty cycle will increase proportional to the voltage drop.  Since the voltage  at VFB is already lower than the hysteretic reference, the off time will be at the minimum and frequency will be increased.  even in this case it may take several cycles for the inductor current to ramp up to the new required value.  So the response can be limited by both the duty cycle and inductor.  TPS568215 can support higher slew rate than 1 A/usec.  0.5 A /usec is just a standard value that we use so users can make apples to apples comparison between parts.

  • 0 in reply to JohnTucker
    Mastermind 7230 points

    Dear JohnTucker,


    I am sorry for late response.

    Thank you very much for reply.

    I was good to knew that TPS568215 was slew rate higher than 1A/us.


    Best regard.
    Bob Lee.