• State TI Thinks Resolved
  • Locked Locked
  • Replies 6 replies
  • Answers 3 answers
  • Subscribers 65 subscribers
  • Views 384 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.

TPS92691EVM-001: Integral compensation for Boost LED driver

Satoshi
Guru 19595 points
Part Number: TPS92691EVM-001
Other Parts Discussed in Thread: TPS92691, TPS92691EVM-752

I trying to change TPS92691EVM-001(SEPIC) to wide voltage range Boost LED driver with low-side sensing. 

Please let me know about external compensation parts; 

①Can boost LED driver apply to use integral compensation? (Datasheet equation (37))

②If ① is yes, which ωp equation for buck or boost by table-1(Datasheet P24)?

　Or, please let me know equation for Boost LED driver's Ccomp/

③ Ccomp changed 20nF to 220nF ⇒ normal operation range was expand

　 Ccomp changed 220nF to 1μF ⇒ specification was not improvement

　I guess that above reason (220nF to 1μF) is limit to TPS92691 internal transient, correct?

【Background】

・Vin: 24V ⇒ Vout: 36V~110V, Iout: trying to reach 500mA_max

　(External parts were changed match high voltage specification, etc)

・At first based on proportional integral compensation circuit (equation (38)~(40)), waveform was oscillation.

・When changed integral compensation, to certain extent was normal operation.

 　When above the certain Vout or Iout (or Pout?) level, it occur oscillation.

Best regards,

Satoshi

  • 0
    TI__Expert 3910 points

    Hi Satoshi,

    Why not trying with TPS92691EVM-752, a boost / buck-boost (boost-to-battery) EVM?

    And remember that with your output voltage being higher than 60 V, you have to make sure all the pins on 691 device cannot see the output voltage directly, especially the CSP, CSN and VIN pins. Otherwise, 691 device could be damaged due to the voltage exceeds absolute maximum voltage.

    I hope it helps.

    Thank you!

  • 0 in reply to Issac Hsu
    Guru 19595 points

    Issac-san

    Thank you for reply,

    The reason for choice TPS92691EVM-001 are two points below;

    ・Higher Vout condition need to change low-side current sensing from below E2E thread.

    https://e2e.ti.com/support/power-management/f/196/p/867907/3242547

    ・About EVM pattern and layout, TPS92691EVM-001 is easy to modify low-side current sensing and high voltage specification. 

    　TPS92691EVM-752 is not secure low-side current sensing area on EVM.  

    Best regards,

    Satoshi

  • 0 in reply to Satoshi
    TI__Mastermind 18890 points

    Hi Satoshi-san,

    Based on your wide load and operating condition, you have to use just a dominant cap (integral) as your compensation.  Because there is a rd (dynamic resistance of the LED) in the equation and your wide range.  In addition, capacitance on a cap will change with DC voltage on them...It best that you do transient response or dominant pole method where you increase the Ccomp until you get the appropriate response. 

    Otherwise, you can use the excel spread sheet on the TI website to help you with the calculation.  Note that cap value that you put down has to be capacitance at the operating voltage ect:  www.ti.com/.../toolssoftware

    Note that you will need to sweep the loop for stability after compensation or do your transient response to cover all your operating conditions.

    Thanks Tuan

  • 0 in reply to Tuan Tran74
    Guru 19595 points

    Tuan-san

    Thank you for reply,

    I trying to set integral compensation: increase Ccomp and remove CHF and Rcomp.

    But Ccomp be larger than a certain capacitance(220nF, 470nF, 1μF, 4.7μF), the same threshold (example is below) become unstable.  

    Example of unstable threshold; 36Vout / 490mA, 52Vout / 450mA, 54Vout / 400mA, 57.5Vout / 350mA, 64.9Vout / 300mA,

    　　　　　　　　　　　　　　74Vout / 250mA, 79.5Vout / 200mA, 95.5Vout / 150mA, 110Vout / 120mA

    　　　　　　　　　　　　　　※110Vout / 100mA is stable operation

    Please let me confirm about calculation method below.

    ①On the boost application, are DC Gain and Pole Frequency equation based on boost condition by Table-1?  

    　Which equation (37) or (38) should I use?

    　※When use boost condition and equation (38), I concern that Ccomp become large value.

    ②Which ILED based on min or max case on these equation?

    ③Please let me know about mean of rD, i can't imagine the principle of "Vout × rD × Cout" on Table-1. 

    　I think that rD = (⊿Vf / ⊿If) × #LED, is it correct? (Vf and If are described on LED's datasheet)

    Waveform for normal operation and unstable operation are attached below;

    ※ch1(Yerrow): Vout, ch2(Green): SW, ch3(Blue): IMON, ch4(Pink): Isw

    【Normal: 24Vin, 53Vout, 400mA_ILED】

    【Unstable: 24Vin, 54Vout, 400mA_ILED】

    By the way, proportional integral compensation(with CHF and Rcomp) based on datasheet and spread sheet, always oscillation.    

    When fine adjustment value, result was also oscillation.

    Best regards,

    Satoshi

  • 0 in reply to Satoshi
    TI__Mastermind 18890 points

    Hello Satoshi-san,

    For wide varying loads I mention on the previous post that you have to use the integral compensation so don't even attempt the proportional integral compensation.

    To answer your question.

    1.  Equations 38, 39 and 40 are proportional integral compensation for your boost regulator.  But remember that you can't use this since the you load is wide and varying thus this will not work for it.  It's used for fixed load and fixed output current

    2 and 3. Your understanding of rD is correct and it's figured out on the LED data sheet base on the operating current and voltage.  You have wide varying loads and conditions so it's not going to be possible for you to use equations here.  You will have to determine compensation emperically or transient response.

    So we are back to try and figure out why you have instability.  Can you verify something for me first.  Is the oscillation only happening when you have greater than 50% duty cycle on the gate drive?  If this is the case then it has to do with slope compensation and not your loop compensation.  Based on one data point of ※110Vout / 100mA is stable operation then when you go up to 120mA it goes unstable.  

    If slope compensation is the case then you have to read about that section in the data sheet to choose the power stage component correctly.  Again, the spread sheet will help you with it also.

    Thanks Tuan

  • 0 in reply to Tuan Tran74
    TI__Mastermind 18890 points

    Hello Satoshi-san,

    We will close this thread for now.  It will open if you response to it with additional request.

    Thanks Tuan