• State TI Thinks Resolved
  • Locked Locked
  • Replies 14 replies
  • Answers 6 answers
  • Subscribers 62 subscribers
  • Views 523 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.

TPS92692: TPS92692-Q1

Ponsonby Rivers
Intellectual 620 points
Part Number: TPS92692

I'm using the TPS92692 as a Boost Converter for a LED application. The converter will control a string of amber LEDs at a certain point in time and a string of white LEDs at another point in time. The two strings will not be on simultaneously and will probably have different forward voltages. The output current is controlled via a microcontroller and an external DAC using the IADJ pin. The output current is monitored via a microcontroller to ensure correct current regulation. The requirement is to have a 5ms soft-start period and an enable/disable signal via the SS pin.
(1) After following the TI Boost calculator spreadsheet to determine the correct component (compensation) values, I have entered these values into my simulator and the system has an over-current IS fault as it exceeds 250mV. When using the compensation parameters of the Evaluation Board, the simulation appears to be functioning correctly, albeit exceeding my 5ms soft-start period requirement which is expected as 8ms is the quoted value in the documentation. How do I correctly configure my simulation in a methodical way so that the converter will function as I expect it to?
(2) I'm using a p-type MOSFET (BSH201) to enable/disable the converter on the SS pin however, it takes approximately 130us to fully disable the converter which is the time I apply a signal from the microcontroller to the time PWM switching ceases. I require a quicker turn-off (almost instanteous) as I'm turning on a bleed resistor on the output to discharge the output voltage below a certain threshold. How do I make the disabling of the converter more responsive?
(3) Can you confirm that a type 1 compensator would be more appropriate for this application?

  • 0
    TI__Genius 17690 points

    Hi Tony,

    1.  The SS time is the time for SS cap to ramp up and it’s different for current mode control and LED than a voltage regulator.  It depends on external components and output voltage of the LED.   Since you ate switching different LED loads you there can be high peak current going from one higher voltage LED to lower voltage LED.  So it’s best to over lap the switching (both on) with turn on and off slowly.

    2.  Remove C63 to help with the switching time of your circuit.  Beyond that it’s the RDS on of your FET discharging the SS cap.

    3.  Type one compensation is the one you should use.

    Thanks Tuan

  • 0 in reply to Tuan Tran74
    Intellectual 620 points

    Tuan,

    Understanding the concept of soft-start, how can I reduce the soft-start time without reaching the VIS current limit? The Boost Calculator provides component values which cause the driver to shut down in my simulation. There has been another post titled 'TPS92692: compensation / stability issue' which explains discrepancies with the Boost Calculator values as well.

  • 0 in reply to Ponsonby Rivers
    TI__Genius 17690 points

    Hi Tony,

    What value are you using for the SS cap?  It looks like you are getting an over current during your soft-start.  You should start with 0.1uF as your SS cap.

    If you are trying to reduce your SS time then you have to have the external components like RIS and Caps along with inductor that does not saturate to support the fast SS start time.  This is not really recommended since it will have high peak current and the calculation in spread sheet assumes that you have adequate SS time.

    The compensation depends a lot on you load and the dynamic resistance of the LED along with your inductor and output caps. 

    Thanks Tuan

  • 0 in reply to Tuan Tran74
    Intellectual 620 points

    Tuan,

    I'm using 100nF for the SS capacitor.

    I have altered RIS and the inductor to handle the current for a quicker SS however, I'm struggling to find appropriate compensation values as my simulation is showing instabilities such as voltage and current oscillations. I have followed the Boost Calculator spreadsheet and the datasheet for correctly identifying these compensation parameters to no avail. What do you suggest I try next?

  • 0 in reply to Ponsonby Rivers
    TI__Genius 17690 points

    Tony,

    The calculator asks you to input the dynamic resistance of the LED at the current that it's being used at...Was that done correctly and is the model of the LED used the correct model? 

    For your simulation use dominant pole to get it to be stable for now...Use 0.1uF for Ccomp.

    Thanks Tuan

  • 0 in reply to Tuan Tran74
    Intellectual 620 points

    Tuan,

    The dynamic resistance entered was 0.9R which was determined at the typical voltage and calculated by looking at the I-V curve in the LED datasheet.

    I have used 150nF for Ccomp.

    After reducing the output capacitance, I have a working simulation. However, the ripple current is five times higher than the value specified in the Boost Calculator. The driver is also very sensitive in the simulation. For example, if I add more capacitance at the output, the driver shuts down from an over-current fault from the CSP/CSN pins. How can I reduce the ripple current?

  • 0 in reply to Ponsonby Rivers
    TI__Genius 17690 points

    Hi Tony,

    Are you running PSpice simulation from your first posting...It does not look like PSpice...It's LTSpice.  Am I missing something here? 

    If you are running PSpice then please send me the file and I'll try to help you.

    Thanks Tuan

  • 0 in reply to Tuan Tran74
    Intellectual 620 points

    Tuan,

    I am not using PSpice for my simulation.

    I have managed to obtain a working simulation at maximum Iadj setting however, after setting the Iadj pin down to 1V, I get an over-current fault from the CSP/CSN again. Can I send you an image of the circuit so that you can duplicate this in your simulation environment to resolve my issues?

  • 0 in reply to Ponsonby Rivers
    TI__Genius 17690 points

    Tony,

    You mentioned simulation so please send me what you are simulating with.

    Thanks Tuan

  • 0 in reply to Tuan Tran74
    Intellectual 620 points

    Tuan,

    I have sent you an email of the simulation files and the Boost Calculator. Please confirm receipt of this email.

  • 0 in reply to Ponsonby Rivers
    TI__Genius 17690 points

    Tony,

    I have receive your email and we will take close this thread and take it offline.

    Thanks Tuan

  • 0 in reply to Ponsonby Rivers
    TI__Genius 17690 points

    Hi Tony,

    I have received your email and will close this post and work offline via email with you on this.

    Thanks Tuan

  • 0 in reply to Ponsonby Rivers
    Intellectual 620 points

    Tuan,

    I have decided to use a different vendor for the LED driver as I felt the TPS92692 would not fulfil my requirements. Thank you for your support in attempting to resolve my issues.

  • 0 in reply to Ponsonby Rivers
    TI__Genius 17690 points

    Hi Tony,

    We understand.  We will close this post now.

    Thanks Tuan