• State Resolved
  • Locked Locked
  • Replies 9 replies
  • Subscribers 63 subscribers
  • Views 3839 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.

TPS92515-Q1: Shunt Dimming COFF Connection

Ryuji
Guru 20090 points
Part Number: TPS92515-Q1
Other Parts Discussed in Thread: TPS92515, LM3409-Q1, TPS92661-Q1, LM3409, TINA-TI

Hello,

Should we add the Diode between Vcc to Roff2 in the figure 14 of TPS92515 datasheet?


Since the voltage of the COFF exceeds the threshold（1V typ), it is discharged,I think that additional diode between Vcc and Roff2 is not necesarry.
However, in the EVM of TPS92661-Q1 , there is the diode between Roff2 and 5V external regulator for ensuring Toff of LM3409-Q1.
Why is this diode inserted in this evaluation board? Please let me know this reason.



Best Regards,
Ryuji Asaka

  • 0
    TI__Mastermind 20590 points
    Hello Ryuji,

    This device is better supported in the LED Driver forum so I'm going to move it there so the experts on this device will see the post and be able to help.

    Best Regards, Scott
  • 0
    TI__Mastermind 25500 points

    Hello Ryuji,

    That is a good question. Basically the diode is not totally necessary, but it may be desirable depending on the application. If the LED stack voltage is less than VCC (or the 5V rail in the LM3409 schematic) then it is not really necessary as you can balance the off timer without it without too much difficulty. But if the stack voltage is higher than VCC it could start to influence the off timer more the higher the voltage is. That can be designed around of course but it is easier to calculate the off time if you use the diode as there is only one resistor in circuit at a time.

    Regards,

    Clint

  • 0 in reply to Clinton Jensen
    Guru 20090 points

    Hello Clint,

    Thank you for your reply.
    I understood.

    For use in LMM applications, VLED is expected to be up to 40 V maximum(12 LEDs).
    What is the calculation of toff in that case without the diode?
    Current will flow to Vcc, but how many mA is allowed?

    Best Regards,
    Ryuji Asaka

  • 0 in reply to Ryuji
    TI__Mastermind 25500 points

    Hello Ryuji,

    The equations for calculating toff and the resistor value from VCC are in the datasheet. There is no concern about current flow into VCC because COFF never goes higher than 1V so current can only flow out of VCC.

    Regards,

    Clint

  • 0 in reply to Clinton Jensen
    Guru 20090 points

    Hello Clint,

    Thank you for the reply.

    Could you please let me know more detail of influence of additional diode in your reply?
    "But if the stack voltage is higher than VCC it could start to influence the off timer more the higher the voltage is. "

    I simulated how the switching frequency changes with the presence / absence of Roff 2 / diode in TINA TI.
    Please see the attached file. I use TINA-TI reference design and I add the 100kohm of Roff2 and the diode only.
    In this result , we couldn't cancel the Roff influence by the additional diode.
    Thus , I think that we don't need additional diode between Roff2 and Vcc.
    Is my understanding correct? 
     

    TPS92515-Q1_Roff2.pptx

    Best Regards,
    Ryuji

  • 0 in reply to Clinton Jensen
    Guru 20090 points

    Hello Clint,

    I simulated the Coff waveform on the TINA-TI. Please see the attached file.

    TPS92515-Q1_Roff2.pdf


    I simulated the Coff with/without Roff2 and diode.
    I think that the influence of the presence or absence of Roff 2 is large in Fsw, and the influence of Diode is small in this result.

    I couldn't understand the effect of the diode for LM3409 Coff in the TPS92661EVM.

    Best Regards,
    Ryuji Asaka

  • 0 in reply to Ryuji
    TI__Mastermind 25500 points

    Hello Ryuji,

    You are correct, I was thinking about it the wrong way. Since COFF is always lower than VCC that diode is always forward biased so it really isn't needed.

    As for the effect of Roff2 it will affect the switching frequency to some degree. The amount depends on the resistor values. Most designs we have done use a pretty low Rds(on) shunt FET. The result of that is that Roff2 is usually 10x or more the value of Roff1 so there isn't much effect. As the Roff2 value gets lower as in your simulation it will have more effect and you may need to compensate for it by adjusting Roff1.

    Regards,

    Clint

  • 0 in reply to Clinton Jensen
    Guru 20090 points

    Hello Clint,

    Thank you for your reply. I understood.
    Just to be sure, let me check.

    Is it OK to understand that the Diode(red circle in the attached) of TPS92661EVM is unnecessary?

    Best Regards,
    Ryuji Asaka

  • 0 in reply to Ryuji
    TI__Mastermind 25500 points

    Hello Ryuji,

    That is correct, the diode is not required.

    Regards,

    Clint