• State Resolved
  • Locked Locked
  • Replies 7 replies
  • Answers 1 answer
  • Subscribers 62 subscribers
  • Views 691 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.

TPS92641: Want to use the UDIM pin as 'Strobe' pin

user3942675
Prodigy 240 points
Part Number: TPS92641

Hello.

I am trying to design 30A LED driver with LM3434 , but I have thought that TPS92641 is better device after I read another thread in this community.

Since some my customers want to use high power LED like flash lamp(e.g. LED-On:100usec. and LED-Off:1 to 10 sec. cyclicly ), I want to use the UDIM pin as a 'Strobe' pin.

So, would you tell me how much time it will take to reach intended LED current after the UDIM pin is set to 'high' ?

I'm sorry I am unused to English sentences.

Best regards,

 

  • 0
    TI__Mastermind 25500 points

    Hello,

    The current rise time when UDIM goes high depends on the inductor value, input voltage, and output voltage (LED voltage plus CS voltage).

    It is governed by v = Ldi/dt. So your rise time is dt = Ldi/(Vin-Vout). I hope this helps.

    Regards,

    Clint

  • 0 in reply to Clinton Jensen
    Prodigy 240 points

    Dear Clint,

    Thank you for the reply.

    > So your rise time is dt = Ldi/(Vin-Vout).

    Follows are my operating conditions:

    Vin=24V,  Vout=4 to 5V,  Iout=30A,  fsw=500kHz and L=1uH.

    In that case, the rise time was calculated at 1.5usec. I appreciate that the response of UDIM pin is fast enough.

    Now there is an additional question.

    Is the response time same If I turn on the LED that has been left turning off for some hours by UDIM pin?

    Best regards,

  • 0 in reply to user3942675
    TI__Mastermind 25500 points

    Hello,

    No, the UDIM pin has a 13ms timer. After 13ms the device goes into a low power shutdown mode. So next time it is pulled high VCC will need to charge above its UVLO threshold and then COMP will have to charge. That could take in the multiple ms range. If it needs to respond quickly after such a long time you would probably need to use the shunt FET which will burn power continuously in the FET while the LED is off.

    Regards,

    Clint

  • 0 in reply to Clinton Jensen
    Prodigy 240 points
    Dear Clint,
    Thank you for the answer again.
    Shunt FET... If I possible, I do not want to use it because there is concern about power loss and heating of the FET.

    There is another question again.
    Is TPS92641 placed into a 'low power shutdown mode' when the UDIM voltage exceeds 370mV (e.g. 500mV)?

    Best regards,
  • 0 in reply to user3942675
    TI__Mastermind 25500 points

    Hello,

    That is a good question. I will try to find out for you. If you can avoid low power shutdown by holding it at a midpoint value that would allow for a fast response after a delayed time.

    Regards,

    Clint

  • 0 in reply to user3942675
    TI__Mastermind 25500 points

    Hello,

    I was able to confirm from a colleague that it will not enter low power shutdown if you keep UDIM above the low power shutdown threshold. I tested it out on the bench as well at a rate of 5mHz (200 seconds between pulses). The rise time will get a little longer the longer you have UDIM low however since the COMP cap slowly discharges over time due to leakage. But charging VCC is the bulk of the startup time and it stays alive. So even after hours the current rise time should still be pretty fast although not as fast as if it were repetitive.

    Regards,

    Clint

  • 0 in reply to Clinton Jensen
    Prodigy 240 points
    Dear Clint,
    Thank you for your ongoing support.
    It is very good news!
    So I consider making 'PMP8016' to evaluate.
    If PMP8016 was available for sale, I could buy it and test right away ...

    However that may be, Thank you very much!