• State
  • Locked Locked
  • Replies 7 replies
  • Subscribers 64 subscribers
  • Views 355 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.

TPS923650: Maximum PWM Frequency Operation

Kevin Gerlock
Prodigy 70 points
Part Number: TPS923650
Other Parts Discussed in Thread: LM3404, LM3406, LM3409,

Tool/software:

Can the TPS923650D1 be operated at a PWM rate of 44.1kHz?  I am looking for a constnat current device to drive LEDs at this rate.  If the subject part cannot support this speed can you recommend one the will?  Either a 1-channel or 3-channel device with seperate PWM inputs is what I am lokking for with current capabilty up to about 1.5 amps.

  • 0
    TI__Mastermind 23190 points

    Hi Kevin,

    What is the dimming ratio that you need under 44.1kHz? In other words, what is the minimum output current pulse period that you need?

    Best Regards,

    Steven

  • 0 in reply to Steven Li
    Prodigy 70 points

    Hi Steven,

    Ideally I woiuld like a ratio of 1450:1 which is a minimum pulse period of 15.6ns.  I know that is a challenge.  I have a working proto using an NS4115 and I am limiting the minimum width in my MCU firmware and just living with limiting the dynamic range on the low end.  I would prefer to use a device from a more well-known manufacturer so that is why I am casting a wider net for devices now that I know my concept works.

    I really don't need 1.5 amp capability since my LEDs draw 350mA but in the future I might want higher current.  Since I started this thread I have found the TI literature on parallel MOSFET PWM and think that is likely going to give me the best result.  I am thinking that an LM3404 or LM3406 would suffice with a parallel MOSFET.

    I do like the fact that the LM3409 comes in a PDIP package that makes breadboarding very easy for me.  If I breadboard with the LM3409 and do the final design with the lower current '04 or '06 will the resulting performance be the same?  In the end I will need a 3-channel circuit to drive RGB LEDs so if there is any device with multiple channels that would simplify my BOM.

  • 0 in reply to Kevin Gerlock
    TI__Mastermind 31585 points

    Hi, Kevin

    Our expert will reply you today.

  • 0 in reply to Kevin Gerlock
    TI__Mastermind 23190 points

    Hi Kevin,

    Kevin Gerlock said:

    Ideally I woiuld like a ratio of 1450:1 which is a minimum pulse period of 15.6ns.  I know that is a challenge.  I have a working proto using an NS4115 and I am limiting the minimum width in my MCU firmware and just living with limiting the dynamic range on the low end.  I would prefer to use a device from a more well-known manufacturer so that is why I am casting a wider net for devices now that I know my concept works.

    I really don't need 1.5 amp capability since my LEDs draw 350mA but in the future I might want higher current.  Since I started this thread I have found the TI literature on parallel MOSFET PWM and think that is likely going to give me the best result.  I am thinking that an LM3404 or LM3406 would suffice with a parallel MOSFET.

    A 15.6ns minimum current pulse is a very challenging target, even for a shunt (parallel) FET solution. But you are correct - Shunt FET is what you should look for as it provides much shorter current rising / falling time in comparison with the main FET dimming like TPS923650. (You can refer to Section "LED illumination" in this white paper - Common LED Functions and LED Driver Design Considerations.)

    I would say it is possible to use LM3404 and LM3406 with a shunt FET, but you need to modify the basic circuit. (For example, you can take a look at the User's Guide for LM3404FSTDIMEV Evaluation board | TI.com.)

    Kevin Gerlock said:
    I do like the fact that the LM3409 comes in a PDIP package that makes breadboarding very easy for me.  If I breadboard with the LM3409 and do the final design with the lower current '04 or '06 will the resulting performance be the same?  In the end I will need a 3-channel circuit to drive RGB LEDs so if there is any device with multiple channels that would simplify my BOM.

    The control mode for LM3409, LM3406 and LM3404 are different, so most likely the resulting performance will not be the same.

    Best Regards,

    Steven

  • 0 in reply to Steven Li
    Prodigy 70 points

    Thank you Steven for the reply.

    I would like to avoid the added complexity of an external FET for PWM dimming so I would like to know what the current rise/fall time is when using the DIM input pin of the device.  Do you have any test data on the current rise/fall times?  The data sheet has no information on this.

    As I said in my previous post, I plan to limit the minimum PWM width and live with the loss of low-end dynamic range.  Test data on the internal PWM timing would be of great help to me.

  • 0 in reply to Kevin Gerlock
    TI__Mastermind 31585 points

    Hi, Kevin

    Our expert will reply you today.

  • 0 in reply to Kevin Gerlock
    TI__Mastermind 23190 points

    Hi Kevin,

    The output current rise/fall time will depend a lot on the external component selection and application conditions (like the inductor, the output capacitor, the LED spec, input voltage). There is almost no delay between the device recognizing the voltage level change of the DIM pin and the device turning on/off the switching FET. This is the reason why we do not specify the current rise/fall times in the datasheet. Section 8.2.2 from the datasheet contains some application curves for your reference.

    I have some questions - Do you need a Buck LED driver (like LM3406) or Boost / Buck-Boost LED driver (like TPS923650)? What is your input voltage and LED voltage? You mentioned two different types of LED drivers above and it makes me confused. Please kindly clarify. Thanks.

    Best Regards,

    Steven