• State TI Thinks Resolved
  • Locked Locked
  • Replies 5 replies
  • Answers 2 answers
  • Subscribers 29 subscribers
  • Views 134 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.

DLP4620SQ1EVM: LM3409 sync period

myungjin song
Prodigy 50 points
Part Number: DLP4620SQ1EVM
Other Parts Discussed in Thread: LM3409,

Tool/software:

Hello 

I have a question about LM3409.

1. In the DLP4620SQ1EVM circuit diagram, Roff and Coff are not connected separately, and the Sync period seems to be set in TPS99001. And 6.8k ohm is connected in series. Is that right?

2. In the LM3409 data sheet, there is an equation for the switching frequency. In order to know Toff, I need to know Coff and Roff, so how do you calculate switching frequency? And what is η? about 0.95?

3. I want to reduce ΔiL-PP as much as possible. How can I do that? I don't know how to calculate Roff and Coff since they are not connected.

4. To design ΔiL-PP close to '0', should I change LM3409 to another part?

  • 0
    TI__Mastermind 27605 points

    Hi Song,

    We don't recommend using the LM3409 as a laser driver because the current is not controlled closely enough with this part.

    Thanks

  • 0 in reply to made4engineering
    Prodigy 50 points

    Please answer questions 1, 2, and 3. These are questions about the LM3409, not related to the laser.

  • 0 in reply to myungjin song
    TI__Expert 6650 points

    Hi,

    Aaron is trying to say that you should not use LM3409 for laser illumination. If it is not recommended, then we do not think we need to answer the initial questions because they are no longer relevant to getting laser illumination to work.

    If you really must know, then below are some short answers for those who may look this information up in the future.

    1. Yes, R_OFF appears to be in series, similar to the datasheet reference.
    2. Eta (η), the efficiency, is assumed to be 85% to 100%
      This is defined below in the datasheet:
    3. Section 9.2 shows an example for an arbitrary subsystem on how one would calculate values. The design process involves selecting your own COFF or ROFF value to calculate the other, and thus TOFF.

    Again, we do not recommend using LM3409 for this application.

    Regards,
    Michael Ly

  • 0 in reply to Michael Ly
    Prodigy 50 points

    Thanks for your reply. Currently you are not recommending LM3409, but we are unfortunately using LM3409 at the moment, so we need a detailed answer.

    2. In the LM3409 data sheet, there is an equation for the switching frequency. In order to know Toff, I need to know Coff and Roff, so how do you calculate switching frequency? And what is η? about 0.95?

    -. alright. η is 85 to 100%. how about Fsw? Coff, Roff are not included in the current schematic, how do I calculate Fsw?

    4. To design ΔiL-PP close to '0', should I change LM3409 to another part?

    -. So what part do you recommend instead of LM3409? 

  • 0 in reply to myungjin song
    TI__Expert 6650 points

    Hi Myungjin,

    The way that we use the LM3409 for our DLP application is a little different from the way the datasheet describes it.

    We have a sync signal that "hijacks" the COFF pin.

    T_OFF and T_ON vary from moment to moment, F_sw is not a constant value--it is ever-changing and non-periodic.

    The team for LM3409 can answer questions regarding this IC, as I specialize in DLP parts.

    TI is also still working on our own laser illumination design, but I suggest you reach back out to our TI engineer who previously messaged your team through email. Those emails already recommended not using the LM3409 and instead gave a concept for a different buck converter design. Please reach back out to that engineer for additional help with the hardware design.

    I will close this thread now because the general answer is the LM3409 will not work for laser illumination.

    The way TI uses this part uses a sync pulse to override the RC timer present in the typical design that pulls the COFF pin to GND via a capacitor. If this sync pulse, which we have little control over, pulses before minimum T_ON time is reached, the device might not properly turn off and continue to ramp up current. This can cause issues such as higher current than expected, and cause potential damage to the laser package. The LM3409 simply does not have a fast enough switching speed, nor does it have the proper fine-current-control capability.

    There are laser drives that TI sells, but not all may meet your needs such as allowing for dimming. Again, please reach back out to our TI engineer. I will try to get him to follow up on the original emails.

    Regards,
    Michael Ly