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.

LM3429: TOF CAMERA's LED DRIVER Schematic Problem Review

Part Number: LM3429
Other Parts Discussed in Thread: OPT9221, , OPT8241

HI Subhash Chandra Sadhu

Test environment

- OPT8241-CDK-EVM kit is used and IR LED circuit is in use Refer to attached circuit

- OPT8241-CDK-EVM Using the firmware for the built-in laser and adjusting the Voxel SDK parameter without firmware modification.

- OPT8241-CDK-EVM OPT9221 ILLUM_N Signal line R99 1K Installed R101 does not use open processing ILLUM_P signal

LED DRIVER Schematic Problem Review

1. Circuits PL7 and ZD1 do not use LED DRIVER but 20V as power supply separately. The LED is driven without any problem.

    However, when 20V is input using LM3429 LED DRIVER, the switching noise is heard and the IR_LED_OUT voltage shakes.

Thank you Subhash Chandra Sadhu6330.tof_sch_170327.pdf

 

  • I am trying to get one of these two identical threads moved to the "optical sensors" forum as nobody in this forum is familiar with that EVM. Thanks.
  • TOF CAMERA's LED DRIVER problem should be solved. Please let me know your engineer name

    Clinton Jensen Thank you

  • Hi!

    I have taken a cursory look at your schematic. I believe that you have a problem of the output voltage of the boost converter LM3429 dropping out of regulation whenever the camera is being driven.
    Can you confirm that you can actually hear the PCB make some sound? Can you try increasing the register value of pix_cnt_max to 1.5 times the original value and see if the sound reduces in its frequency?

    Regards,
    Subhash
  • hi Subhash Chandra Sadhu

    - You said the boost converter output voltage is a problem. Which part is the problem?

      Led directly to IR_LED_VOUT. Then the output voltage is not a problem.

    - Please look at the sending circuit again. And answer

     And if you connect the ILLUM_EN signal from the tof camera to Figure 2, you hear the switching noise and the output voltage shakes.

     tof_sch_170329.pdfWhat is the problem

  • Hi!

    Given your schematic I am guessing that you have an issue of LM3429 falling out of regulation. A potential reason could be that the input supply to LM3429 can't sustain the power requirement. Can you compare the IC power requirement to the power source capability?
    Also, we expect the output voltage of LM3429 to reduce when LM3429 is disabled by the ILLUM_ENz signal. So that is an expected behavior of the IC.
    When you say you hear switching noise, can you confirm if you actually hear sound from the PCB? Highly varying voltages on the power supply can cause the capacitors to generate some sound because of mild piezoelectric nature of the dielectric in the capacitors.

    Regards,
    Subhash
  • hi Subhash
    Thanks for the answer.
    - The pix_cnt_max parameter value was changed and the switching noise was reduced. What is the cause
    - The input power of the lm3429 is 5V
    - Switching sound seems to be heard on PL8 coil parts
    Thanks ,Subhash
  • Hi!

    The cause of the noise is a combination piezoelectric nature of the ceramic capacitors and magnetostrictive effect of the inductor. This is a well known problem. Please refer to the following links:
    - www.capacitorindustry.com/.../
    - en.wikipedia.org/.../Coil_noise

    Can you verify if your 5V supply can give enough current for the peak power requirement of LM3429?

    Do you see if the camera image is looking bad? If not, your only problem may be that your board makes some acoustic noise. If you see excessive flickering in your camera image, you may have to do further debug.

    Regards,
    Subhash
  • hi Subhash

    Thanks for the answer.

    - 3A power supply to give the LM3429 5V power. The LM3429 has enough current.

     Also, as you can see from the circuit diagram, there is no problem in driving the LED.

    - The camera image does not look bad. Switching noise sound is a problem.

    - I connected the PWM signal applied to the LM3429 NDIM pin to the beacon instead of the ILLUM_EN signal. There is no switching noise and the LED works well.

    What is the cause

    Thanks

  • hi Subhash

    - The OPT8241-CDK-EVM firmware is using firmware for the laser. Is there firmware that uses the ir LED applied to the OPT8241-CDK-EVM kit?

    - If you have the firmware for the LEDs applied to the OPT8241-CDK-EVM kit, can you send it to me

    Thanks

  • Hi!

    The OPT8241 chipset is reconfigurable to work in multiple modes of operation. To make the LEDs work, you would have to keep the modulation frequency at a value less than 20MHz. The LEDs could dissipate a lot of heat. So I would recommend that you start using the board at a relatively low (4%) integration duty cycles. Once you get the image and you are sure that the board is not overheating, you could carefully increase the integration duty cycle to improve depth performance.
    These changes are possible by just modifying register settings in Voxel viewer. A firmware change is not required for these changes.

    Regards,
    Subhash
  • hi Subhash
    I've been doing a lot of LED lamp design in the meantime. In particular, we have developed automotive lights using TI LED drivers. For example, headlights, work lights, etc.
    So, considering the heat that Subhash was worried about, we designed the LED PCB by applying metal PCB from the beginning and also made a heat sink for heat dissipation.
    The LM3429 power input is designed and calculated as SFH4715S_OSRAM IR_LED 1S_6P VF_2.9V_1000mA at 12V input. The output current is 1.7A [20.4W]
    If I connect the ILLUM_EN signal from the opt8421 to the LM3429 nDIM pin, why do I hear switching noise?
    For example, if you connect another PWM signal [beacon], you will not hear any sound.
    The value of the pix_cnt_max parameter that Subhash advised has changed. So the switching sound is getting smaller. But the sound did not completely disappear.
    What is the cause? may be difficult, but I need help with Subhash.
    Thanks
  • hi Subhash

    I've been doing a lot of LED lamp design in the meantime. In particular, we have developed automotive lights using TI LED drivers. For example, headlights, work lights, etc.

    So, considering the heat that Subhash was worried about, we designed the LED PCB by applying metal PCB from the beginning and also made a heat sink for heat dissipation.

    The LM3429 power input is designed and calculated as SFH4715S_OSRAM IR_LED 1S_6P VF_2.9V_1000mA at 12V input. The output current is 1.7A [20.4W]

    If I connect the ILLUM_EN signal from the opt8421 to the LM3429 nDIM pin, why do I hear switching noise?

    For example, if you connect another PWM signal [beacon], you will not hear any sound.

    The value of the pix_cnt_max parameter that Subhash advised has changed. So the switching sound is getting smaller. But the sound did not completely disappear. 

    What is the cause? may be difficult, but I need help with Subhash. Thanks

  • In reply to Subhash Chandra Sadhu:

    hi Subhash

    I've been doing a lot of LED lamp design in the meantime. In particular, we have developed automotive lights using TI LED drivers. For example, headlights, work lights, etc.

    So, considering the heat that Subhash was worried about, we designed the LED PCB by applying metal PCB from the beginning and also made a heat sink for heat dissipation.

    The LM3429 power input is designed and calculated as SFH4715S_OSRAM IR_LED 1S_6P VF_2.9V_1000mA at 12V input. The output current is 1.7A [20.4W]

    If I connect the ILLUM_EN signal from the opt8421 to the LM3429 nDIM pin, why do I hear switching noise?

    For example, if you connect another PWM signal [beacon], you will not hear any sound.

    The value of the pix_cnt_max parameter that Subhash advised has changed. So the switching sound is getting smaller. But the sound did not completely disappear. 

    What is the cause? may be difficult, but I need help with Subhash. Thanks

  • Hi!

    Sorry for delaying this.
    The sound that you hear is coming because some large value capacitor has its voltage changing a lot or some large value inductor has its current changing by a large value. This happens because the capacitors/inductors expand and contract a little whenever their voltage/current change. If they expand and contract, they create sound waves in air at the frequency of expansion or contraction.
    If the nDIM pin is switched at frequencies which are in human hearing range, approx. 20Hz to 20kHz, you will be able to hear the sound from the PCB. So when you try to inject a PWM signal into the nDIM pin you may not hear anything because of its frequency. Can you tell us the frequency of the nDIM pin?

    Regards,
    Subhash
  • Hi Subhash,
    I also have acoustic noise problem, the frequency of ILLUM_EN is 60Hz@15fps and 120Hz@30fps.
    How could I do to improve the noise?

    Regards,
    Eric
  • Hi Eric,


    Here is a post which explains how to reduce the sound from the capacitors: 

    Regards,
    Subhash

  • Eric, Minsu Park,

    Please let us know if following the guidelines helps reduce the sound.
    These guidelines are specific to the ceramic capacitors whose voltage changes a lot.

    Regards,
    Subhash