• State Resolved
  • Locked Locked
  • Replies 4 replies
  • Subscribers 64 subscribers
  • Views 281 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.

TPS922052: Unable to smoothly fade in/out using Flexible Dimming mode

Elan
Prodigy 80 points
Part Number: TPS922052
Other Parts Discussed in Thread: TPS92641,
  • Using TPS92205xEVM
  • Input voltage 48V
  • Connecting to a single COB LED (18V / 500mA)
  • PWM frequency (for ADIM + PWM input) 4KHz

I'm initializing Flexible Dimming mode as described on the forum, and then setting ADIM pin to 12.5% duty cycle for 4 A * .125 = 500 mA

I'm then doing PWM ramp on the PWM/EN pin to fade in (over 2 seconds) and out (over 2 seconds). For 13-bit PWM this means I change the duty cycle every 247uS approximately. This is roughly the 10000:1 dimming ratio advertised (13-bit = 8096 levels). I've also tried doing it over 4 and 8 seconds as well in case this matters, but it doesn't seem to have an effect.

There are two effects I observe:

  1. It's impossible to completely turn the LED off, likely due to the current leakage issue described in https://e2e.ti.com/support/power-management-group/power-management/f/power-management-forum/1241615/tps922055-leakage-current-issue-when-driver-is-in-off-state
  2. More importantly, the dimming quality below around 5% is not smooth at all. As the PWM sweeps smoothly between those last 5% - 0%, the actual brightness level jerks considerably, seemingly to stair-step down. On the way up from 0% - 5% I see similar stair-casing.

Any advice? I'm coming from a Shunt FET design with TPS92641 which was super smooth the entire range, and looking to see if these new chips can provide the same (or very similar) quality.

Only thoughts I had were around output capacitance, and also perhaps running ADIM @ 12.5% affects ability to do PWM smoothly down to dark.

  • 0
    TI__Mastermind 19715 points

    Hi Elan,

    I am currently out of office celebrating Chinese New Year. I will reply to you after Feb.18 when I am back to office. Thanks for your understanding.

    Best Regards,

    Steven

  • +1
    TI__Mastermind 19715 points

    Hi Elan,

    Thanks for your patience.

    Elan said:
    I'm then doing PWM ramp on the PWM/EN pin to fade in (over 2 seconds) and out (over 2 seconds). For 13-bit PWM this means I change the duty cycle every 247uS approximately. This is roughly the 10000:1 dimming ratio advertised (13-bit = 8096 levels). I've also tried doing it over 4 and 8 seconds as well in case this matters, but it doesn't seem to have an effect.

    It would be helpful if you could provide a graphical illustration on the timing sequence of your control signal so that I do not misunderstand your meaning.

    Based on my current understanding, you are going to drive the EN/PWM pin with 4kHz PWM with a 13-bit (8192:1) resolution / dimming ratio. Am I correct?

    If this is the case, then please be noticed that this is out of the capable range of TPS922052, which has a minimum PWM on time of 150ns (Please refer to Section 7.5 Electrical Characteristics from the datasheet. A screenshot is shown below). That is to say, theoretically, the maximum dimming ratio under 4kHz PWM dimming is around 250us:150ns = 1666:1. However, the 150ns is only the minimum on time of the device, but not the minimum on time of your system (including the inductor and capacitor, etc., which will slow down then turning-on/off of the current output). So the actual maximum dimming ratio you can achieve will be lower than the theoretical 1666:1.

    Elan said:
    More importantly, the dimming quality below around 5% is not smooth at all. As the PWM sweeps smoothly between those last 5% - 0%, the actual brightness level jerks considerably, seemingly to stair-step down. On the way up from 0% - 5% I see similar stair-casing.

    Most likely this is why you see this star-step down in your 5% to 0% transition as this short pulse (5% - 0%) is out of the minimum on time that the device can recognize and also your system can react. You can try with a lower PWM dimming frequency or lower dimming ratio and see if this helps.

    Best Regards,

    Steven

  • 0
    Prodigy 80 points

    Hi Steven, thank you very much for the reply. I appreciate the clarifications and I think based on your reply and some more experimentation on my part, I'll stick with the Shunt FET solution.

  • 0 in reply to Elan
    TI__Mastermind 19715 points

    Hi Elan,

    You are welcome.

    If you take both resolutions of EN/PWM pin and ADIM/HD pin together in flexible dimming mode, there is a chance that TPS922052 can achieve higher dimming ratio compared to TPS92641 when only shunt PWM dimming is used. And you can save the shunt FET in TPS922052 solution.

    Best Regards,

    Steven