• State
  • Locked Locked
  • Replies 1 reply
  • Subscribers 63 subscribers
  • Views 799 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.

LM3409 Output capacitor

Gonzalo Garcia
Prodigy 150 points
Other Parts Discussed in Thread: LM3409

Hello all:

I have one design of constant current led driver using LM3409

Values of different componets are:

Coff:470pF

Roff:32,2Kohm

L1:22uH

Rsns:0.05

Cin:54uF

Ruv1 6.98kOhm

Ruv2 49.9 kOhm

I expected to have almost 5A in the Led string. I barely reach 4A, but this is not my problem.

The problem is :I am strobing pulses about 1 ms long, 25 times per second. Doing that, I can see (using osciloscope) a high peak of voltaje between led terminals over the first useconds of each pulse, following by a quite high ripple.

I have to check about the high ripple, because is maybe a component choice matters. But the first peak is worrying me.

My questions are:

- How could I protect output led string from "peaks" of current? Just using some kind of diode? or do you have any sugestion

- Should I use an output capacitor to smoooth the final wave? If so, why in the datasheets examples sometimes the output capacitor is not used?

- Do you have some suggestions about how to reduce the ripple current "peak to peak" without reducing total current?

Best regards

  • 0
    TI__Mastermind 25500 points

    To better understand I would need to know what the input voltage is and what the LED string voltage is. But usually a peak like that is a noise issue, possibly due to layout. But at such a low frequency you are probably initiating a startup sequence each cycle since EN is low for long enough for VCC to discharge and go into low power shutdown mode. Try using the UVLO pin instead, it functions the same way but will keep the part alive so the response time is much faster and the device reaction time is faster as well. You can still use Ruv1 and Ruv2, just either OR the signal in with a schottky diode (anode to UVLO pin, cathode to PWM signal) or use a small FET in parallel with Ruv1. If there is still an issue I would be happy to check your schematic and layout for any issues.

    As for the output capacitor, it is optional in a buck converter since the output current is continuous. It will lower the current ripple but it will slow the LED current rise and fall times. It could cause some overshoot as well. They are generally better for analog dimming applications. You can also reduce the ripple current by increasing the inductor value and/or the switching frequency.