• State TI Thinks Resolved
  • Locked Locked
  • Replies 8 replies
  • Answers 1 answer
  • Subscribers 62 subscribers
  • Views 352 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.

LM3445: Low dim level issue using LM3445

Wade Novin
Prodigy 40 points
Part Number: LM3445

I am using the LM3445 in accordance with the application note SNVS570M

At low dim levels the buck converter transitions from regulating the current to constant on state of the switching transistor.

From the application note:

As the TRIAC fires beyond 135°, the DIM decoder no longer controls the dimming. At this point the LEDs will dim gradually for one of two reasons:

1. The voltage at VBUCK decreases and the buck converter runs out of headroom and causes LED current to decrease as VBUCK decreases.

2. Minimum on-time is reached which fixes the duty-cycle and therefore reduces the voltage at VBUCK. The transition from dimming with the DIM decoder to headroom or minimum on-time dimming is seamless. LED currents from full load to as low as 0.5 mA can be easily achieved.

The problem is that the output of the LED is not smooth between these two states.The result is that when dimming from full off slowly up there is a noticeable "pop" or "flash" 

The brightness steps down noticeably when the buck converter takes over.

Is there a way to adjust this point to achieve a smoother transition. 

I believe this would require altering the input voltage to FLTR1 or adding a bias voltage.

The schematic page 29 of the application note there is a resistor R14 between C3 and D11.

I assume this R14  adds a small voltage bias or?

Thanks in advance for any advice

 

  • 0
    TI__Mastermind 28655 points

    Hello Wade,

    What you are describing is a powering up issue which is different then when you are dimming down.  Are there issues when you are turning the dimmer down or is that smooth?  There are passive components in the triac dimmer that causes the duty cycle of the decoder to be higher when there is no load.  When the dimmer is turned up from off the LEDs don't initially load the triac causing the phase angle to read higher than when the LEDs are on.  This causes the flash when first turned up.  R14 creates an offset allowing the LM3445 to pull more current when dimmed further down.  This will change the dimming behavior but the flash issue may persist.

    If you have an oscilloscope you can look the rectified AC when first turning the dimmer up as well as FLTR2 and ASNS to see what is causing the flash.

    If you place an incandescent bulb in parallel with your LED light does it work correct when dimming down and back up?  Interfacing with triac dimmers can be difficult.

    What is your LED stack voltage?

    Best Regards,

  • 0 in reply to Irwin Nederbragt
    Prodigy 40 points

    Irwin,

    It does happen when dimming down as well, but perhaps different behavior

    Adding an incandescent load does not alter this phenomenon. I happens predictability when stepping from 4% to 5% levels on a control device that has this capability.

    Adding R13 did eliminate the pop on but did alter the dim curve as suspected. Also it added a flickering in the level.

    I did scope the 'GATE"  pin and at the 4% level it was constant high and the at 5% it starts to switch hi to lo.

    Perhaps adding a circuit that turns off the LED at levels below 5% dim is the solution.

    The stack voltage at full on with 2.4 Amps is 38 volts

    Thankd

  • 0 in reply to Wade Novin
    TI__Mastermind 28655 points

    Hello Wade,

    Are you using the valley fill circuit?  If so what are your capacitor values?  I ask because 38V at 2.4A is 76W.

    The LM3445 is a peak current sense hysteretic controller, if the current doesn't reach the peak threshold the MOSFET will stay on.  If the inductor current ripple is high the current can jump quite a bit in this mode since the LM3445 just keeps the MOSFET on.  The reason it is doing this is due to the available voltage not being high enough to regulate current/reach the peak current trip threshold on the LM3445.  Turning it off before it gets to this mode is one way to avoid it.

    Best Regards,

  • 0 in reply to Irwin Nederbragt
    TI__Mastermind 28655 points

    Hello Wade,

    I haven’t heard back from you, I’m assuming you were able to resolve your issue.
    If not, just post a reply below (or create a new thread if the thread has locked due to time-out)

    Best Regards,

  • 0 in reply to Irwin Nederbragt
    Prodigy 40 points

    Irwin,

    The valley fill is two stage with 150uF caps. The inductor value is 680uH.

    I did not see the capacitor hold up voltage going below the LED stack voltage.

    The inductor ripple is below 10%.

    One idea I had was to add a resistor from FLTR2 to ground to effectively lower the 750mV current sense trip point and then reduce the current sense resistors to compensate for the lowered trip point. This would presumably shift the dim curve down a little a bit so the transition point when the controller starts current regulating is lowered 

    Thanks 

    Wade

  • 0 in reply to Wade Novin
    TI__Mastermind 28655 points

    Hello Wade,

    What is your design based on, the datasheet?  SNVS570M is the datasheet for the LM3445.  Have you looked at the rectified AC waveform when these transitions happen?  Is this design 120 VAC or 230/240 VAC?

    Changing what you mention above won't change anything.  If you divide FLTR2 down the voltage will be lower, then lowering the current sense resistors will increase the current back up.  Effectively it will operate the same.

    I might be useful if I could look at your schematic.  There are circuits that have been added that help with keeping the triac conducting during dimming.  See LM3445EVM-695 for example.

    Best Regards,

  • 0 in reply to Irwin Nederbragt
    Prodigy 40 points

    The design is based on the SNVS570M application note for 120 VAC operation.

    I can send the schematic to you in email as there is some proprietary IP involved.

    I do not believe that the triac is going out of conduction as I have a circuit that drives a fan to keep current through it at low dim levels

    Thanks!

  • 0 in reply to Wade Novin
    TI__Mastermind 28655 points

    Hello Wade,

    I'll close this thread since we're communicating via email.

    Best Regards,