• State
  • Locked Locked
  • Replies 4 replies
  • Subscribers 62 subscribers
  • Views 698 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.

LED Drivers

Tom Woscek
Prodigy 70 points
Other Parts Discussed in Thread: TPS92560, UCC28810

 I have 2 LED Driver designs built and am using: PMP5112 Rev E and TPS92560 _120V_Boost_A. They both work but want to know how to adjust the current to dim the LED strings.
  • 0
    TI__Mastermind 28655 points

    Hello,

    Are these two separate designs?

    It looks like current turned to voltage via R2 is fed back to Vsense.  There is a pull-up of 100K, R20 and a series resistor with the voltage representing current of 100K, R22.  Changing these values will change the output current.  Vsense does need to be above the enable threshold of 0.72 volts (worst case) to start.  That's one reason R20 is on the board.  So sourcing current to the Vsense node will cause it to dim.

    As for the TPS92560 board input voltage 120 VAC?  Is it based on a TI design?

    Thanks,

  • 0 in reply to Irwin Nederbragt
    Prodigy 70 points
    Thanks for the reply. Yes, these are 2 separate designs. The TPS92560 is based on document PMP6008. I used that reference design to build it. The input is approx 135vac to get the 190v output I need to light the string. For that one I am thinking that changing R8 to a variable resistor would vary the current and dim the LED string. For the PMP5112 I am going to try changing R22 to a variable resistor and see what result I get.
  • 0 in reply to Tom Woscek
    TI__Mastermind 28655 points
    Hello,

    Be careful of 135 VAC to 190 volt, on a boost. Nothing limits current if the peak of the rectified AC is above the load voltage. 135 VAC peak is 190.9 Vpk. Also take care in replacing R8 since the output power can increase if the resistance is higher than the present R8 value. This is also true with the UCC28810 circuit. I'd leave the 100K and add the variable resistor in series with it or replace R22 with a 100K variable resistor that can only go lower in resistance.

    Thanks,

    Irwin
  • 0 in reply to Irwin Nederbragt
    Prodigy 70 points

    Thanks, Irwin

    Yes, the LED string for the TPS92560 design is 109 orange leds in series of course and the max current is 70ma. It takes about 135vac input to get the 190v output before they light up. They operate from 190 - 230v. This ckt works great. So, yes I may try something such as a 250ohm variable pot with a 120ohm fixed resistor is series with it to vary the current. Good to know that going over 374ohms will increase the current. With this design I get a max of about 40ma and the LEDs are rated at a max of 70ma. So, I got a little bit of overage current to play with. Varying the input voltage will vary the current and brightness but, I don't want the techs having to vary the input voltage. I'm afraid they may kick it up too high. I want to set it at 135vac and then have them just turn some switches and turn a knob to see of the LED string changes brightness. These ckts are just for testing to make sure they operate before they go back to the customer and they install the assembly back in their final product.

    About the other, PMP5112, yes that was exactly what I was thinking: replace R22 with a 100K pot to vary the current and brightness. Those LED strings are white LEDs and have a max of 1.4A. These are mounted on a heatsink with a thermal pad between them. At 250-300ma they get very hot when operated off the heatsink.  

    Best Regards,

    Tom