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.

LM3404HV Not Delivering Expected Forward LED Current

Other Parts Discussed in Thread: LM3404HV

I am not getting the expected If in my LM3404HV breadboard prototype. I have attached a drawing of my schematic & the desired circuit features are:

Input Voltage = 52 Vdc

LED Vf = 33 Vdc

LED Current = 1000 mA

Switching Frequency = 215 kHz

The circuit turns my series string of LEDs on & I am able to dim using a 0-5 Vdc PWM signal. Right now, I only measure around a 200 mA forward current when I am expecting ~1000 mA.

02021600.PDF

  • First question: Are you actually breadboarding the prototype? If so a switching regulator (any switching regulator from any company) will never work correctly on a breadboard, there is simply way too much noise and the parasitics will cause all kinds of problems. If so I would recommend getting an EVM and modifying it to your needs to test, if not let me know and we can investigate further.
  • Clinton -- I am indeed using a breadboard. Not the best option, but I am in the early stages of evaluating a design. I did discover that the breadboard was causing a lot of parasitic resistance, which resulted in my Rsns path having a higher resistance. I bypassed the breadboard resistance, going straight from IC to resistor to ground, & obtained much better results.

    In terms of using an EVM -- I looked at the evaluation board for the LM3404HV, but it seems that all of the parameters are fixed? Capacitors, inductors, etcs. How would I go about modifying it to test my setup?

    Thanks!
  • It's not just added resistance, it is also added parasitic inductances/capacitances and provides all kinds of little radiating antenna. It is quite likely you are exceeding the -0.3V abs max rating on some, if not all, of the IC pins due to the high frequency ringing produced by breadboarding. It's also likely that the low current was more due to noise on the CS pin tripping the comparator prematurely. Breadboarding often causes damage if not immediately it will eventually. So it is not recommended.

    As for modifying the EVM, surface mount components can be removed and replaced with any decent soldering iron. Just use some tweezers to put upward pressure on the component and heat each end up back and forth until one side pops up. Then grab it with the tweezers and remove the other side with the iron. You will usually damage the component you remove, but you are replacing it anyway.

  • Clinton,

       Thank you! I will give that a shot.

    -Matt