• State TI Thinks Resolved
  • Locked Locked
  • Replies 2 replies
  • Answers 1 answer
  • Subscribers 63 subscribers
  • Views 807 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.

LM3414HV: Having to compromise between unacceptable ripple current without an output capacitor or squealing at certain loads with one!

Henry Roe
Prodigy 15 points
Part Number: LM3414HV
Other Parts Discussed in Thread: LM3414

Hi,

I'm currently designing a 4 channel DC-DC LED driver using four LM3414's for a domestic lighting application.

It is required to fit into a rather small poorly ventilated enclosure, so the switching frequency has been brought right down and a large value inductor has been  increased to accommodate the lower switching frequency and try to reduce the ripple current as much as possible. The load will also be connected via an unknown length of cable which could be quite long.

Design parameters:

Vsupply: 12 - 48v

Vled: 4.5v - 45v

Iout 350mA - 1A (selectable) with 10 bit PWM dimming

Fs: 51KHz

L: 470uH

Cin: 3.3uF

Cout: 1uF

I have also included a 1mF capacitor on the input to decouple the supply from the previous CV input supply. 

In terms of current regulation, dimming and power dissipation, the circuit performs admirably across all load conditions without Cout, but the ripple current on the output is totally unacceptable.

When Cout is included however, the supply whines horrendously when the load voltage is around 39-43v. I have used small heatsinks on the ICs which seem to act a bit antenna like (not too bad mind) and the input loop between the catch diode, power switch and Cin is larger than it could be to stop the heatsinks fouling with Cin for each channel. Not sure if this could be the problem, the loop area is still pretty compact.

I thought I'd just post this here to see if anyone has any words of wisdom whilst I carry on trying to solve the problem. I can provide schematic and extracts from the board layout if it helps.

Thanks in advance.  

  • 0
    TI__Mastermind 25500 points

    Hello Henry,

    Does this noise happen at steady state operation or during PWM dimming? My guess is PWM dimming and if so what is your PWM frequency? Ceramic capacitors are often noisy when used with PWM dimming since they have piezo characteristics. If that is the case you can usually get rid of it with some experimentation. One way is to use aluminum electrolytic, tantalum, film, or some other non-ceramic technology that does not make noise. Another way is to play with case sizes and number. For example, you can replace one 1206 capacitor with two 0805 capacitors (or any other variations on case size and/or number). That accomplishes two things: Different case sizes buzz at different frequencies, and going to 2 rather than 1 (for example) splits the RMS current between the two and makes them less likely to buzz in the first place.

    If it happens when you are not PWM dimming (100% duty cycle on DIM) then there is something else going on we would need to investigate.

    Regards,

    Clint

  • 0
    TI__Mastermind 25500 points

    Hello Henry,

    I am not sure why I don't see the information here, but I did get your waveforms over e-mail. The scope shots make it look very much like there are some noise issues, usually those are due to layout. I would be happy to look over it if you can send me the schematic and layout. If you would rather not attach it to a public forum you can send me a friend request and we can do private messaging (or share direct contact info). Let me know. Thanks.

    Regards,

    Clint