• State
  • Locked Locked
  • Replies 3 replies
  • Subscribers 62 subscribers
  • Views 559 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.

LM3405 oscilation? inestability? what is happenning?

Julian Antonacci
Prodigy 70 points
Other Parts Discussed in Thread: LM3405

Hello!

I'm having problems with font based on lm3405.

I have the boost derived from a 4.7V zener with a capacitor .01uF

capacitors are (following the data sheet naming):

C1= 10uF
C2 = C5 = 1uF

the inductor is large


When R1 = 0.8ohm
IL = 0.2V / 0.8ohm = 250mA. The average current works well. But if I put the oscilloscope SW I see I have three cycles at 100% and 0% two cycles.

but when i change and put R1 = 0.4ohm duty cycle varies permanently in SW. If I put the tip of the oscilloscope on FB I see a triangular signal 35 to 40khz that never reaches 200mV.


Does anyone have the same problem? Does anyone know what is happenning?

thanks

  • 0
    TI__Mastermind 25500 points
    Without further information, such an Vin, Vout, all component values, etc... it is hard to say. But often erratic switching like that, particularly if it gets worse as the output current goes up, is either due to noise issues due to layout or insufficient gate drive. Are you using an EVM for the layout, or something else? If something else and you have pictures I'd be happy to look at it and comment. But first, can you confirm the zener is fully biased to 4.7V? Have you tried increasing the gate drive capability by using a 5.2V zener and confirm it is fully charged to 5.2V. The rule of thumb in switching converters is the more gate drive the better.
  • 0 in reply to Clinton Jensen
    Prodigy 70 points
    Hi, thanks for your reply. I have vin=12V, Vout arround 5V. The zener is fully biased at 4.7V with 5mA. I didn't try with 5.3V zener but i'll do it.
    In this link: "www.dropbox.com/.../AADSfaKasDGLVU2A7Oaj47CFa" there is a folder with the board and schematic images.

    What is the minning of EVM? i'm ussing Pertinax.

    Thanks!
  • 0 in reply to Julian Antonacci
    TI__Mastermind 25500 points
    Hello. I was able to generate the pictures of the top and bottom layers, but I'm not sure what the middle file is. Dropbox doesn't seem to come through our network well for some reason. In any case I am pretty sure this is a layout issue. Buck converters have discontinuous input current, meaning during the switch on time it is pulling current from the input supply and during the off time it will pull none. This means it is the biggest possible noise generator in the circuit so proper and careful layout are very important as with any switching power supply. My main concern is the grounding in this case. The input capacitor grounds are seperate from the IC ground which is seperate from the power diode ground etc...
    You need to have a solid a ground as possible between CIN, the IC, the power diode, and the current sense resistor. If the layout is too difficult to connect them directly together on the top plane (which it often is) this can still be accomplished using a bottom (or inner) layer ground plane and place at least a couple vias directly by the ground pad for each. You still want to get their ground points as physically close as possible and do not cut the copper between them on the ground plane with any traces. You might be able to test this out a little by using thick, short wires soldered to the board to tie all these ground together and see if there is improvement.
    Two more things you will want to check just to make sure everything else is good is the inductor and the capacitors. Just make sure the inductor saturation current rating is high enough to handle the peak currents at full power and make sure the capacitors are a good dielectric, like X5R or X7R type. Y5V and those like it will lose significantly more capacitance both when a voltage is applied and also over temperature. This can cause noise issues as well.