• State Resolved
  • Locked Locked
  • Replies 6 replies
  • Subscribers 64 subscribers
  • Views 865 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.

Original question:

LM5118: Buck-Boost Converter Dying

LM5118: Parallel DCDC Converters drop effeciency

Danny Andreev
Intellectual 310 points
Part Number: LM5118

Hello,

I designed a dcdc converter board designed to pulse power a set of high power LEDs.

I have 4 LM5118 Circuits which charge up a 1200uF capacitor which is then switched on across a LED providing a 11A,52.5V pulse at 3.7% Duty.

This works great when I have a single LED driver activated and a single LED being flashed. In this state the DCDC driver consumes 580mA at 48V to power the capacitor. This translates to an efficiency of (52.5*11*.037)/(0.550*48)=80%efficiency (which is not the best but I think I can improve it by using capacitors with a much smaller gate capacitance).

However, when I power up multiple drivers and LEDs in parallel the efficiency drops dramatically. I end up having the same output power on each led but at 3.3A at 48V input. This translates to an efficiency of (4*52.5*11*.037)/(3.3*48)=53%effsiciency

I suspect this may be an issue with the synchronization of the supplies through the SYNC pin, however when I detach this pin, the entire converter starts shorting out. Both mosfets open and let power flow freely through the inductor. 

I am at a lose, please let me know if you have any ideas whats going on.

Best,

-Danny

  • 0
    Intellectual 310 points

    I am currently using a pair of STD70N10F4 mosfets for the switching circuit. They may be the problem because they have a 85c gate capacitance whereas the webench recommendation is for 6.5c gate capacitance fets.

    Here is the webench recommended circuit.

  • 0 in reply to Danny Andreev
    TI__Mastermind 31125 points

    Hi Danny,

    I've assigned this question to the responsible engineer. He will get back to you by this week or next, since there is Thanksgiving holiday this week.

    Thanks,

    Richard

  • 0 in reply to Richard Fu22
    Intellectual 310 points

    Thanks! I will roll out a new version of the pcb with lower capacitance fets and try to get some more data for you to work with.

  • 0 in reply to Danny Andreev
    Intellectual 310 points

    Here is a picture of the SYNC(yellow) pin and HO(blue) (high side mosfet output) pins

    he HO signal goes to 55.1 Vout at its peak and during the wavy part it stays a tor under 48V (vin = 48) But I still dont like how is oscillates, lmk if that is normal.

    This is all captures will a 1mA load. and 50Vout.

  • 0 in reply to Danny Andreev
    TI__Guru* 78220 points

    Hi Danny,

    You circuit is not stable according to your scope pictures. Could you restore the circuit back to the state when it was working, and try out the synchronization first? 

    When you said you paralleled multiple drivers, show do you do the connections between the drivers?  Can you show us a connection diagram?

    Thanks,

    Youhao Xi, Applications Engineering

  • 0 in reply to Youhao Xi
    Intellectual 310 points

    FIXED!

    The issue was shoot through!

    I forgot to include enough low esr capacitance on the output. The electrolytic cap was unable to accept the massive spike generated when the inductor was slammed off. I had a 150V+ 100ns spike which would shoot past my 100V shotkey output diode and couple across my 80V lowside mosfet and damage the lm5118 lowside driver.

    Adding 1x 0.1uF 2x 1uF 1x 10uF ceramic caps on the output of each driver fixed all my issues, including effeciency (which is now at 90%), voltage instability, and driver failure.

    Stupid issue, that took my friend 30 seconds to see after I showed him my schematic.