Because of the holidays, TI E2E™ design support forum responses will be delayed from Dec. 25 through Jan. 2. Thank you for your patience.

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.

LM5116: Erratic behavior in a 20V 20A buck

Part Number: LM5116

I am trying to make a DC/DC buck converter from 24...29V to 20V 20A output (The output will be eventually made into a CC source later). But for now, I have difficulties to make it work properly as a buck.


The issue is that beyond some 9...13V (depends also with the current output, seems to be stable until attain some 150-200W output) the circuit start to oscillate (?) , the output seems to become something alike 2-3 pulses are sticked together and after that there are 1-2 discharges of the energy accumulated. The waveform is not stable, not periodic, simply buzzing around erratically...

I have read that the circuit make a simulation of the charging coil current using a ramp generator, and a capacitor at pin Cramp. It is using also a voltage collected through pins Cs and Csg, using kelvin measurement (parallel independent traces from Rs). I did this with accuracy. The problem may be if the coil does not have exacly the inductance calculated, but i tried using a ferrite core to vary it's inductance and could not obtain a result; still erratic behavior.

Attached the schematic and PCB. The PCB is made by using three pieces; the controller is implanted through terminals in the power board, to have possibility to change it, and i had done much attention to have very short paths for ground and for the gate traces to MOS's. Also the power board is made from two halves; one including most electrolytic capacitors, one only with a few and some ceramics. THe two power boards are united by thick straps of copper,  for low resistance and inductance. The thick power traces are also strenghted with copper wires appropiate for 20A current.

I really can't figure out how to make it to work normally. It behaves fine at low power outputs; but when i pass over 10-13V at 8-10A, it start fuzzing and the oscilloscope show just madness...

I am just guessing that may have also something to do with the RAMP injected current, but i tried adding various amounts of current (tens of microamp) to the Cramp, from Vcc through a potentiometer to Cramp, also even from Vout; but, except there is some improvement until voltage output can be raised to maybe 15V and also the output current to some 12-15A, beyond that is reacting again erratically and cannot go further...

  • Of course the 6 and 4 capacitors figured there at input and output are properly interconnected, and on another piece of PCB.

    The coil has 20 turns of thick mutithread wire, 2,5 mm CSA. Is in air, without core. The internal diameter is approx 35mm, and is approx like a Brooks coil. Has a calculated inductance of about 35uH, but in realty is probably 25uH because of the wire insulation (PVC insulated; is just a double-core speaker wire, i am used it for the tests only)