• State
  • Locked Locked
  • Replies 3 replies
  • Subscribers 64 subscribers
  • Views 1164 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.

LM5116 stability

Gabriel Guex
Prodigy 60 points
Other Parts Discussed in Thread: LM5116

Hi,

I'm using an LM5116 for a wide input buck converter that works with a 20 to 80V input, 5V / 50 to 600mA output ranges.Here is the schematic:

All capacitors (except the 47uF/100V) are ceramic.

I'm experiencing some stability problem in some ranges of operation. At ~100mA output, 60V input, the output voltage drops below 4V. Some times it's every few ms and some time every x seconds! On the image below, the first trace is the output voltage and the second is the comp input.

I have 20 cards mounted and the problem doesn't appear the same on every card! I didn't notice this problem on the prototype and the first series of 10 pieces but it was found on the  second series of 20 pieces. In fact the problem was present on some cards of the first series, but it wasn't revealed as the voltage drop is not the same on each card. If it is too low, the display hangs, so I detected the problem.

I've tried a lot of Rcomp/Ccomp/Chf combination, but it seems to have quite no effect, even with a variation (2,7k to 180k, 350pF to 15nF, 0pF to100pF). Of course these values were calculated on the Excel sheet or on the Webbench with different input and output parameters. I've also tried to add some output capacitors.

Can you please help me finding a solution?

Best regards.

  • 0 in reply to Gabriel Guex
    TI__Mastermind 46141 points
    Hi
    Would you please check the issue is caused by loop stability or anything other ?
    Why is there a glitch on COMP signal at 500uSec ? Can you measure SS and UVLO signal together with Vout and COMP ?
    Regards
  • 0 in reply to EricLee
    Intellectual 565 points

    Interesting...just wondering if using a 5V output connected to VCCX is not a problem...according to the data sheet, it should work, but it is very near the borderline.  When some units work, and other units work most of the time, it is always good to see if you have enough margin.  Also, do you have a good decoupling cap close to the VCCX pin?  Perhaps a bit of noise at VCCX is causing VCC to switch from using VCCX to using VIN, or perhaps VCC (which must be a bit less than VCCX) is falling slightly below 4.5 volts, just enough to activate the soft-start (which is activated if VCC falls below 4.5 volts for any reason).

    Interesting to hear from TI people about this, because I have seen the SS get activated in situations where all the input pins seem to be at levels that this should not be happening.

    Another thing to check...in one case which I have seen, a similar behaviour was related to the hiccup timer...on the scope I could count the 256 switching cycles before the event...this despite the fact that there load current was quite far from the point where it should have been activating cycle-by-cycle current limit.