• State
  • Locked Locked
  • Replies 3 replies
  • Subscribers 63 subscribers
  • Views 546 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.

LM5116 PWM buck controller

jarod salsman
Prodigy 30 points
Other Parts Discussed in Thread: LM5116

I designed two different buck converters in 2010 using the LM5116 PWM controller.  One was a 5V, 5W supply.  The other was a 12V, 2W supply.  I have been struggling with the 12V supply ever since.  Today I was at a different computer than my own so I re-downloaded the data sheet.  The newer data sheet has extra design considerations compared to the old data sheet.  Specifically, the enable blurb on page 12 now includes directions to ensure the EN pin rise time is less than a single switching cycle.  The previous version allows for the EN pin to be tied directly to Vin (which is what I did).  Can anyone tell me the reason for this change?  My Vin slew rate is much slower than my switching frequency. 

  • 0
    TI__Intellectual 940 points

    Jarod,

    You can tie enable to VIN  through a resistor without a problem.  The inclusion of the warning in the datasheet applies to situations when the enable pin is used as an on/off  control.  Be sure you tie it through a resistor (about a Meg Ohm for high voltage inputs and about 100k for low voltage inputs).

    Regards,

    Ron Crews

  • 0 in reply to Ronald Crews
    Prodigy 30 points

    Hi Ron:

    I put a 100K pullup from EN to Vin with a MOC207 open collector opto-isolator as an on/off switch to determine if this change was the reason for my issues.  I do not believe it made any difference.  I actually had to drop the resistance to 10K to get the EN pin to slew from 0 to 3.3V in one switching cycle (3us).  Either way, I have the same problem which is the following:  The SS pin maintains a voltage of 0.54V.  I have checked the VCC, EN, and UVLO voltages for any sign that they are seeing a loss in voltage.  All signals look fine.  Also, the SS pin voltage is DC which to me says the LM5116 internal logic isn't toggling the SS pin due to a fault condition.  I have tried 0.1uF < Css < 2.2uF.  The higher Css value seems to make a difference but the converter still shuts down and then restarts at a constant frequency (about 50Hz give or take).

    I have another PWB with the same circuit on it.  I had the same issue with this board.  I finally popped my output inductor off the board to make sure I had not saturated the core.  I measured the inductance to be exactly what it was supposed to be.  I had my tech put it back (the same part) on the board.  I powered the circuit out of exasperation and got 12V out.  I asked the tech what he did.  He had used a hot air solder station to remove the inductor.  The entire circuit is in close proximity.  I figured there must have been a bad solder connection.  However, I have now replaced the current LM5116 chip 3 or more times because it looks like the SS pin cannot power the capacitor.  Brand new chips have the same issue. 

    Another interesting tidbit... The circuit that works doesn't soft start.  The capacitor is populated and = 0.1uF (measured).  This should be about a 10ms startup time.  Actual startup time is less than 100us.  I am starting to suspect the internal 10uA source for Css.  Any thoughts?

    Regards,

    Jarod.

  • 0 in reply to jarod salsman
    TI__Intellectual 940 points

    Jarod,

    This has become more than a question about the datasheet change.  I need some more information regarding what you are seeing as a failure mode.  Does the output quit switching or it it some other symptom?  Also I need a schematic with values or a schematic and a BOM.  Any other information you have regarding the failures would be helpful.

    Also, is VCCX tied to the output in this application.

    Thanks,

    Ron Crews