• State
  • Locked Locked
  • Replies 3 replies
  • Subscribers 63 subscribers
  • Views 542 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.

LM25069 Fails to Start

George Branck
Prodigy 190 points
Other Parts Discussed in Thread: LM25069

 I'm using the LM25069-2 Hot Swap Controller to supply 12V to our CPU board which typical draws 6 amps. The current limit is set to ~40amps and we've seen no issues with start-up until we've added the ability of the board to also provide the switched 12V to an external fan. The fan is rated at a max current of 1.2amps.  The board  fails to start with the fan connected.  Looking at the fan current waveform when the switched 12V is supplied I don't see the fans initial start-up current exceeding 2.5 amps and the shutdown occurs well before the LM25069 fault timeout period.  Has anyone experienced a similar issue when driving a DC fan . Any insight would be appreciated.

Thanks

  • 0
    TI__Expert 5940 points

    Hi George,

    Many items affect start up.  Total Cout used (did the Fans add capacitive loading, even if part of an LC filter?), Plimit setting, if any load is present during Vout rise (does the overall system take advantage of Power Good to delay load current until after Vout rise has completed).  Rshunt value and routing technique (true Kelvin sense).  Obviously layout technique used (are the FET, Shunt, LM25069 all on the same side of the pcb with short routes?).  Is there an RC used on the sense line to filter out nusiance CB trip? 

    Verify that Plimit inception voltage is > 4mv in the design.  This is the voltage on the Rshunt at start up (Vin = full value, Vout = 0v) where the Plimit engine activates.  {Plimit/Vin (max) * Rshunt} has to be > 4mv.  This is due to Vos of the current sense amp used for CL and Plimit.  I don't think this is your issue since you say shutdown occurs prior to timeout but you should verify none the less.

    Without a schematic of board layout file, it is hard to get much deeper.  If you would like to post the schematic details, I can look at next week.

    Brian

  • 0 in reply to Brian Daugherty
    Prodigy 190 points

    Brian,

      Thanks for your input.  The design is not using the PowerGood output to delay fan startup. The fans are directly connected to the Hot-Swapped 12 V. I've done some independent testing of the fans. These are smart fans with an operating range of 5 -13.2 volts. Using an external lab supply the fans turn on at ~ 4.8 volts. At the point of turn-on (4.8v) I'm observing  ~ 4.9 amps current spike for a duration of 10 usec. After this spike the fan current drops to 0 and ~3msec later current ramps as rotation begins. If I apply 12V to the fan the initial current spike increase to ~12 amps with duration remaining at 10usec.  Is there a way to filter out current spikes during startup?  The board is driving 4 fans ..... with 4 fans attached and assuming all turn on at the exact same time(probably unlikely) there could be a ~20amp, 10usec current spike. In your earlier response you suggested an RC on the sense line to possible filter out nusiance CB trips. Can you recommend RC values to accomplish this.?

    Thanks,

    George

  • 0 in reply to Brian Daugherty
    Prodigy 190 points

    Brian,

      Thanks for your input.  The design is not using the PowerGood output to delay fan startup. The fans are directly connected to the Hot-Swapped 12 V. I've done some independent testing of the fans. These are smart fans with an operating range of 5 -13.2 volts. Using an external lab supply the fans turn on at ~ 4.8 volts. At the point of turn-on (4.8v) I'm observing  ~ 4.9 amps current spike for a duration of 10 usec. After this spike the fan current drops to 0 and ~3msec later current ramps as rotation begins. If I apply 12V to the fan the initial current spike increase to ~12 amps with duration remaining at 10usec.  Is there a way to filter out current spikes during startup?  The board is driving 4 fans ..... with 4 fans attached and assuming all turn on at the exact same time(probably unlikely) there could be a ~20amp, 10usec current spike. In your earlier response you suggested an RC on the sense line to possible filter out nusiance CB trips. Can you recommend RC values to accomplish this.?

    LM25069 schematic (pdf)

    LM25069_sch.pdf

    Thanks,

    George