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.

TPS23753A won't run

Other Parts Discussed in Thread: TPS23753A, TPS23753

We built a prototype board with this part, using the reference design information in the data sheets, etc.  It worked.  Now we have a production run of boards, and they are not starting / running.

Went through, checked the resistors, they measure as per the schematic.  The FET isn't blown.

We have the same result if powered by POE, or the local adapter input.  So far, everything looks like it is as it should be, but no activity.

Any ideas what to look for? 

  • Hi David, I need to move this to the correct forum.
  • David,

    Can you send any waveforms to give us a clue? Please show the output voltage waveform and also the waveform at TPS23753A Vc pin. Seeing the TPS23753A CTL pin can also help. It is quite possible that a wrong value passive component is installed during manufacturing flow.

    Also, let us know which TI reference design or schematic was used for this design. Or, you can send your schematic.

  • Thanks for moving to the correct forum.

    The signal on Vc is near zero volts.  The signal on CTL pin is near zero.  Nothing is running.

    I have a .pdf of the schematic, can't figure out how to attach it yet.

    We did an unconventional test just now - I took a big FET and a driver, and connected it in place of the FET in the normal design.  It's driven by a 50% square wave.  I then slowly turn up the lab supply feeding the local power jack on our board.  The transformer and output stages produce 5V, [obviously not regulated] and doesn't draw unexpected current.  During the test, the Vc supply also has 5 V on it.   The FET drive pin of the TPS23753 sits at 0V during this. The voltage on the CTL pin followed the 5V output.

    The supply is disconnected from the normal circuit it is intended to run during this kind of test, to avoid risk of blowing it out with unregulated voltages.


    There must be something not right with the startup / enable and all that of the  TPS23753....    The output parts from the transformer on do what I expect.

  • Here are voltage readings from a working prototype board, running with a 10 ohm load on 5V.
    TPS23753A Pins:
    1- 1.88V
    2- 5.05V
    3- .04V
    4- 12.78V
    5- 3.64V
    6- 0 RTN [reference for the measurements]
    7- 0.51V
    8 & 9- 46.6V
    10- 46.6V
    11- 0.51V
    12- 4.32V
    13- 1.24V
    14- 1.24V

    I'll add readings from a dead board tomorrow.
  • It appears that something is over-loading either the Vc or Vb pins. At startup, Vc will ramp up via an internal current source from the VDD1 pin until Vc hits about 9V. When Vc gets to about 5 or 6V, Vb should also come up to 5V. When Vc hits 9V, switching should start.

    If neither Vc or Vb are active then there must be something loading them. Is it possible to measure load resistance from these pins? Maybe you can compare static meter measurements side by side with a working board to see what may be the cause?

  • I checked 2 dead boards and the good one. 

    Pin 4 = VC   -  The good one reads megaohms / open, except for seeing the capacitor charge.

    The bad ones, read 70 and 170.   So it seems something is wrong there.  ...

  • I lifted the VC pin on one of the bad boards, and the low ohms was on the IC. Replaced it, and it's working.
    ...
  • For some reason, it appears that the ABS Max voltage of the Vc pin (19V) was violated. This could occur if the control loop was not properly setup or went open loop for some reason. Now that you have a working board, try observing the start up behavior (Vc pin and DC output) with a scope when operating from a lower input voltage (from a DC power supply set up as an adapter input). Check for overshoot or other mis-behavior at start up as the DC input supply voltage is gradually increased.

  • I now have 4 new ones working. As best I can figure right now, the initial board I was working with was not what I thought it was.
    After blowing up 1 or 2, I was brought into the troubleshooting. I discovered that the main inductor was installed backwards, which puts a dead short where the primary should be. Of course, I said put those aside, and use a new unpowered one, fix the part, then try it. That board didn't work, but I wasn't working on it yet... Maybe it got mixed up, I don't know. I just took 2 un touched ones, put the inductor correctly, and then powered them, and they work. Still, all of this information is very helpful to know, not only for this time, but in the future when trying to figure out why one won't run.