• State TI Thinks Resolved
  • Locked Locked
  • Replies 15 replies
  • Answers 3 answers
  • Subscribers 64 subscribers
  • Views 1553 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.

Original question:

TPS23754: TPS23754PWPR

TPS23754: TPS23754

Sunil N
Prodigy 115 points
Part Number: TPS23754

Hi,

Further to the issue pertaining to making the changes to get a 28V output instead of 24V, i have made few changes in the feedback loop. Below are the changes that are done

1. R32 changed to 4.02K

2. R29 changed to 12K

After making the above changes the circuit outputs 28V but without any load connected to it. But once the rest of the circuit is connected(which consumes 200mA approximately) the 28V output is not constant. Below is the snap shot of the 28V output with and without load

1. Without load

2. With Load

Also the VC voltage changes with and without load as shown below

1. Voltage at VC, without load

.

2. Voltage at VC, with load

In case of the designed 24V output, the voltage is constant at 24V in any case. After making the changes in the feedback circuit to get 28V output i am facing the issue and there is no issue if i revert back the changes to get 24V.

Can you kindly suggest if any changes that needs to be done to solve the issue. 

Thank you.

  • 0
    TI__Mastermind 41760 points
    Hi Sunil,

    I recommend changing the opto to FOD817A. It may be the CTR on your current design.
  • 0 in reply to Darwin Fernandez
    Prodigy 115 points

    Hi Darwin,

    The change had been made earlier itself. Sorry that does not reflect in the schematics. We are using FOD817A (U3) on the board. 

  • 0 in reply to Sunil N
    TI__Mastermind 41760 points

    Okay I understand, next I recommend checking to see if the sense resistor is populated correctly on your board like it is on the schematic. It maybe that your converter is folding back early as you increase the load. Or I recommend decreasing the sense resistor.

  • 0 in reply to Darwin Fernandez
    Prodigy 115 points

    Thanks Darwin, will try it out. Meanwhile do let me know how is the sense resistor value calculated. I followed the datasheet but i am not getting the proper value according to the design.

  • 0 in reply to Darwin Fernandez
    Prodigy 115 points
    Hi Darwin,

    Tried reducing the sense resistor value, but still the issue exists.

    I doubt it is not the sense resistor issue because, for the 24V configuration the circuit works fine without/ with load also. Only after changing the configuration for 28V i am facing the issue.
  • 0 in reply to Sunil N
    TI__Mastermind 41760 points

    Hi Sunil,

    I just saw that the R29 you changed is the opto biasing current. 12k is too large. Can you try with 6.1k? This will keep is similar 4mA biasing current like it was with 24V.

  • 0 in reply to Darwin Fernandez
    Prodigy 115 points

    Hi Darwin,

    Changed the R29 resistor to 6.1K (5.1K+1K in series). Still the same fluctuations exists when the rest of the circuit is connected. Just the POE section disconnected from the rest of the circuit works fine with output voltage constant at around 29V.

  • 0 in reply to Sunil N
    TI__Intellectual 1005 points
    Hi Sunil,
    Darwin is traveling so I'll fill in while he is away. Here are some questions & suggestions:
    1) Were you starting from a TI design / EVM or your old design?
    2) The 50V on C29 feels very marginal for 24V and especially at 28V output. Your diode is rated for 150V. Maybe you can try with a 100V cap.
    media.digikey.com/.../ES3A D.pdf
    3) One theory is that you are drawing more power at 28V than 24V, which could be causing the power supply to shut down. You are drawin 200mA at 28V. Can you try drawing 300mA at 24V to see if the supply continues to operate properly. If it does, we can rule out the "power is too high" theory.

    -Artem
  • 0 in reply to Artem Rogachev
    Prodigy 115 points

    Hi Artem,

    Thanks for filling in.

    1. For one of our processor based application, we have designed a POE based power supply section based on  the TPS23754 chipset, the schematic of which i have shared earlier. It was designed for a 24V output which is working fine. But now we require a 28V output instead of 24V. After making some obvious changes we are facing the issue. During the issue, for debugging purpose we have tried few steps. There is a series diode at the POE output. If the diode is removed, then the POE output is stable at 29V approx. If the diode is connected (power goes to rest of the board) then the output starts fluctuating (output waveform already shared). The same works fine with 24V output.

    2. Will try with 100V rating capacitor 

    3. 24V output is tested with load close to 1A, so the power output is fine i guess.

    Thanks,

    Sunil

  • 0 in reply to Sunil N
    Prodigy 115 points

    Hi Artem,

    We changed the C29 capacitor with 100V rating, still the same issue exists.

  • 0 in reply to Sunil N
    Prodigy 115 points

    Hi Artem / Darwin,

    The transformer used (POE300F-24LD) output is 24V. Will it cause a issue or can we still tune a 28V output

  • 0 in reply to Sunil N
    TI__Intellectual 1005 points
    There are a few things to check when going to a higher output voltage:
    1) Can the transformer handle the power -- You told me that you can do 24V, 1A without a problem so no reason for 28V, 200mA to have an issue.
    2) Will VC get too high -- Your VC seems below the 18V recommended max of Vc.
    3) Will you hit max duty cycle -- Max duty cycle is 78% so you should still have some margin based on your transformer turns ratio of 1:067 (from coilcraft website)

    At this point I can't think of a good reason why you'd have an issue here...
  • 0 in reply to Sunil N
    TI__Intellectual 1005 points
    Okay. Next I would suggest that you look at some other critical signals. i would recommend looking at all of these together to see if there are any dependencies. you may need to play with zoom levels .
    * CTL
    * VB
    * VC
  • 0 in reply to Artem Rogachev
    Prodigy 115 points

    Hi Artem,

    During the issue, VB is fluctuating. Which may cause the CTL/output to fluctuate too, right ?. When the POE section is isolated (with no load on POE output) then VB is constant. Below are the graphs under both the cases

    1. VB - Without load

    2. VB - With Load

    What causes VB to fluctuate? 

  • 0 in reply to Sunil N
    TI__Intellectual 1005 points
    Sunil,
    Vb is derived from Vc with an internal linear regulator. From your previous waveforms we could see that the VCC was not stable. Thus I am not surprised that you are seeing this fluctuation on Vb.

    You need to look at VC, CTL, and Vb on the same scope shot to understand the cause and effect between these waveforms. It would be good to take a zoomed out scope picture and to also zoom in on when VCC is starting to drop off.

    -Artem