PMP22477: PSR MODE Design question

Hi Zen,

Thanks for reaching out to us! May I know the purpose of your effort (like passing the EMC test)?

I just checked this external control loop. I felt I also have some questions and need to confirm with the author. I will get back to you in the next week!

Seems you applied dithering and try to disable the external control loop. Changing R67 from 0.05 ohms to 0.015 ohms increases the loop gain by 0.05/0.015 = 3.33 or 10.4dB.  The loop may be unstable with this much increase in gain.

Best regards,

Diang

Hi Diang

     thanks for your reply.

     Because of cost and system applications, I try to do this test.

     The power consumption is between 35W and 40W in my system. the class of power will be fixed type 3 class 5 that I'm sure.

      I think reduce type control circuit, CS current is not enough on type 3.

      The power transformer sound gets louder as the load increases. So I try applied frequency dithering to reduce sound.

       Finally, I measure MOS' gate signal that is unstable.       

        How to calculate open loop gain (0.05/0.015 = 3.33 or 10.4dB)? 

        How range is the open loop gain on type 3?

         From PMP22477 test report (page 13), I don't know if it's related.

B.R

Zen

Hi Zen,

Thanks for your additional information! 

The current sensing gain (positively correlated with the sense resistance) is at the denominator of the control-to-output gain (I modified this sentence after rechecking the document, really sorry for the potential troubles). So when you applied smaller current sense resistor the gain will be increased (by 20*lg(0.05/0.015) in your case). The following link could be helpful:

https://www.ti.com/lit/an/snva555/snva555.pdf?ts=1666046164924&ref_url=https%253A%252F%252Fwww.google.com%252F

I will consult the author of PMP22477 and get back to you this week. 

Best regards,

Diang

Hi Zen,

Thanks for the patience. 

The purpose of this circuit is for input power limitation. It is discussed in the paper below:

https://www.ti.com/lit/an/slyt787/slyt787.pdf?ts=1609952767818&ref_url=https%253A%252F%252Fwww.google.com%252F

For me it looks like for power-sharing applications, so you may not need this circuit if you have one PoE input only. 

One thing you might miss is D9 which is connected with COMP pin. Could you try to remove D9, change the R67 to 0.05 back, and disable dithering? Let me know if the transformer sound has any differences.

Best regards,

Diang

Hi Diang

    Thanks your information.

     I try to remove D9, disable dithering and change R67 to 0.05, the sound has difference that better than before.

     The sound like switch on/off,I measure NMOS' gate signal.

    I study TPS23730 functional block diagram  and try to change some elements (COMP/CS/DTHR...) about NMOS' gate.

     Do you have some suggestion for my testing?

thanks

B.R

Zen

Hi Diang

    I do loading test on TPS23731 EVM (TPS23731EVM-095) which the Architecture is same as PMP22477 (PSR mode and remove input power limitation circuit).

    when disable/ enable dithering, this has the same question (switch on/off sound).

    The sound gets louder as the load increases.

    Can you help confirm?  thanks

B.R

Zen

Hi Diang

   Thanks your testing reference. understand your testing.

   I tested  TPS23731EVM  (W/ dithering) which has two sounds (high frequency & similar switch on/off)

   If I change C39 to 1.8nF, the small high frequency sound is gone. But the switch on/off sound still exists.

   could you check if change C39 to 1.8nF?

  Note:  PMP22477 EVM has the same sound, But this EVM had modified( remove in put power limitation circuit, w/ dithering).

            so I confirm both EVMs at the some time.           

thanks

B.R

Zen

Hi Zen,

Thanks for your updates!

I tried both 2.2 nF and 1.8 nF cap with TPS23731. Under 5-A load, W/ dithering, I could hear small sounds for both cases and I could not tell the difference. Using the phone App helped me to find a difference during 10 - 15 kHz. I am wondering if you need to use dithering for EMI test or only for solving the hearable sound. Seems when I do not use dithering I will not hear the sound which seems is from the transformer. 

Changing that cap will cause different modulation frequencies. 1.8 nF will result 13.44 kHz and 2.2 nF will result 11 kHz modulation frequency. The upper limit hearing frequency for human adult normally is 15 - 17 kHz. 

1.8 nF with 13.44 kHz modulation frequency:

2.2 nF with 11 kHz modulation frequency:

Best regards,

Diang

Hi Diang

   OK, you can close this discussion.

    thanks your support

B.R

Zen

Hi Zen,

Glad to help! If you have any further questions please feel free to contact us!

Best regards,

Diang