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.

TPS23754: Inquiry regarding frequency noise problem of TPS23754

Part Number: TPS23754

Hi, TI experts

There is a request for technical inquiry from a customer regarding the use of TPS23754PWP in an RFID model.

The customer's RFID model operates by ORing the POE power and DC power to the TPS23754 to produce 5V output.

DC power is located in Option 2 of Datasheet_Figure 26 of TPS23754, and POE is designed to operate even when DC power is applied.

CE certification of the RFID model is currently in progress, but it is failing when DC power is applied.

The problem is that noise at a frequency that is a multiple of 240KHz bounces around.

It has been confirmed that disconnecting the connection between TPS23754 and the DC line, and instead connecting an external DC-DC converter, eliminates the problematic noise when the 5V output is operational.

The conclusion is that the issue is attributed to the switching frequency (240kHz) generated during the operation of TPS23754PWP.

Below is the debugging content tested to solve the problem. (There was no effect at all)

1. Remove noise by connecting a common mode inductor (240KHz / 6K Ω) to the DC power line → No effect.
2. Remove noise by connecting a 22uF decoupling capacitor to the DC power line → No effect.
3. Connect Bead to DC power line (240KHz / 1K Ω) → No effect.
4. Wrap Ferrite Core around Adapter Power Cable and Test → No effect.


Q1) Could you please confirm if there are any additional debugging considerations other than the test contents above?

Q2) Is there a guide to reduce noise?

Please check how to improve or what needs to be reviewed.

Thank you.

  • Hi Grady,

    Thanks for reaching out to us!

    - What is your DC power adapter's voltage? If it is too low the converter may not work well due to flyback transformer saturation / duty cycle limiting / Rcs limiting... These may cause abnormal noise amplitude.

    - You mentioned "DC power is located in Option 2 of Datasheet_Figure 26 of TPS23754, and POE is designed to operate even when DC power is applied." Do you use the APD pin prioritize the PD? If so, the PoE may not keep operating when DC adapter is applied. If not using APD, you may need additional external hotswap in that oring circuit

    - Could you share the schematic?

    Best regards,

    Diang

  • Hi, Diang

    We will provide you with the customer's response to any additional inquiries you may have.

    - We are using a 24V/4A adapter for the product in question, and there is no shortage of current.

    - Regarding the APD, when both the adapter and POE are connected simultaneously, the adapter power is utilized. When connected individually, each operates with its respective power source.

    When connected to the 24V/4A adapter, it is connected to VDD1 in the attached circuit diagram (Option2).

    - The circuit diagram is also attached.

    SCHEMATIC1 _ 10_POE.PDF

    Please check. Thank you.

  • Hi Grady,

    Thanks for your schematic.

    In the circuit the POE300F-50L is used. Looks like it is mainly designed for 36-72V. I putted your input condition and 5V/5A out then get 2.61-A max current on the primary side winding. Since current will be lower at lighter load, do you see a RFID pass at light load condition.

    ...

    By the way, how much is R315?

    Best regards,

    Diang

  • Hi, Diang

    1. In the circuit the POE300F-50L is used. Looks like it is mainly designed for 36-72V. I putted your input condition and 5V/5A out then get 2.61-A max current on the primary side winding. Since current will be lower at lighter load, do you see a RFID pass at light load condition.

    → In the actual review model, when powered by an external adapter with 12V/4A, the RFID operates with an output of 5V/2A during operation and 5V/500mA when not in operation. During CE certification tests, it was confirmed that the same issue of noise at 240kHz frequency persists even when the RFID is not functioning.

    Additionally, I am having difficulty understanding the waveforms provided above. Could you please provide an explanation?

    2. By the way, how much is R315?

    → The value of R315 is 0.18 Ω.

    Please check. Thank you.

  • Hi Grady,

    Thanks for your reply.

    Are you talking about "24V/4A" adapter or "12V/4A" adapter on the Adapter Option 2 (J16)? "12V/4A" may not work on Option2.

    The waveform is the primary winding current and voltage, where the configuration is shown below. The winding current indicates the core flux. Since 24-V is lower than the 36-72V input range of POE300F-50L. And lower input voltage results in higher input current at the same power. So if it is related to the core flux, I am wondering if under a light load will this RFID pass.

    Another point I noticed is you having both PPD and APD circuits on the same adapter. I think you need to make one of this circuit not soldered. If Adapter Option 2, you can try to set R362 OPEN.

    Best regards,

    Diang

  • Hi, Diang

    Thank you for reply.

    This is what the customer has confirmed regarding the information you have provided.

    1. Are you talking about "24V/4A" adapter or "12V/4A" adapter on the Adapter Option 2 (J16)? "12V/4A" may not work on Option2.

    → The circuit diagram of the review model was designed with reference to the reference circuit in the SLVU301 document, and a 24V/4A adapter of 21.6~57V is used as shown in the attached photo below.

    →  This is a photo taken of Figure 1 inside the SLVU301 document.

        

    2. The winding current indicates the core flux. Since 24-V is lower than the 36-72V input range of POE300F-50L. And lower input voltage results in higher input current at the same power. So if it is related to the core flux, I am wondering if under a light load will this RFID pass.

    →  Even in a low current consumption state (low load) when RFID is not in operation, noise problems occur.

    3. Another point I noticed is you having both PPD and APD circuits on the same adapter. I think you need to make one of this circuit not soldered. If Adapter Option 2, you can try to set R362 OPEN.

    →  Let’s check this part.

    Please check. Thank you.

  • Hi Grady,

    Thanks for your reply. 

    Please try to set R362 OPEN to disable the PPD.

    Besides, have you tried to increase Q6's gate resistance? Increasing it to 20-50Ohm could also help to reduce the switching ripple.

    Best regards,

    Diang