• State Resolved
  • Locked Locked
  • Replies 4 replies
  • Subscribers 62 subscribers
  • Views 456 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

TPS54394 AM frequency interference

Eng Lih Khaw
Expert 1990 points
Other Parts Discussed in Thread: TPS54394, TPS54395

Hi,

Customer of mine is evaluating TPS54394 for mini compo product. Per my understanding, TPS54394 has fixed switching frequency of 700kHz. This would cause TPS54394 to interfere with AM frequency. Attached please find the AM beat measurement being carried out by customer. The requirement is not to exceed 68dB.

Vin = 13V

Vout1 = 5.2V; Iout1 = 2A

Vout2 = 3.3V; Iout2 = 1A

Could you advise the following:

- Is there any way to adjust the switching frequency lower than 700KHz to avoid interference with AM frequency?

- if it’s not possible to adjust the switching frequency, is there any other counter measure to improve the noise?

- In datasheet (pg1), I noticed that the GND and PGND pin being grounded differently. However in EVM, both pins are grounded at the same plane. Could you advise which I should follow?

Thanks in advance!

  • 0
    TI__Guru**** 191445 points

    The TPS54394 does not have an internal oscillator.  It uses DCAP2 control with adaptive on time based on Vout and Vin and hysteretic off time.  Also the two channels are independent and are not sychronizeable.  The switching frequency is not adjustable.  And the actual operating frequency will vary with line and load conditions.

    Regarding your data, I do not see any correlation.  It seems somewhat random.  I would expect to see a large peak at the switching frequency, then additional peaks at the harmonics (1400 kHz, 2100 kHz, etc).  Can you tell me how the data was measured?  If you can explain the test method, I may be able to suggest something.

    As far as ground, the PGND1 and PGND2 pins (4 and 13) are the main power grounds.  They are connected together directly under the IC and are connected with multiple vias to ground copper on the internal and bottom layers.  GND is analog ground (pin 8).  It is connected to the sensitive analog returns for VFB and SS (TPS54395).  This small island of ground is connected at a single point to the main power ground plane near pin 8.  Single point connection prevents unwanted current circulation.

  • 0 in reply to JohnTucker
    Expert 1990 points

    Hi,

    Thanks for your fruitful information and pardon me for delay

    I've attached the test method showing how my customer carry out the TPS54394 evaluation. Please help to review and let me know if you need further clarification.

    Thank you!

    6406.TPS54394_AM_Fsw_issue_150313.pdf

  • 0 in reply to Eng Lih Khaw
    TI__Guru**** 191445 points

    You show a block diagram with "Tuner" an "D-amp" but do not show any power supply circuit.  So it looks like what you are doing is broadcasting an AM carrier signal into an AM tuner and then measuring the relative AM frequency signal at the amplifier  output.  Why would you think it is related to the power supply rather than some other leakage path thru the tuner or amplifier?

  • 0 in reply to JohnTucker
    Expert 1990 points

    Hi,

    My aplogoyy for the confusion made, customer actually took off their existing DCDC (from other competitor) from the audio set and powered up the amplifier, tuner (also other circuitry) with TPS54394 then measure the AM noise level

    I’ve updated the diagram with better illustration as well as the existing DCDC result. This should be clearly shows that TPS54394 was having interference with AM frequencies.

    3051.TPS54394_AM_Fsw_issue_revised.pdf