• State Resolved
  • Locked Locked
  • Replies 4 replies
  • Subscribers 63 subscribers
  • Views 335 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.

LM22673 Whistling like a Tea Kettle

Anthony Rolando
Prodigy 90 points
Other Parts Discussed in Thread: LM22673

I have a board where I am using two LM22673 Simple Switchers. One is configured as a non-inverting regulator generating +12V at 1.4A and the other is configured as an inverting regulator generating -12V at 800mA. Both are fed from same +15VDC 4A regulated power supply. I used the Web Bench application to help design the circuit. 

The prototypes worked perfectly for us. We ran 300 units. and about 20% of them exhibit a very audible whistling when tested under a load. The whistling sounds exactly like a tea kettle. It is possible to hear this sound coming right off the board. When we use the PCB to power our device, the whistling sound is coupled into the signal path, so it is a problem. 

My guess is that this whistling sound is a beat frequency of the two switching frequencies. Since there is a tolerance of 10% in the switching frequency on the LM22673 (I think, I cannot remember exactly), perhaps there is some combinations of two switching frequencies that results in a beat frequency that is in the audible range. 

Any help will be greatly appreciated.

Tony

  • 0
    TI__Genius 16025 points

    Hi Tony,

    Your guess was right on! The difference in the switching frequency between the two IC's working together causes the ceramic capacitor on the input to behave as a loudspeaker. There is a good probability that the pitch would match said difference. Solution?

    Easiest way is to put an electrolytic in parallel with the ceramic caps you have on your board. Other options would include adding a little inductance to act as a filter on the input. If you use the inductor method, please add a little resistance to it so as to reduce the Q factor.

    Hope this solves your problem! Let me know if I can be of further assistance!

    Thanks,

    Anston

  • 0 in reply to Anston Lobo
    Prodigy 90 points

    SPent a bunch of time with 2 units that have the problem. 

    We already have a 47uF Electrolytic in parallel with the ceramics at the input of the inverting regulator. The inverting regulator is where the tea kettle noise is coming from and overall it is noisier. 

    Just to give it a try, I removed the ceramic caps, one by one, and where I could I replaced them with electrolytics. I understand this is not an ideal solution to the problem, but what I was trying to do was to figure out which caps were singing. Even with all the ceramic caps on the input swapped for electrolytics, I was still getting the tea kettle. So I removed the ceramics at the output. No improvement. The last ceramic in the circuit at this point is the 10nf bootstrap supply cap. Is it possible this cap is the culprit?

    I even cut in an inductor in series before the caps at the input of the LM22673

    I have ordered some 10nf C0G caps to swap into the circuit. IN the mean time, is there any other suggestions you could offer? 

    Thank you,

    Tony

  • 0 in reply to Anthony Rolando
    Prodigy 90 points

    Another idea, is there a 2267X chip that operates at a different switching frequency? Something we could drop into this board and try?

  • 0 in reply to Anthony Rolando
    TI__Genius 16025 points

    Hi Anthony,

    Did the C0G caps solve the problem? I think the inductor could also be the singing culprit. If your circuit has Large current pulses it could contribute to the issue as well. Can you share with us your schematic and layout and also mention what load conditions the part is operating at? If you have any reservations sharing this information, you could always drop me a message on A%n%s%t%o%n%[at]ti.com [remove %] [security precaution to minimize spam]

    Thanks,

    Anston