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.

LM27762: At low levels of output current, it is generating strong spurious tones within the loop bandwidth of the PLL

Part Number: LM27762

Hi, 

we are using LM27762 to generate the negative power supply for some op-amps which are placed on a PCB along with a PLL. The output of the PLL shows very strong spurious tones caused by the (switching frequency of) LM27762 which lie within the loop bandwidth of the PLL i.e. 41 kHz. The total output load current is ~ 70 mA . 

After some initial discussion with the TI support team, the following solutions were tried: 

1. Reduce the value of the fly capacitance. ---> We reduced the value to the smallest allowed value i.e. 470 nF. It changed the frequency of the spurious tones from 41 kHz to 50 kHz which is still inside the loop bandwidth. Therefore, this method did not help to resolve the issue. 

2. The LM27762 switches (from PWM mode) to PFM mode for smaller loads i.e. to restore the PWM mode the load should be increased. --> We increased the load by adding additional resistors in parallel to the actual load to increase the output current of the LM27762 and it did move the spurious tones outside the loop bandwidth. However, this solutions comes with the drawback of increased power dissipation. Please, see the attached images to have a look at the problem and the measurement result of the recent experiment with the additional load. 

Are there any other solutions to supress the spurious tones or to push them outside the loop bandwidth with 70 mA of current. The switching frequency of the LM27762 needs to be set high at low current levels also i.e. we have to somehow avoid the PFM mode. 

Waiting for suggestions. 

Best regards, 

Abdul Rehman Javed

1. LM27762 without additional load 

2. measurement result without additional load showing high levels of spurious tones

3. LM27762 with additional load connected

4. Measurement results with additional load 

  • Hi Javed,

    I will look into your request and give you a feedback by tomorrow. Meanwhile coukld you please share the latest schematic and layout just to be sure of any possible issues there.

    I hope you will be fine with this. 

    Regards,

    Febin

  • Hi, 

    here are the files that you asked for. 

    Best regards, 

    Abdul Rehman Javed

  • Hi Javed,

    Thanks for all the details. I could not find a problem with the schematic and layout.

    Could you please share the Vin, +/-Vout, Cfly and CCp with load and no load for LM27762?  Could you confirm the input and output is stable and please check the ripple as well.

    Regards,

    Febin

  • Hi Febin, 

    what we understand is that with the 80 mA load the LM27762 goes to PFM mode and hence we get the spurious tones at 41 kHz. We experimented with different loads and found that with an additional 12.5 mA load the switching frequency increases to a high enough value that is outside the PLL loop bandwidth. This is not an optimum solution but something that we can live with for now. However, it is very important to separte the ground connection of the LM27762 and the global analog ground with a large inductor to ensure good phase noise at the outptu of the PLL. We are fine with the current situation but if you have any suggestions for improvement you can let me know. 

    Regards, 

    Abdul Rehman Javed

  • Hi Javed,

    Thanks for confirming!

    At the moment, it sound good. If you require any further help please reach out to us in the future!

    Regards,

    Febin