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.

Absence of decoupling capacitors in LMK03806's EVAL Board

Hi,

I can't find decoupling capacitors in LMK03806's EVAL Board. Why?

http://www.ti.com/lit/ug/snau075a/snau075a.pdf

www.ti.com
��0�/�4$d�N$%%�6H!�>%�F�.e�Dԉ��B�N���9�藈��u�XO����/M����ᲫIJ����'*?� � *���z��o�!�� �� A��SV9"?�`�II�^�V���_ck�2�t���������+NV�ye�*3� P���� @���u���:tn�Dƍ伉Nd��!QX����]!F�΀9:��Ҡ�2�����-��Xx�� �������f��� 3X��7�__�1-w��DZ�L/�W��+��|��e[5�����,9���D��...

For example, in the datasheet, i found that core current is 122 mA

In the EVAL Board, Vcc_core comes from the regulator via a ferrite bead which has 120 ohm at 100 MHz. The inductance of power supply is 120/(2*pi*0.1 GHz)=191 nH.

Cdecoupling must be greater than Power supply inductance/(impedance target)^2

Ztarget=voltage ripple/core current=300 mV/122 mA

finally, Cdecoupling= 32 nF.

The minimum number of capacitors n is equal to 2*Lparasitic/(Zt*trise).

trise in lmk03806 is 200 ps.

Is this approach of estimating decoupling capacitors incorrect?

If  decoupling capacitors are not necessary as it is the case in LMK03806 EVAL Board, can someone, please, give me an explanation.

I have to make a choice this day between lmk03806 and cdce72010 (in ADS41XX EVAL BOARD). And, both, has stranger decoupling capacitor values which don't match with values of currents specified in datasheet. Please, can someone, give me an explanation if my estimation was wrong. I have an urgent need for explanation today.

Applying guidlines in this document, www.ti.com/.../scaa048.pdf 
I can't find the same values of decoupling capacitors for lmk03806 and cdce72010. I need explanation please; an example of calcul if possible. This is very urgent.

Thank you.

  • The EVM decoupling/bypassing caps on the EVM was based on having a clean supply from the onboard LDO regulator (without voltage ripple).  There are 10uF and 1uF decoupling caps (C43, C44) on VccCore.  However, I don't know the exact method for selecting the bypass caps on these EVMs.  It could be been chosen bsaed on emperical test results, and under a some specific supply and device test conditions.

    Note that decoupling and bypassing cap values are not one-size-fits-all, and can be different depending your power delivery network design.  I think the approach in the app note you referenced can be applied.

    Regards,
    Alan

  • Hi Alan,
    Thank you very much for your response.

    Best Regards,
    Mourad