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.

LMK04808: SNR degrades in ADC due to PLL clock amplitude changes

Part Number: LMK04808
Other Parts Discussed in Thread: ADC16DV160

Hi,

PLL(PART NO :LMK04808B) is tested with the following configurations.

Here PLL CLKout0 is used as source clock for ADC1(PART NO :ADC16DV160), PLL CLKout2 drives ADC2 of the same part.
ADC's sampling clock is same frequency as that of source clock.
(i)As per our requirement ADC1 is configured for 160 MHZ and ADC2 is configured for 144MHZ (Refer 3rd test case).
With the above configuration SNR degrades, in comparison with other frequency combination test cases.


(ii) With our default configuration(ADC1-160M, ADC2-144M), clock amplitude was reduced to 1200mVpp in ADC2 and tested(Refer 7th test case),
it was observed that ADC1 SNR is better compared to test case-3.

(iii) With our default configuration, clock type was changed to LVDS in ADC2 and tested(Refer 8th test case),
it was observed that ADC1 SNR is better compared to test case-3.

(iv) To idenitfy the source of problem to be ADC or PLL, ADC2 chip was power down and tested, even with it ADC1 SNR degrades with default configuration.
Then, we power downed clkout2, here ADC1 SNR improves(Refer 9th test case)

Test case Configuration CLKout0(ADC1) CLKout2(ADC2)
  Frequency Variation
1 Frequency 160 MHZ 160 MHZ
Type LVPECL, 2000mVp LVPECL, 2000mVp
SNR Observed 57.01 55.14
2 Frequency 144 MHZ 144 MHZ
Type LVPECL, 2000mVp LVPECL, 2000mVp
SNR Observed 55.65 56.7
3 Frequency 160 MHZ 144 MHZ
Type LVPECL, 2000mVp LVPECL, 2000mVp
SNR Observed 51.84 55.87
4 Frequency 144 MHZ 160 MHZ
Type LVPECL, 2000mVp LVPECL, 2000mVp
SNR Observed 53.93 54.51
5 Frequency 160MHz 120 MHZ
Type LVPECL, 2000mVp LVPECL, 2000mVp
SNR Observed 59.54
6 Frequency 160MHz 96 MHZ
Type LVPECL, 2000mVp LVPECL, 2000mVp
SNR Observed 56.74
Clock amplitude variation in Clockout2
7 Frequency 160MHz 144 MHZ
Type LVPECL, 2000mVp LVPECL, 1200mVp
SNR Observed 54.88
Clock type Variation in Clockout2
8 Frequency 160MHz 144 MHZ
Type LVPECL, 2000mVp LVDS
SNR Observed 56.8 54.44
Clockout2 Power down
9 Frequency 160MHz Power down
Type LVPECL, 2000mVp
SNR Observed 56.08


why ADC1 chip SNR varies with Clock frequency, amplitude and Clock type of ADC2?

  • Hi Auxilian
    I'm not sure the reason for the behavior you are seeing.
    My first theory is that noise from CLKOut2 could be coupling into the the adjacent LDO Bypass capacitor connections and adding jitter into the clock seen at CLKOut0 for ADC1.
    Can you try adding capacitance to the LDO Bypass points, or shielding them from CLKOut2 to see if that has any effect?
    Best regards,
    Jim B