• State Resolved
  • Locked Locked
  • Replies 1 reply
  • Subscribers 49 subscribers
  • Views 1142 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.

LMH6629: Signal Chain Noise Analysis

xyz xyz
Intellectual 1000 points
Part Number: LMH6629
Other Parts Discussed in Thread: LMH6552

Hello

I'm sorry to post this question!

I've studied some TI's document about the noise analysis but in my final design I have some problem.

the circuit is like below:

the noise spectral density at point 1 is 4nv/rtHz

I think the noise spectral density at point 2 is:

en2=A*sqrt(en_LMH^2+en1^2)=10*(0.65^2+4^2)=40.52 nV/rtHz

But what happen to the noise spectral density of 50 Ohm resistors and impedance matching?!

and on the other hand, in a document I've studied that the SNR of driver+filter is:

SNR_af=20*log10(Vrms/(En_driver*sqrt(1.7*fc)))

and

SNR_system=-20*log10(10^-SNR_adc/10+10^-SNR_af/10)

now How should connect this concept to my design?

is there such concept for calculating harmonic distortion in a signal chain? 

Please help me as fast as you can. I'm gonna order them from digikey.com

Regards

  • 0
    TI__Genius 15975 points

    Hi,

    Sorry for the delay in reply.

    The total output noise at point 2 (Eo2) will be 10.82nV/rtHz at point 2 in your figure which accounts for the 50-ohm resistor impedance matching. This is calculated by adding the 4kTR noise of the 50-ohm series resistor to the LMH6629 cascaded output noise (Eo_Amp), and then multiplying by 0.5 due to 50-ohm impedance matching between the 2nd and 3rd stage. This is shown in Figure 1 which I have modified to show the noise contribution at each stage.

    The LMH6629 output noise (Eo_LMH6629) output noise is calculated by the expression provided in Figure 2. In your circuit, the Rs is 0-ohms because it is filtered by the 100nF cap. If you just try to calculate the output noise contribution due to the LMH6629, it results in 8.25nV/rtHz using the equation given in Fig 2. For the LMH6629 device, the Eni = 0.69nV/rtHz and in = ip = 2.7 pA/rtHz.

    In a similar fashion, you should be able to calculate the total output noise at the LMH6552. You can refer to this App-note for the output noise calculation of an FDA such as the LMH6552:

    Once you have calculated the total output noise due to the driver, you should be able to back calculate the ADC noise from the ADC SNR value and add it to the SNR equation as below.

    SNR_system = 20*log10(ADC_FullScale (rms)/sqrt(En_driver^2 + En_ADC^2))

    I don't think there is a similar concept for calculating the harmonic distortion in the signal chain because the harmonics are never in-phase between stages which makes it difficult to predict the harmonic distortion.

    Figure 1:

    Figure 2: Op-Amp Output Noise Calculation


    I hope this helps understand things better.

    Best Regards,

    Rohit