• State Resolved
  • Locked Locked
  • Replies 3 replies
  • Subscribers 49 subscribers
  • Views 582 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 Noise Figure Calculations

Vinit Dhulla
Prodigy 20 points
Other Parts Discussed in Thread: LMH6629

en

Noise voltage

Input referred f > 1MHz

0.69

nV/√Hz

in

Noise current

2.6

pA/√Hz

NF

Noise figure

RS = RT = 50 Ω

8.0

dB

 Above numbers are from the LMH 6629 datasheet. 

I am just trying to calculate the noise figure and arrive to the same NF number published in the datasheet. 

It seems like I should be able to combine the input-referred voltage and current noise, given a 50-Ohm source, to come up with a total input-referred voltage noise.  And I ought to be able to then calculate the amplifier’s noise figure.  However, my numbers don’t seem to work out when I attempt this.

Per my calculations, the total input-referred voltage noise comes to 0.7nV/Rt Hz

Then I use this formula: 

NF = 10log(1+NA/Ni)

For NA is use the calculated input-referred voltage value of 0.7 and for Ni I use 0.91 nV/RtHz (50 Ohm source thermal noise)

Thank you in advance for the help.  

  • 0
    TI__Mastermind 22825 points

    Hello Vinit,

    Some comments:

    1. NA is the square of the total input referred voltage noise (with Rs in place but assumed noiseless). This includes the LMH6629 noise voltage, noise current flowing through the equivalent source resistance Rseq (25ohm assuming input 50ohm termination), noise current flowing through parallel combination of RF and RG (RF || RG) , thermal noise of Rseq, and thermal noise of RF || RG. I used Equation 3 of the datasheet to compute this to be (1.14nV)^2.

    2. NI is the square of the source resistance noise (0.91nV/RtHz) that actually gets to the LMH6629 input. For the datasheet condition (Av= 10V/V, input terminated in 50ohm for matching), that would be [(0.91nV/RtHz) / 2 ]^2, due to the fact that only 1/2 the source noise voltage reaches the input. This is (0.455nV)^2.

    So, the Noise Figure computes to:

    NF= 10*log (1+ NA / Ni) = 10*log (1+ (1.14nV/ 0.455nV) ^2) = 8.6dB

    So, the datasheet number of 8dB is probably based on actual measurement and it' is a little too optimistic compared to calculation!

    Regards,

    Hooman

  • 0 in reply to Hooman Hashemi
    Prodigy 20 points

    Hi Hooman -- thanks a lot. 

    Best,

    --V

  • 0 in reply to Hooman Hashemi
    TI__Genius 13935 points
    Hello Vinit,

    Hooman is correct. The voltage noise and current noise were measured on a spectrum analyzer. The noise figure was measured on a noise figure meter. So, both are actual measurements. It is not unusual for the two instruments to give slightly different results.

    Regards,
    Loren