1. PF Contest - Win "Conquering the Physics GRE" book! Click Here to Enter
    Dismiss Notice
Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Calculating power over a frequency band

  1. Mar 17, 2010 #1
    I am interested in taking a measurement with a Spectrum Analyzer from an amplifier circuit I'm putting together and am interested in calculating the total power over a certain bandwidth. Is is just a matter of taking all the readings in dBm, converting to watts and then taking the RMS values taken from the Spectrum Analyzer across my band of interest? The only power in to the amplifier will be the noise power. The amplifier has 45 dB gain.

    I understand some Spectrum Analyzers have an option that allows you to measure this directly, but the Spectrum Analyzer I have access to does not have this functio

  2. jcsd
  3. Mar 17, 2010 #2
    I will guess that the noise power is about -111 dBm per MHz**., including 3 dB noise figure, and 45 dB gain to get -66 dBm per MHz. You don't need to convert to watts. Just use dBm.

    Bob S

    ** the noise power is kTB where k= 1.38 x 10-20 millijoules per deg kelvin, T=293 kelvin, and B(bandwidth in Hz)= 1 MHz

    So noise power is 1.38 x 10-20 x 293 x 106 = 4 x 10-12milliwatts per MHz = -114 dBm per MHz.

    Add 3 dB noise figure to get -111 dBm per MHz
  4. Mar 18, 2010 #3
    Bob - thanks for the reply. That's noise power, but once I hook up the amplifier, I'll get some trace that represents the output noise power versus frequency. How do I calculate total power over that bandwidth? Say I have the following readings versus F1 - F2:

    -55.6 dBm, -55.7 dBm, -56.2 dBm, -57.2dBm, -57.8 dBm, -56.9 dBm.

    I guess thr question I have is how I integrate (think that's the right term) power over that bandwidth?

  5. Mar 18, 2010 #4
    Do you know the frequency resolution of the analyzer? Presumably, values of -55.6 dBm (i.e. 2.75 nW), etc. correspond to the energy contained within the window around the frequency to which the analyzer is tuned. The size of the window is either adjustable or can be found in the manual.
  6. Mar 18, 2010 #5
    hamster143 - I want to get the Noise Floor of the Spectrum Analyzer down to -65 dBm or lower, and I planned on setting the Resolution Bandwidth (RBW) and the Video Bandwidth (VBW) to 30 KHz.
  7. Mar 18, 2010 #6
    If you back-terminate the input to the spectrum analyzer with a good matched resistor, the noise level should be in the range -114 to -111 dBm per MHz bandwidth (not per MHz resolution). See my post #2.

    Bob S
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook