1. Blog>
  2. Calibration of Spectrum Analyzers

Calibration of Spectrum Analyzers

by: Apr 29,2022 3014 Views 0 Comments Posted in Technology

Spectrum Analyzer Calibration pink noise white noise brown noise

Summary:       This article will explain how Spectrum analyzers are calibrated. It explains the use of different kinds of noise and the way human perceive it.

Calibration of Spectrum Analyzers


Noise patterns

First let met define noise: An endless bunch of different frequencies with random amplitudes. By changing the spectrum of the noise spectrum, we can ‘color’ it.

White Noise

This is the most common and well-known form of noise. White noise is comprised of an equal amount of every frequency, all played at once. It happens in real life when your TV or radio makes static noise. However, because of our ears’ frequency response, we perceive higher frequencies as louder, so it doesn’t sound particularly balanced to most people.


Pink Noise

Pink noise is essentially bass-boosted white noise, where the amplitude decreases as the frequencies get higher. This is the perceptually ‘balanced’ white noise that many of us are comfortable with.


Brown Noise

Brown noise boosts bass a bit more then pink noise, just to further warm things up.


Blue Noise

Blue noise is, the opposite of pink noise. It’s treble-boosted, and the frequencies are so high that most of them are barely perceptible to the human ear.


Violet Noise

This is blue noise with even more treble boost


Gray Noise

It’s basically pink noise that has been adjusted to a individual. Tailor fitted to your ear, so to speak.


(Source of above noise information was derived from : https://splice.com/blog/difference-bewteen-noise/ )


Daily noise in the environment

To give you an idea about the spectrum of some daily noises in your environment, take a look at the following graph:


How this effects the spectrum analyzer and its calibration

When you look at the spectrum of white noise you can see that it is a straight forward noise pattern with similar amplitude for each frequency. In short, it is pretty balanced. However, this is not how the human air perceives it. Because of our ears’ frequency response, we perceive higher frequencies as louder, so it does not sound particularly balanced to most people. This is where all the color patterns fall into place. We can trick the human air to perceive the sound spectrum differently. Ideally, gray noise should be used for calibration because it would be perfect for your ear. However, this is also a problem because it would be ideal to your ear only. Therefore, if we would calibrate audio amplifier or spectrum analyzer etc. to a gray noise source, the outcome would be perfect to one specific person but it might sound awful to all others. This is not the outcome we want for the calibration of a spectrum analyzer. However, using some other color patterns like pink or brown noise in addition to white, will give us enough variation to calibrate the spectrum analyzer to your likings.


Calibration method

Calibration of your spectrum analyzer is needed to compensate for your hardware’s frequency response. Your microphone might have a good response for frequencies between 300 and 3400 HZ but below and above that narrow band, the response will be a lot less. This all depends on the quality of the microphone. Another example is your pre-amplifier or even your input connector and cables. Each item has a frequency response that will influence your total setup. This might be noticeable by a lessor response at certain frequencies.

The basic way of calibrating the spectrum analyzer is to hook it up to a white noise source. Ideally, all frequency bands will have the same average amplitude. Because of what I just explained, your analyzer will not respond ideally. This is where the calibration kicks in. With the white noise source connected, you can now compensate each frequency channel by a multiplication factor that brings it up to the level you want. You can choose to scale it if you know the amplitude of the input source. For example, if you use a 1db input signal, you can now define a scale on your analyzer of 1db and you could do the same for different levels. Many spectrum analyzers are only used as visual enhancers and for those version, I scale is not as important but that is all up to you.

Now back to our calibration. Instead of using white noise, we could also use a pink or brown noise as source for our calibration. The calibration method is the same but this time we are tricking our analyzer as it does not know that our input noise is no longer ‘balanced’ it will compensate more on the frequency bands that has a lower input amplitude. After calibration, those specific channels have a extra amplification and as a result, those specific bands will show a better response.

 

 

ing. Mark Donners 29-april-2022

www.theelectronicengineer.nl

For diy manual of Spectrum Analyzers, check out my youtube channel:

www.youtube.com/c/TheElectronicEngineer


Join us
Wanna be a dedicated PCBWay writer? We definately look forward to having you with us.
  • Comments(0)
You can only upload 1 files in total. Each file cannot exceed 2MB. Supports JPG, JPEG, GIF, PNG, BMP
0 / 10000
    Back to top