A Spectrum Analysis of Your System

[musicfirst1]

What should a spectrum analysis of white noise played through one's system taken from the main listening chair look like?

 

[Robh3606]

What should a spectrum analysis of white noise played through one's system taken from the main listening chair look like?

Pink would be flat octave to octave and white should be rising I believe. Why are you using white noise? Assuming a flat frequency response

Rob

 

[musicfirst1]

Not assuming anything here, just trying to understand what various noise types played through one's system result in various spectrums.

 

[Bruce B]

Or you can just do a freq. sweep

 

[Robh3606]

Not assuming anything here, just trying to understand what various noise types played through one's system result in various spectrums.

What are you trying to do or are you just curious. If you were say setting up a system using say a traditional graphic equalizer and an RTA and your target was flat in room response you would use pink noise which has the same power in each octave. What are you using to see the in-room response?? Do you have a measurement system like CLIO as an example?

If you use white noise it is linear as far as power per bandwidth in hertz so it depends on what you are using to see the response. That's why it is rising on say a traditional RTA as the bandwidth doubles per octave.

Rob

 

[musicfirst1]

Just Curious Rob to see what I would expect on a basic RTA app on my phone. It appears to be weighted bottom to top for real music, but I want to try some test tracks of white, pink, brown etc.

 

[Mark Seaton]

Just Curious Rob to see what I would expect on a basic RTA app on my phone. It appears to be weighted bottom to top for real music, but I want to try some test tracks of white, pink, brown etc.

If using an RTA, pink noise provides equal energy per octave, and would display flat from a speaker that would measure flat in a frequency response sweep. When you use the term spectrum analysis some might think you are referring to an FFT of the spectrum which would be downward sloping from pink noise, and flat from white noise.

 

[microstrip]

Just Curious Rob to see what I would expect on a basic RTA app on my phone. It appears to be weighted bottom to top for real music, but I want to try some test tracks of white, pink, brown etc.

If you just curious, fine. But using such rough test to make any changes or take conclusions concerning your system would be very misleading.

A calibrated microphone and a good USB sound card will cost less than $300 - used with REW or Spectraplus they are the minimum needed to make decent measurements.

 

[sbnx]

Just a note. If you are using a phone app to show you the spectrum as you listen to music just be aware that you should expect a drop off at higher frequencies as music has less energy at higher frequencies. Like others pointed out you can use pink noise to see if your overall frequency response is flat.

Be somewhat careful with playing white noise. It has a lot of energy being sent to the tweeter and could damage your tweeter if you are playing it for extended periods of time.

 

[DonH50]

Pink noise (equal energy per octave) is more commonly used for audio. Sounds more pleasant, works with most audio analyzers, and doesn't blow up as many tweeters. It will look "flat" on many audio spectrum analyzers but fall off on a "normal" spectrum analyzer at roughly 10 dB/decade IIRC.

White noise (equal energy per decade) will produce a rising response on a typical audio analyzer (set up for octaves instead of decades) and will sound overly "bright". It will look "flat" on a normal spectrum analyzer.

There are many other noise spectral densities ("colors"), of course. See e.g. https://en.wikipedia.org/wiki/Pink_noise

HTH - Don