Looking for info on how and when the following should be used. Sorry if this question has been asked but after days of searching and reading I feel I'm still missing something.
1. White Noise.
2. Pink Noise
3. Sine wave
4. Square wave
5. Saw wave
6. Frequency sweep.
The basic question is how would a person who is deaf. Test his audio gear, From Power supply to pre amp to tone control to amp to speaker. Using nothing more than a scope and a function generator. (China based or Phone / PC based.).
1. White Noise.
2. Pink Noise
3. Sine wave
4. Square wave
5. Saw wave
6. Frequency sweep.
The basic question is how would a person who is deaf. Test his audio gear, From Power supply to pre amp to tone control to amp to speaker. Using nothing more than a scope and a function generator. (China based or Phone / PC based.).
I'd think REW generates most of them. It of course does the analysis too. I'm pretty sure pink noise corresponds to nth octave analysis and white noise goes with a "straight" FFT. Square waves go hand in hand with a scope display, which REW has. The bandwidth of which depends on what you have for an audio ADC. I'd say learn about REW and its capabilities; that would cover a majority of what you're looking to do.
Maybe there should be an REW sticky, if there isnt already one in place.
Maybe there should be an REW sticky, if there isnt already one in place.
Measurements very depends of what do you want to test/measure.
For many things "Using nothing more than a scope and a function generator" is not enough. Especially for the digital equipment.
With a simple scope and functional generator the tests will be more "go/no go" (or "works/failed").
You need at east very good sound card and the software like RMAA to provide better tests of existing devices, and much more complicated equipment to develop the new ones.
For many things "Using nothing more than a scope and a function generator" is not enough. Especially for the digital equipment.
With a simple scope and functional generator the tests will be more "go/no go" (or "works/failed").
You need at east very good sound card and the software like RMAA to provide better tests of existing devices, and much more complicated equipment to develop the new ones.
Thanks for the REW tip. But there are tons of android and iphone apps that do all this. How to do it is the least of my worries.
How to use the data is.
Recently I learned that Square waves are good to test tone control boards.
Than white noise is good to also test tone control boards as well as to test parametric equalizers as well as bass boost circuits.
Sine waves are good to look for PSU noise.
Square waves are good to test opamps.
But there is still a lot more I dont know. It took me months of playing with a tone control board before I learned these things.
Like today I need to test if my PSU would benefit from a snubber circuit or not.
How to use the data is.
Recently I learned that Square waves are good to test tone control boards.
Than white noise is good to also test tone control boards as well as to test parametric equalizers as well as bass boost circuits.
Sine waves are good to look for PSU noise.
Square waves are good to test opamps.
But there is still a lot more I dont know. It took me months of playing with a tone control board before I learned these things.
Like today I need to test if my PSU would benefit from a snubber circuit or not.
To use White/Pink Noise, and also for the PSU testing - you do not need a scope, you need Spectrum Analyzer.
REW is a good choice.
Also for PC the good (free) choice is WaveSpectra (and her brother - WaveGene, as signal generator).
REW is a good choice.
Also for PC the good (free) choice is WaveSpectra (and her brother - WaveGene, as signal generator).
It has been said that it is easier to hear resonances when you can hear more of the spectrum at once. This is the reason that speakers are sometimes demonstrated using solo music or light electronic music.
As a result, pink noise is sometimes used as a simple way to test for the preferred polarity of tweeters, or for tonal balance.
As a result, pink noise is sometimes used as a simple way to test for the preferred polarity of tweeters, or for tonal balance.
Tons of apps do that.. but simply to be an app. Out of those apps most have then got either limitations on FFT control (windowing), can't make a reference, use proven maths or support higher than 16bit.. REW would be my go to (although it costs for 2 channels - only wish it could do bode (frequency sweep response using two channels) without needing to pay 😀
THD - this is mean to both the signal generator and the analyser.. I have plans for a -140dB 1Khz oscillator to test for THD as the SDG has more noise.
As it was already stated your choice of source will depend of what you try to troubleshoot.
Rather than giving a list of case i think it's better to understand what those signals are.
It should help clarify some possible use...
What is a white noise? A signal in which there is all possible frequency playing at once with equal intensity. It's a signal with constant power spectral density.
What is a pink noise? A 1/f signal ( other name it have) with inversely proportional power spectral density (with reference to frequency). In other words it's a filtered white noise... with more weight on the low end part of spectrum.
What is a sine wave: it's a smooth periodic function whose graphical representation is a sine wave. In audio it relate to the fundemental of a sounds. It's the 'purest' signal as it doesn't display harmonics ( theorically).
Etc,etc,...
Take a look at definition it'll help.
I'll gives you some shortcuts though: a triangle contains only even harmonics+ fundemental., square only odds harmonics ( plus fundemental) sawtooth a mix of 50% of both harmonics+ fundemental ( not convinced? Take a a look at a square and triangle waves and then sawtooth...).
From this def you'll already have an idea of what they could show use or be of interest. There graphical representation could help imagining what they could be good at: a sine won't easily show a ringing if it occurs (as it doesn't have sharp edge). A square would be of better use in this case as if ringing occur there is a 'mute' in the cycle and anything appearing during it is probably unwanted...
An eq is applied to a frequency 'area' or interval. A static sine could give info on a given freq but not an interval: a sweep in the other hand will run through an interval... a white or pink noise would be useful too in this case,...
If you want to test something 'dynamic' ( with variation with amplitude) then none of the signal will be of interest without modifying their length: with one cycle of a square or sawtooth however and if you can adjust amplitude to compare between them...
Well, You got the idea... grossly simplified but..
Last edited:
op, if you are able to make frequency response tests - what is the problem in using the resulting curves to verify correct operation of circuits under test?
🤔
With respect to Joeseph Giovanelli, profoundly deaf people probably do not own audio gear.
However that chain you propose, NObody can hear the sound inside the amplfier boxes, only at the speaker. And several types of audio systems work without routine speaker-monitoring (radio network feeds, security recordings, now court audio records made and never requested because everybody agreed).
Think about what these signals may be good for. Single sine tells one frequency steady-state, and simple harmonic distortion. Sweep tells all frequencies sequentially. Hiss tells all frequencies all the time. White-noise hurts ears and tweeters, pin is kinder. 2-tone is simple intermodulation distortion. What do you want to know?
For someone with a pile of tone-controls that don't work: simple single site tells if sound will pass. Sweep-sine is traditional for plotting response but Noise with FFT gives a good peek faster.
I wish that someone had told me that it is NOT a good idea to test the limit of my high frequency hearing with speakers and foobar2000's frequency generator.
Set the thing to 20kHz. "Huh? I'm not hearing anything".
Set it to 19kHz. "Oh, there it is. Let me turn it up a little, the level seems quite low."
A titanium dome tweeter died that day.
I did this over 15 years ago, when I was an ignorant teenager. Since then I've grown into an ignorant adult but I have not blown up any more tweeters.
Set the thing to 20kHz. "Huh? I'm not hearing anything".
Set it to 19kHz. "Oh, there it is. Let me turn it up a little, the level seems quite low."
A titanium dome tweeter died that day.
I did this over 15 years ago, when I was an ignorant teenager. Since then I've grown into an ignorant adult but I have not blown up any more tweeters.