Sound Quality measuring question

[smithce]

For my science fair experiment i am testing if the price of a stand up or bass vial affects it sound quality. What areas of sound should i test?
Any guidance is welcome
Thanks!

[kgudger]

Hello and Welcome to the Forum!

I used a search engine to look for "measuring sound quality" and found a number of pages that describe ways to measure sound quality. If you're looking for a quantitative measurement of the sound quality for several of one type of instrument, I have some suggestions. You can use a computer to look at the waveforms of a specific note from each instrument. You will probably need a good microphone and a good sound card to be able to do this. Once you capture specific notes from each instrument, you can use a sound program (like Audacity, which is free) to analyze the waveform (and plot the spectrum.) (Control for as many variables as possible, such as always using the same bow, the same type and age of string, etc.)

Another way to "measure" sound quality is with a "jury". This is less quantitative and more qualitative, but might be what you want. Participants should not know which instrument they are listening to.

Let us know if you have more questions.
Keith

[smithce]

Thanks! Do you think a spectrum analyzer would work, or would that be the wrong sort of thing?

Thanks again!

[kgudger]

Hi:

Yes, an audio spectrum analyzer should work.

Keith

[smithce]

Another question, would THD work?

Thanks so much!

[gannonad]

Hi. I am doing a similar project although I am measuring the tightness of a bow's effect on a violin. Would the same factors apply to my project? I was planning on using Audacity (like you said) to graph the sound waves. I already have a microphone and computer to do this. I just need to figure out how I want to compare the sound waves. Any advice? Thanks so much!

[Craig_Bridge]

Another question, would THD work?

THD is one measurement of sound reproduction/amplification quality. Unfortunately, harmonic distortion is also an aspect of the distinctive tone some instruments make. In other words, if the original sound has a specific harmonic distortion, a high quality recording / reproduction will maintain that tonality without further distortion. THD measurements are typically done on electronic equipment using pure sine waves of various frequencies as an input and looking for any distortion in the output. This technique isn't viable for measuring or comparing tonal quality of different instruments.

I am doing a similar project although I am measuring the tightness of a bow's effect on a violin.

As someone who plays a violin, the tightness of the bow will be a difficult independent variable to work with. If the bow hair is loose, you can't apply as much normal force without having the wood make contact with the bow hair or string and there will be considerable changes from frog to tip. Once bow tension reaches a certain point, the stick will start deforming or warping, often out of the plane between frog and tip. In other words, bow tension changes affect lots of other variables in the way other forces come in to play (puns intended).

[smithce]

Thanks! Do you recommend anything that might work better? I was thinking of using Garage band on a mac. Would this be something I could use or would you recommend something better?

[robertreavis]

I want to take a different view of your question.

Sound quality of what? The microphone, amplifier, and speaker system or of the pleasing sound of a particular instrument?

A few examples: An accoustical guitar and an electric guitar with no amplifiers sound almost the same. What makes the electric guitar's distinctive sound is intentional distortions intruduced by an highly non-linear amplifier or a fuzz box. Without distortion you would probably not like the electric guitar.

Consider a horn, a french horn and a trumpet can produce the same note (same fundamental frequency) but they sound very different because of harmonic content. What is distorted?

A long time ago I attended a lecture by the head of the Stanford University Music Department. The topic was electronic music. One point he made is that a pure note (an undistorted sine wave) is easy to produce electronically and sounds about the same as a tuning fork for a simple plucked string. But(!) that is not music (he said). Music also includes the distortions, variations in tempo, attack, sustain, and decay of notes, squeeks of chairs, clatter of feet, fingers sliding on strings, fingernails on the keys, and all that stuff beyond pure tones. Leave that out an it sounds dead.

So, what do you want to measure?

A bass viola makes a sound, do you like it? Record, save, playback, amplify it, run the electrical signal into a speaker or earphones, and so on. Do you still like it? Is it better or worse? Speakers and earphones sound very different.

Another aspect you can measure is the effect of the environment. An open-air bandstand in the park sounds a lot different than a hard-surfaced room. The addition of echos adds another layer of distortion. Do you still like it?

Sound quality is what you call it. Once you define it, then you can measure it.

[smithce]

for my project I was wondering how I could measure sound quality... without bringing in a jury to listen and rate what they think sounds better. I was testing if a more expensive bass would sound better than a cheaper bass.
Anything you recommended will be greatly appreciated!


Thanks!

[gannonad]

Thanks Craig. I understand doing this project will be difficult. I am planning to test the experiment soon to figure out what I need to change. Do you have any suggestions on how I could change my project a little bit or keep the variables under better control? Also, I still am confused on how to compare the sound waves.

Here is my procedure for the experiment. I thought it might help...
1. Loosen bow hairs until the screw comes completely out.
2. With the bow screw out, mark one of the vertices on the hexagonal shaped screw.
3. Holding the bow with bow hairs facing up, insert the screw with the marked vertice (also facing up) and turn one complete rotation so the marked part of the bow is in the original position.
4. Record all open stringed notes: G, D, A, and E, with the bow at this tightness.
5. Turn screw one complete rotation.
6. Repeat steps 4 and 5 until you have recorded 15 times, each with different bow tightnesses.

[smithce]

My project is very similar but instead of the stuff with a bow, I would switch out the basses (After playing the 4 strings).

Thanks!

[Craig_Bridge]

gannonad,
How are you going to control the normal (or contact) force that the bow hairs make with the string?
How are you going to control the number of bow hairs that make contact with the string?
How are you going to control the width of the contact area?
How are you going to control the speed of the bow hair movement?
I repeat, it is going to be difficult to eliminate all the factors that are affected by bow tension. A person playing the instrument will naturally compensate for different bow tensions. This is practiced behavior because the tortional force required at different bow positions to sustain a note throughout the bow travel differs. An experienced musician will naturally change things to get the sound and maintain the sound that they expect/desire.