Measuring sound

Discussion in 'The Projects Forum' started by magnet18, Mar 7, 2011.

  1. magnet18

    Thread Starter Senior Member

    Dec 22, 2010
    1,232
    124
    Hey, does anyone know the best way to measure sound?
    I'm mainly concerned with volume. It would be a fairly quiet volume coming off of a plasma speaker I'm making. I tried my oscilloscope hooked to a mouthpiece off an old, military, sound-powered phone with a ton of impedance in it, but it only picked it up off my computer speakers at a loud volume (plasma speaker is still under construction).
    Anyone know a good way to do this?
     
  2. bribri

    Member

    Feb 20, 2011
    133
    5
    a good way would be to buy a sound-level meter. i think they can be found for fairly cheap: velleman, etc.
     
  3. wayneh

    Expert

    Sep 9, 2010
    12,101
    3,033
    A sound level meter is the quick and cheap way. You'll want to supply a broad spectrum, or white noise signal that you can repeat precisely, otherwise your measurements won't mean much.

    A more useful and more elaborate measure is given by testing equipment that chirps your speaker with a frequency sweep and measures sound level against frequency. That's what a modern audio-video receiver does when you set it up, so that it can adjust to your room and speakers accurately. A cheap meter will show dB, put is a poor substitute.
     
    magnet18 likes this.
  4. magnet18

    Thread Starter Senior Member

    Dec 22, 2010
    1,232
    124
    thanks for the info, I was planning on using a specific frequency but I can probably find something on the internet that will provide a frequency sweep.
    I was hoping I could do this without buying anything though, are there any ways to do this with stuff I might have? (already dished out $100 for the speaker parts)
    I was also wondering if anyone know if the plasma arc itself and its very nature might interfere with measurements?
     
  5. wayneh

    Expert

    Sep 9, 2010
    12,101
    3,033
    You can approximate a frequency sweep by just using, say, 5 different frequencies and manually measuring volume at each. You can use any microphone to measure sound - the problem is calibration. Maybe for your purposes, you could use a nice hi-fi speaker, play the tones at a fixed distance from your microphone, and call whatever you get the "standard". Your "test" is your plasma speaker at the same distance, same hi-fi, same microphone.

    There are tone files out there you can use to set equalization of iPod headphones and such. I've used one with tones from low to high, all at the same wave-peak volume. Could be handy.
     
    magnet18 likes this.
  6. magnet18

    Thread Starter Senior Member

    Dec 22, 2010
    1,232
    124
    Whats the difference between hi-fi and normal and why couldn't I use normal computer speakers to get a standard?
     
  7. bribri

    Member

    Feb 20, 2011
    133
    5
    'hifi' does tend to be used as kind of a generic term for music systems.

    computer speakers tend to be quite limited in their frequency ranges from what i've seen. something like a 'full-range' speaker might be what you want for side by side testing...

    the mic from an old telephone device is also going to be very limited in terms of which frequencies it can reproduce.

    what is the desired frequency range of your plasma speaker? i guess this must be fairly high since low frequency reproduction would need a BIG arc.
     
    magnet18 likes this.
  8. Wendy

    Moderator

    Mar 24, 2008
    20,765
    2,536
    Something you may want to consider, many of the free sound card oscopes (which are pretty lousey in and of themselves) also include spectrum analyzers, which can be pretty handy.
     
    magnet18 likes this.
  9. magnet18

    Thread Starter Senior Member

    Dec 22, 2010
    1,232
    124
    I was origionaly planning on using somewhere around 1kHz, but now that I think about it testing at multiple sounds like a good idea.

    The frequency range is hopefully going to be throughout the audio range, not sure exactly what range it is, I'm basically building the plasma speaker off of www.teravolt.org/Plasma_Speaker_2.htm

    My experiment is changing the distance of the spark gap and measuring the efficiency (volume).
     
  10. Kermit2

    AAC Fanatic!

    Feb 5, 2010
    3,779
    932
  11. wayneh

    Expert

    Sep 9, 2010
    12,101
    3,033
    My thinking was that if you use high quality sound reproduction equipment - a "hi-fi" - you can argue that it is accurately producing a sound level from a given input, at least in the range of 50Hz or so up to 15kHz or so. Even good speakers and good hi-fis may struggle at lower and higher frequencies (20 Hz is generally considered the bottom of audible, 20kHz the top, for a young ear).

    Anything less than linear response (volume plotted against frequency) within that range is arguably due to only the microphone, which you can characterize by repeated measurements.

    If you use cheap speakers, no one would accept that they necessarily output a linear response curve. In fact that wouldn't be expected. It would look more like a mountain range.

    I like Bill's idea about using a software oscilloscope to monitor audio in for this project. Cheap and easy, lots of insight into what you're getting. It may not give you anything quantitative but a picture is worth a thousand words.
     
    magnet18 likes this.
  12. bribri

    Member

    Feb 20, 2011
    133
    5
    i'm pretty darn sure you'll be limited to higher frequency ranges.

    oh neat.
    i wonder what kind of radio interference you'll be getting from that.
     
  13. bertus

    Administrator

    Apr 5, 2008
    15,646
    2,345
    magnet18 likes this.
  14. Audioguru

    New Member

    Dec 20, 2007
    9,411
    896
    If you measure sound levels then it must be done by using random pink noise or in an anechoic chamber if continuous sine-wave tones are used.

    I measured the frequency response of speakers with tones by laying the speaker on its back outdoors in a field with the front grille level with the ground (the speaker was placed in a shallow hole in the ground) and the measuring mic hanging above the speaker. Then there are no echoes to interfere.
     
    magnet18 likes this.
  15. t06afre

    AAC Fanatic!

    May 11, 2009
    5,939
    1,222
    In principal. It is not that hard to construct sound measuring equipment. And in theory a soundcard may give you good results. However the problem is to get the setup calibrated. Without pro equipment this is not doable. It is like using a DMM that you know is off scale. But you do not know how much it is off scale
    Also take a look here http://en.wikipedia.org/wiki/A-weighting The human ear do not have "flat" frequency response ;)
     
    Last edited: Mar 9, 2011
    magnet18 likes this.
  16. Audioguru

    New Member

    Dec 20, 2007
    9,411
    896
    The human ear is used to its response and will sound wrong if you try to "correct" the response.
    But I like the automatic bass boost provided by a loudness control.
     
  17. magnet18

    Thread Starter Senior Member

    Dec 22, 2010
    1,232
    124
    I don't think that for my purposes I really need it calibrated, as long as it is consistent.
    I'm trying to find the most efficient (i.e. loudest) distance of the arc gap, so as long a I get a curve I should be able to find the best point.
    Right?
     
  18. wayneh

    Expert

    Sep 9, 2010
    12,101
    3,033
    I hope you'll tell us.

    FWIW, I saw a demonstration of this at an Engineering Open House at the U. of Illinois ~5 yrs ago.
     
  19. magnet18

    Thread Starter Senior Member

    Dec 22, 2010
    1,232
    124
    A plasma speaker or measuring the efficiency of a speaker?
    or both?
     
  20. wayneh

    Expert

    Sep 9, 2010
    12,101
    3,033
    A spark-as-speaker demo. I don't recall if they were collecting any data, or just doing show-and-tell.

    ps: I don't accept friend requests or anything else off-forum. Don't take it personally.
     
Loading...