2 questions about the 3802...possibly

Discussion in 'Archived Threads 2001-2004' started by ben hunt, Jul 18, 2002.

  1. ben hunt

    ben hunt Stunt Coordinator

    Joined:
    Jun 12, 2002
    Messages:
    82
    Likes Received:
    0
    Trophy Points:
    0
    Ok I just got a 3802 a week and a half ago. Now I assume that reference level is when the volume indicator is at 0 (the 3802 goes from -70 to +18). Is that right? Because if it is then I can't imagine listening to anything that loud for an extended period of time in my house. I usually hover between -24 and -10.

    Second question, I have Axiom m60ti's for mains. At what point is there too much power going through the speakers. I doubt my 3802 will tax the speakers that much but what if I added an amp for some extra power? At what point is my power source too much for my speakers and how do I tell when I get to this point?
     
  2. AustinKW

    AustinKW Stunt Coordinator

    Joined:
    May 30, 2002
    Messages:
    169
    Likes Received:
    0
    Trophy Points:
    0
    Hi Ben,

    Eventually, you're going to need an SPL meter to balance speaker levels in your surround setup. So, bite the bullet and get the Radio Shack analog meter (about $30). Next time you go to the movies, take the meter along with you and see what kind of SPLs are being blasted at you. When you get into HT, you'll find that the sound effects blow you out of your chair at -24 or -10 but that you're leaning forward struggling to hear dialog on your center channel at the same volume level. Music is generally listened to at MUCH lower volumes - even your -24 to -10 is a bit much for my tastes.

    Re the power to the speaks - your ears will give out long before your speakers will PROVIDED your amp does not go into clipping or your tweeters into compression. Check the specs on your Axioms. They should show a watt rating (say 200W) and a peak rating (say 350W). Stay within these limits without clipping or compressing and you're golden. Clipping is where the amp can no longer reproduce the input signal and "chops off" or clips the top of the waveform. A sine wave then starts looking like a square wave which has theoretically infinite energy in the high frequencies. Lots of energy directed at a tiny tweet - not good! Compression is where an increment in power does not get you an increment in SPL output. The speaker driver therefore turns the additional power into heat. Speakers can make heat or sound but not both!

    Austin
     
  3. Aslam Imran

    Aslam Imran Second Unit

    Joined:
    Mar 1, 2002
    Messages:
    286
    Likes Received:
    0
    Trophy Points:
    0
    0 db as reference level is only arelative and not an absolute level. 0 db in your room might sound louder than reference and might sound softer than reference in someone else's room. In a small room with low ceilings you might hit reference levels at -20 db. So get an SPL meter and calibrate your speakers.

    Also I dont think any amp overloads a speaker. Speakers cant be overloaded (unless it happens for a long period of time) they can only be clipped. The more powerful the amp the less likely it will go into clipping. Most speakers of 90 db nominal sensitivity will only require a watt of power to play loud its only during those dynamic peaks of 105 db that you need about hundred times more power and speakers can take brief moments of higher than rated power without being damaged provided that power is clean and unclipped, and the bigger your amp the more the likelihood of that power being clean and undistorted.
     
  4. RonGecan

    RonGecan Agent

    Joined:
    Jun 19, 2002
    Messages:
    32
    Likes Received:
    0
    Trophy Points:
    0
    Austin,
    Do you know of a source that explains clipping in a little more detail than what you have there? I've been asking around on issues just like this -- ie. what happens when the receiver clips, most people answer, "well the speaker can't handle it", why? "well, it just can't". So where, theoretically, does the spike in energy come from when the speaker faces a squaring of the sine wave? I assume it's at the kink, but why would a kink lead to an infinte energy? Maybe asking for too many details, but this sort of thing interests me.
    Also, when does compression kick in?
    Thanks, very helpful. [​IMG]
    Ron.
     
  5. Aslam Imran

    Aslam Imran Second Unit

    Joined:
    Mar 1, 2002
    Messages:
    286
    Likes Received:
    0
    Trophy Points:
    0
    from Fourier theory a square wave can be resolved into infinite sinusoidal terms of varying frequencies. So in other words a square wave has an infinite frequency content.

    Compression is when your dynamic range is being restricted i.e one end of your freq spectrum (usually lower freq) is being restricted while your other end (usually higher freq)is still being amplified leading to compression. And because woofers can handle about 5-6 times more power than the tweeters, the tweeters burn out even though the woofers are being distorted. So clipping occurs on the woofers while the tweeters are still fed clean power but much more than they can handle resulting in their destruction.

    The way drivers burn out is that drivers are always producing heat because they are not 100 % efficient. So one watt of electrical energy is only partly converted to acoustic energy while the rest is dissipated as heat in the voice coils. More energy = more heat = melted voice coils. Also distortion leads to nonlinear motion if the drivers that can destroy them.

    Hope that helps
     
  6. AustinKW

    AustinKW Stunt Coordinator

    Joined:
    May 30, 2002
    Messages:
    169
    Likes Received:
    0
    Trophy Points:
    0
    Thanks, Aslam. I LOVE these forums - you learn so much from smart people!

    Ron, I think there's a bunch of tech stuff on Rane's site. You might try there or ask Aslam for other tech sources.

    Austin
     
  7. RonGecan

    RonGecan Agent

    Joined:
    Jun 19, 2002
    Messages:
    32
    Likes Received:
    0
    Trophy Points:
    0
    Thanks for the responses. I've never really studied Fourier analysis, though I understand in principle what it's about. My recollection is, however, than ANY curve can be approximated arbitrarily closely as an infinite collection of sinusoidal waves, so that a square wave is no exception. In fact, doing a quick internet search yields many examples of using Fourier series to approximate, square waves, step functions, etc. So unless there's some special property relating to audio waves, I wouldn't think a square wave is really different than anything else.

    What is Rane's site?

    Thanks,

    Ron.
     

Share This Page