1. Sign-up to become a member, and most of the ads you see will disappear. It only takes 30 seconds to sign up, so join the discussion today!
    Dismiss Notice

Sub placement

Discussion in 'Archived Threads 2001-2004' started by Rick Radford, Sep 19, 2001.

  1. Rick Radford

    Rick Radford Supporting Actor

    Joined:
    May 12, 2001
    Messages:
    642
    Likes Received:
    0
    Trophy Points:
    0
    Today, I had some free time and experimented with sub placement. I have a 39" wide alcove either side of my entertainment center where the L & R mains and the sub sit (in the corner). Room is 11.5 x 12 x 8.
    I moved the SVS 20-39CS from one alcove to the other and picked up 6 dB of headroom (I dialed the sub down 6dB to keep the SPL #s the same).
    I assume the difference must be room acoustics. The former location has a wall 12' away. The new location instead of a wall 12' away has a 4' opening into another room 12' away.
    I was quite surprised at the difference that made.
    ------------------
    --RR
     
  2. Cameron Seaman

    Cameron Seaman Supporting Actor

    Joined:
    Apr 26, 1998
    Messages:
    708
    Likes Received:
    0
    Trophy Points:
    0
    Rick, I had no idea you had a SVS sub! Let me know if you wish to audition it. [​IMG]
    I'm getting a 50H81 delivered tomorrow that I'd be more than happy to show off for you some time! [​IMG]
    Ok, now back to regularly scheduled topic...
    ------------------
    [​IMG]
    The HTF Lowest Post Champion!
     
  3. ling_w

    ling_w Second Unit

    Joined:
    Sep 3, 2001
    Messages:
    426
    Likes Received:
    0
    Trophy Points:
    0
    Are you measuring the response with narrow band (1/3 oct) pink noise/warble tone or is it just a wide band pink noise.
    If you are measuring with wide band pink noise, you new placement either
    - excited more room modes so much that the avg level is effected by it.
    - enhanced the low freq so that it did not roll off as early. e.g. -3dB @ 25hz instead of 40hz.
    - room mode null has been eliminated.
    It usually is not the 2nd choice, since 6dB increase means the overall level is much higher than the what it was prior. A pink noise of 20-80 hz indicates if you quadrupled your sound output, keeping peak/nulls the same, if the current response is flat to 20hz, the prior rolloff would have to been @ 56hz. A highly unlike situation.
    You should do narrow band measurements of both location and see what is causing the increase in sound level. More is not always better, don't think corner placement is always the best, especially in a square room like yours where room length and width mode could potentially cause twice the problem in the amount of bass peak
     
  4. Rick Radford

    Rick Radford Supporting Actor

    Joined:
    May 12, 2001
    Messages:
    642
    Likes Received:
    0
    Trophy Points:
    0
    Cameron, private email coming your way.
    Ling, I'm gonna hafta digest your post a bit. Back at ya later...
    ------------------
    --RR
     
  5. Rick Radford

    Rick Radford Supporting Actor

    Joined:
    May 12, 2001
    Messages:
    642
    Likes Received:
    0
    Trophy Points:
    0
    >>Are you measuring the response with narrow band (1/3 oct) pink noise/warble tone or is it just a wide band pink noise
     
  6. ling_w

    ling_w Second Unit

    Joined:
    Sep 3, 2001
    Messages:
    426
    Likes Received:
    0
    Trophy Points:
    0
    I don't have the avia disc, so I don't know what that is. Maybe you could use the Low Freq Sweep for the LFE channel under Verification/Evaluation. You will have to watch the meter as it sweeps across in freq and take note at what freq is there peak and null readings.
    I use the Stereophile test disc that has fixed (as opposed to sweep) warble tone at 20,25,32,40,50,63,80,100,125hz and so on.
    The real way to measure sub level -- if you set your main to small -- is to match the level of the L+R level and the sub while playing 1/3 octave tone/warble tone centered at the crossover freq (if the HP is 2nd order and HP is 4th order, you might have to set the L+R level 3dB louder.) This ensures the transition from the main to sub is seamless with no sudden increase or decrease in perceived level. A general broadband pink noise is just not accurate enough, since it average the peaks and dips into one SPL reading.
     
  7. Rick Radford

    Rick Radford Supporting Actor

    Joined:
    May 12, 2001
    Messages:
    642
    Likes Received:
    0
    Trophy Points:
    0
    The LFE sweep shows all kinds of activity on the SPL meter. My strongest peaks now occur in the low 40 Hz area vs mid 50's when the sub was in the opposite corner.
    >>while playing 1/3 octave tone/warble tone centered at the crossover freq (if the HP is 2nd order and HP is 4th order
     
  8. ling_w

    ling_w Second Unit

    Joined:
    Sep 3, 2001
    Messages:
    426
    Likes Received:
    0
    Trophy Points:
    0
     
  9. Rick Radford

    Rick Radford Supporting Actor

    Joined:
    May 12, 2001
    Messages:
    642
    Likes Received:
    0
    Trophy Points:
    0
    Let me get out my VE disc and try. I've been using Avia for the sound tests, but really like the presentation (video and audio) of VE.
    Avia offers more audio setup options, however, which is why I've been using it for sound setup.
    Regarding Digital VE, it's not out yet.. and appears to be delayed until sometime early next year. [​IMG]
    I didn't record any figures on how high the peaks were.. but it's safe to say the room is... active. [​IMG]
    >>at what dB was the SPL hovering when the meters were not hitting peaks, in both sub position? Was one corner 6dB louder than the other
     
  10. Rick Radford

    Rick Radford Supporting Actor

    Joined:
    May 12, 2001
    Messages:
    642
    Likes Received:
    0
    Trophy Points:
    0
    Ok, here's what I did today.
    Plugged in VE and recalibrated all 5.1 channels at 75 dB. I decided to reset everything with my master volume at 0 Relative. (is there any reason NOT to do this?... earlier results had the 75dB level at -4 on the MV)
    Here's where all levels are with the MV at 0 for VE and -8 for Avia (75 dB):
    LMain: -4.0
    CC : -3.5
    RMain: -5.0
    RS : -3.5
    LS : -2.5
    Sub : -14.0 (consistent 76-77 dB readings)
    I had to dial the Samson back to -5dB (from 0db) because I ran out of adjustment room on the Onkyo 989. The Onkyo 989 limit on the sub adjustment was -15dB.
    It was interesting to note that I used VE first to set levels. Then I dialed the MV down by 10dB to compare Avia (since it should be calibrated at 85 dB). But to get the same SPLs, I had to crank up the MV to -8 dB. I guess that's close enough, but am curious as to why there'd be any difference unless my receiver is giving inaccurate +/- relative dBs.
    Will MV 0 give me reference level now? I thought I had that figured out but I've confused myself.
    ------------------
    --RR
    [Edited last by Rick Radford on September 23, 2001 at 02:07 PM]
     
  11. Guy Kuo

    Guy Kuo Supporting Actor

    Joined:
    Mar 6, 1999
    Messages:
    581
    Likes Received:
    0
    Trophy Points:
    0
    Electrically the levels are 10 dB louder in AVIA and in a anechoic room with perfectly flat responding speakers you would note a 10 dB difference in master volume settings between VE and AVIA to get the same SPL readings. However, the tones are not spectrally identical and this means that room effects and speaker voicing reproduce the signals with slightly differing characteristics that are basically an imprint of your room and speakers' effects. A 2 dB difference between two differing test tones is quite normal. You'll note such a difference between built-in tones and test disc tones. Which tone is correct? They all should be in roughly the same area.
    One thing to note is that pink noise is not recommended for speaker balancing due to the unequal frequency responses of the various speakers in a surround system. That is why shaped noise is used. Shaped noise is band pass limited to a range which should be well reproduced by all the speakers in the system.
    AVIA's speaker level tests are bandpassed from 35-70 Hz (one octave) for the subwoofer and 500 to 2kHz (two octaves) for the main 5 channels. The rms levels of the two portions of the tests are identical so they will read the same level on your meter. Per David Ranada, both bandass noises were brickwalled at something like 14th order. As you have already discovered we also included warble and phase matching noise at the crossover frequency range to aid in sub integration.
    ------------------
    Guy Kuo
    www.ovationsw.com
    Ovation Software, the Home of AVIA DVD
    [Edited last by Guy Kuo on September 23, 2001 at 02:28 PM]
     
  12. Rick Radford

    Rick Radford Supporting Actor

    Joined:
    May 12, 2001
    Messages:
    642
    Likes Received:
    0
    Trophy Points:
    0
    >>A 2 dB difference between two differing test tones is quite normal.
     
  13. Guy Kuo

    Guy Kuo Supporting Actor

    Joined:
    Mar 6, 1999
    Messages:
    581
    Likes Received:
    0
    Trophy Points:
    0
    If I recall the correctly the wide band pink noise are actually pink noise rather than shaped noise. That's why those signals were not featured in the section for setting speaker levels. The pink noise is on the disk for those who wish to examine frequency response with a spectrum analyzer.
    White noise is a signal which has equal energy at every frequency. We use Pink Noise in audio work because of our propensity to relate pitch in a logarithmic matter or octaves. As you know each octave is at double the frequency of the one below it. An RTA or equalizer usually splits its bands in an octave like matter. Since the number of possible frequencies in each octave is NOT constant due to each octave doubling in frequency, white noise does not have equal energy in each octave. Pink noise has a frequency content shaped for increased amplitude in the lower frequencies to make each OCTAVE equal energy. That makes it read "flat" on RTA's which split the bands in the usual octave manner. If you compare the sound of pink noise against white noise, you'll notice that pink noise has a lot more low frequency component compared to white noise.
    Shaped noise as in those used for setting speaker level in a surround system is pass banded to have a limited range of frequency content. The speaker level setting tones in AVIA are shaped noises, but the wide-band pink noises are pink noise. As I recall, there are also low frequency pink noises on the disc which are a truncated variant of pink noise which has the upper frequencies removed to allow isolation of the test to just the bass frequencies.
    If you have graphic equalizer or RTA, you can examine the various shapes of the noises in AVIA by directly feeding the signal into the EQ or RTA. If you feed it in via an microphone you'll see the room and speaker effects superimposed on the underlying signal.
    ------------------
    Guy Kuo
    www.ovationsw.com
    Ovation Software, the Home of AVIA DVD
     
  14. Rick Radford

    Rick Radford Supporting Actor

    Joined:
    May 12, 2001
    Messages:
    642
    Likes Received:
    0
    Trophy Points:
    0
    Great info, Guy.
    Thanks for the clarification.. and the education. [​IMG]
    ------------------
    --RR
     

Share This Page