Page 2 of 3 FirstFirst 123 LastLast
Results 11 to 20 of 29

Thread: Bass localization: an experiment

  1. #11
    Junior Member sssnake's Avatar
    Join Date
    Oct 2011
    Posts
    28
    Both of these findings together would have important implications when it comes to the "up front bass" topic. One of them might be that time alignment is more important than phase coherence. Another might be that shallower slopes, which should enhance overall correlations and reduce the separation in frequency, might improve the illusion of up-front bass. Something to think about.
    Time alignment more important than phase coherence.
    Check, got it, makes sense.

    Another might be that shallower slopes, which should enhance overall correlations and reduce the separation in frequency, might improve the illusion of up-front bass.
    This one is a little harder for me. If we use a shallower slope on the sub (the driver in the wrong location) the spectral content will shift up in frequency improving our ability to localize the sub (to the wrong or at least undesired location). If you are suggesting that decreasing the slope on the midbass (assuming it is up front)would help, I am definitely on board. I'm just not sure on the sub, maybe... I might have to test that one myself. So many ideas to test and so little time :gr_grin:

    BTW - GREAT info and thanks for sharing.

  2. #12
    Senior Member
    Join Date
    Oct 2010
    Posts
    169
    Yeah, I don't know what to make of the idea of shallower slopes on the rear sub just yet. You'll get more high freq information coming from behind you, but will the increase in correlation trump it?

    Still working on this and hopefully will have some updates soon.

  3. #13
    Senior Member cvjoint's Avatar
    Join Date
    Mar 2011
    Location
    Northern California
    Posts
    801
    Real Name
    George
    Well, I would think that there is a third factor here. The shallower the slope the more the sub source ends up being it's own stand alone fullrange. Even though the correlation increases you now have dual soundstages, or closer to it.

  4. #14
    Founding Member Subwoofery's Avatar
    Join Date
    Sep 2010
    Posts
    353
    Real Name
    Kelvin
    George, am just wondering if you've tried to play some test tones on your LAT to see when you localize bass... Since the LAT helps in minimizing rattles, the experiment should be a good one IMO

    Kelvin

  5. #15
    Senior Member cvjoint's Avatar
    Join Date
    Mar 2011
    Location
    Northern California
    Posts
    801
    Real Name
    George
    Well, the idea would be to isolate rattles completely. The LAT doesn't do that since it still needs to move air. There are still rattles present from the trunk compressing or the cabin and rattling the soft top. If someone is willing to mount a lat in a sealed box in a fairly large hardtop car with the windows open I think it would work. That way there is no baffle or box strain, the baffle by virtue of push-pull and the cabin by virtue of size, strength and widow openings.

    My best bet is to try it with the trunk open and the top down and EQ. in the output lost to cancellation. It could work. Maybe my next amp won't be limited to 180hz either, that would be needed as well.

  6. #16
    Tester Extraordinaire ErinH's Avatar
    Join Date
    Aug 2010
    Posts
    1,247
    nothing terribly important... just a FWIW.

    I've been dealing with something in my subwoofer response that has been driving me nuts for years. Never really had the ability to go after what I thought it was so just left it alone.
    I've recently gotten in the mosconi 6to8 dsp which allows for varying Q adjustments and shelving filters. Finally, the chance to go after some specific issues. I spent a few hours in the garage with everything except the subs turned off. I turned the LPF off and let them play naturally as high as they'd go. I ran a few sweeps on TrueRTA with 'peak hold' set on. I made sure to use sloooowwww sweeps to try to excite the cabin well. I found a lot of issues that I expected but one that I didn't expect... a serious cabin related mode due to the trunk itself (proved this by finding it now where else except when a sub was in the trunk) at about 210hz. The Q wasn't terribly narrow. I wound up using a Q of about 2 with an attenuation of 12dB to remove it from the car as much as possible. Sure enough, it resolved a problem that has been bothering me for years. The thing is, since it is so far out of the passband the subs play, I would have never gone looking for it. I actually found it by accident.

    But, that wasn't the only thing I found.
    To make a long story short, this is what I had before I started experimenting:
    • Midbasses: No HPF (used aperiodic) - 200hz/24dB L-R
    • Subs: No HPF - 50hz/24dB L-R



    And, after experimenting, I wound up with the following settings:
    • Midbasses: 60hz/24dB - 200hz/24dB L-R
    • Sub LPF @ 70hz/24dB L-R
    • Sub HPF @ 40hz/12dB L-R*
    • Sub High Shelf Filter @ 75hz, -6dB**


    Note: I used no EQ during this research project and spent a lot of time with other crossover/slopes when I landed on this final configuration. The final result does have some EQ here and there but nothing like what I had before and especially not in the 100-160hz range.


    *Yep, a subsonic filter, but not at subsonic frequencies. Who actually uses this? Well... not me... until now. The reason I am, however, is that the increase in cabin gain - at least in my car - is much different below 40hz than it is from 200hz-100hz. Cabin gain allows a nice swing up in the low end and it helps get more out of the low end response. I always assumed that it was best to let it carry out but it seems either due to efficiency in the low end or just flat out thanks to the shape of the cabin, below 40hz, the car bumps the slope of the response up a notch in to a realm where it just flat out doesn't allow for a well-meshed response with the midbasses, no matter if I let the midbasses roll off naturally or used a HPF on them. So, using a HPF on the subs at 12dB slope really helped to blend things and keep from having overbearing low end while *not* having enough impact from the 60-80hz region.


    ** The car is stupid. I get a serious null from about 50-70hz causing a relatively large bump in response from about 80-120hz thanks to position seated, in the sub response. This was just another reason why I had been experiencing that "pull to the rear" response when I increased the sub LPF. It wasn't a rattle so much as the frequency response. Attenuating the response above 75hz basically resolved the upper swing in response from 80-120hz and brought it to a point where it was more in line with the rest of the subs' response. The LPF at 70hz helped to line everything up a bit more.
    It's a lot easier to explain with a graph, but I didn't take a screenshot. I drew it up off memory and attached an example of how the original response with the subwoofer only, no filter applied, looked.
    The other attached picture, in red, is an idea of how it looked after I applied the filters to shape up the response a bit more. Obviously it's not exact, but it should give you an idea of what I had (green) vs what I now have (red).

    This post doesn't even begin to really cover what else was done. It just highlights the major things done that seemed to have the most effect on getting everything to blend. I increased my midbass HPF from *none* to 60hz/24dB L-R. As it stands, my car has the absoulte best sound it ever has. I have had a very good response above 300hz for a few months now... IMO the best it's ever sounded. I have had a lot of trouble getting the impact out of the system it needed and it's solely because I was having to tune away some of it to get rid of the pull to the rear effect. After 3 hours, I managed to regulate a very large majority of that issue and as a result, now have the most impactful and well blended, centered up front, response I've ever have. And, as I toot my own horn, I can say it does it without pulling your attention to your butt cheeks or your shoulder blades like so many cars do. My whole attitude toward my audio system has now changed from jaded to excited again. Probably the best 3 hours tuning session I've had in a few years.


    I rambled a lot. I really only intend to say: use a mic. Experiment. Play with things and you may be surprised at what you find and even more so surprised at what helps to get a better system response. I think if you can determine where a null in response is - which is going to cause a noticeable difference to those frequencies that are higher in response relative to the null - and you can also determine where modes are and you have the ability to manage them, you'll get a much more pleasurable sound and probably have a much more realistic/impactful system without feeling like your subs are overpowering or your attention being pulled to the rear or your butt. ;)
    Attached Images Attached Images   
    Last edited by ErinH; 05-14-2012 at 04:41 PM.
    Your ears: The best tools you have... and they're free, too!

  7. #17
    Senior Member
    Join Date
    Oct 2010
    Posts
    169
    Quote Originally Posted by cvjoint View Post
    Well, I would think that there is a third factor here. The shallower the slope the more the sub source ends up being it's own stand alone fullrange. Even though the correlation increases you now have dual soundstages, or closer to it.
    I'm not sure that "dual soundstages" can exist with 100% correlation.

  8. #18
    Senior Member cvjoint's Avatar
    Join Date
    Mar 2011
    Location
    Northern California
    Posts
    801
    Real Name
    George
    Maybe I'm misunderstanding the use of correlation here. In the limit the subwoofer would reproduce the same material as the front speakers identically. Correlation is 1 because if you hear some program material in one you will hear it on the other with certainty. They are dual soundstages because the brain can pick them up as individual sound sources.

    Example: two subwoofers playing 40hz and below with a steep slope=corr 1 between them, but assuming they can't be localized the soundstage is still intact, no doubling.
    Last edited by cvjoint; 05-17-2012 at 01:24 PM.

  9. #19
    Senior Member
    Join Date
    Oct 2010
    Posts
    169
    If you were to interpret two correlated sources as completely distinct, then most of the problems we're talking about in this thread wouldn't exist in the first place. For example, rattles from the back would have zero influence on our localization of bass, yet most people would acknowledge that they do. This is because they're correlated with the subwoofer's "beat". If you just had a rattle in the same place that occurred whenever you went over bumps, it wouldn't alter your bass perception.

    Part of the reason that stereo is such a neat effect is that the temporal envelopes of the signals are broken to the point where we can begin to isolate individual sources. If our brains just integrated everything over a very broad time window, stereo wouldn't work. Even ILD effects require correlation.

  10. #20
    Senior Member cvjoint's Avatar
    Join Date
    Mar 2011
    Location
    Northern California
    Posts
    801
    Real Name
    George
    What I think happened is that I was referring to correlation of program material whereas the purpose of your quest is temporal correlation. I was a bit off topic then.

    I forget if it was you that recommended a neat experiment, introduce a correlated buzzing/rattle in the front speakers to pin the stage down in front. My car is sort of a realized experiment of this. With LATs in the back and conventional midbass up front, the rattles up front completely overwhelm the ones in the back (which are also less correlated temporally, not just lower output). The stage is pinned up front better than before but I suspect if I change my midbasses to LATs some of the correlation will disappear and maybe the back will be more localizable again. That would be unfortunate.

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •