Ultra Low Tuning for Tempest EBS

Discussion in 'Archived Threads 2001-2004' started by Nate_Herring, Oct 25, 2002.

  1. Nate_Herring

    Nate_Herring Auditioning

    Joined:
    Jan 5, 2002
    Messages:
    5
    Likes Received:
    0
    Trophy Points:
    0
    Well I guess not ultra low but lower than 15.8 Hz.

    I'm going to finalize my inlaws' DIY speaker set with an EBS Tempest. I'm using the Adire LspCAD software. I'm going to be using the 340L box size but I'm toying with the idea of a low tuning frequency of 13.1 Hz with a 30" long 6" PVC pipe. This alignment is over damped but yields a flatter in-room response as reported by LspCAD. Of course the group delay is reduced as well.

    Should I put any trust in the room response numbers in LspCAD? I also have the full version of LspCAD, would it be a better judge of room response?

    Thanks.
     
  2. Michael R Price

    Michael R Price Screenwriter

    Joined:
    Jul 22, 2001
    Messages:
    1,591
    Likes Received:
    0
    Trophy Points:
    0
    Nate, I for one do not trust that room response sim. It predicted way too much low-frequency boost for me (causing a performance compromise) and has for some others as well. I think it maybe because it assumes your room is perfectly sealed off, or something like that.

    AFAIK, most rooms tend to have a gain of 3-6db at 20Hz, depending on the size. In that case a 13.1Hz tuned Tempest would probably roll off a little.

    Note that whether the frequency response is flatter or not, you'll lose headroom above tuning because the port takes over less of the driver's work in the important 20-30Hz range. I recommend you tune to 16Hz or thereabouts.
     
  3. SteveHood

    SteveHood Agent

    Joined:
    Nov 29, 2001
    Messages:
    26
    Likes Received:
    0
    Trophy Points:
    0
    From what I can tell the room response curve does nothing. I tried changing the size of the room to 2Mx2Mx2M and the curve didn't change one bit. Not only should have the 20hz point increased but since the room is perfectly squared there should have been large peaks and valleys in the curve. I then changed the room to 10Mx10Mx10M and again no change.

    I've also seen weird in-room SPL result with PR configurations. (ie. the PR design having 10db more output than the equivalent vented design)
     
  4. Isaac C

    Isaac C Stunt Coordinator

    Joined:
    Aug 1, 2000
    Messages:
    140
    Likes Received:
    0
    Trophy Points:
    0
    The model used by LspCAD is not the most accurate and it does not try to be so. However, it is good enough and better results can be attained with better input data. It is important to use the most relevant reflection coefficients for your room surfaces; and any object -- sofa, coffee table etc. -- will have varying effects on the actual room response, which cannot be taken into account by the room image model.
    As for the seemingly strange and unchanging response with changing room dimensions, this is just the characteristic of the image model and it should not be mistaken for program inaccuracy [​IMG] For example, changing the height of the virtual room without including ceiling reflections won't affect the curve. Click here for more information on image analysis and here is the main room acoustics page of A. Ludwig.
    hth [​IMG]
     
  5. Nate_Herring

    Nate_Herring Auditioning

    Joined:
    Jan 5, 2002
    Messages:
    5
    Likes Received:
    0
    Trophy Points:
    0
    Thanks for the help.


    I'm going to try to figure out a way that I can temporarily test the room response with different port sizes. The sub will be corner loaded, as there really isn't ant room for this monster any place else. I'll report back if I get this to work if anyone is interested.
     

Share This Page