Writing a program for the BFD..need input

Discussion in 'Archived Threads 2001-2004' started by Anthony_Gomez, Aug 4, 2002.

  1. I am in the middle of writing an EXCEL based program that will hopefully get added to the FRC Consortium. Ed West started me up with the basic parametric EQ formulas.
    What it does is model the BFD gain and phase with your measured or simulated responced.
    For calculated responces, this is set up currently for a sealed box only to see what settings will work as a psuedo linkwitz transform. Best results work with Low Qtc systems (must be = to or lower than 0.6 IMO, but you may get away with up to 0.707). The equations I am using right now are just the basic ones from Brian Steele's website. I hope to later figure out what forumulae to use to account for box stuffing and leakage (Isaac????).
    For measured responce, you can use it to try and predict the best settings for the BFD to counteract room interactions. Since the room does not respond linearly with modes and nulls, this will work as a guide and not a one shot wonder.
    I will have the measured responce be selectable to use the RS SPL correction value and possibly your own correction values.
    Later I hope to work with Paul V on this to work on an importing feature from Unibox.
    Here is a screen shot of it SO FAR. Formating is not a concern at this point. I still do not have driver phase worked out, so it is not working yet in the sum.
    http://beyond_gomer.tripod.com/bfd.html
    Now for your Input
    1) What type of resolution do you want? The higher the resolution, the more accurate the low bandwith gains appear. The higher the resolution, the more measurements you have to take. 1/6th octive is minimally adiquare IMO
    2) Is there any additional thing I should try to incorperate, or is there anything I should change from what I have already meantioned?
     
  2. Isaac C

    Isaac C Stunt Coordinator

    Joined:
    Aug 1, 2000
    Messages:
    140
    Likes Received:
    0
    I'll get back to you on that, perhaps after my class today. Have you tried the Windows help file version of "A Unified Model for Closed-Boxes, Vented-boxes and Passive Radiators", a paper (Preprint # 2841 (H-2) presented by Bullock at the AES 1986 Convention? I think Unibox was based on the paper, I'm not really sure.
    I've my own set of equations, let me know if you still want them. It would help if you can define complex number variables and apply the basic mathematical operations (MDAS) directly on them. I'm only familiar with C++, so I'm not sure if VB+Excel will let you do this exactly. But the simple all-acoustical model of a sealed-sub is really just a series RLC circuit -- assuming well-sealed box and negligible voice-coil effects at the bandwidth of interest. A complex number class makes things easier if you're after just a bit more model complexity.
    cheers,
    Isaac
     
  3. Thanks Isaac. I think I got most things down, but there are 3 things that I need some help on (I am ditching phase extraplation since people here won't even use it).
    How do I incorperate the following? (formulas[​IMG])
    Ql
    Qa
    Le
    The only formulas I understand right now are on Brian Steeles site.
    Thanks
     
  4. Isaac C

    Isaac C Stunt Coordinator

    Joined:
    Aug 1, 2000
    Messages:
    140
    Likes Received:
    0
    Have you had a look at Bullock's paper above? The equations he presents makes use of Ql and Qa (and I think even Le).
    Let me know if you need more help, good luck! [​IMG]
    Isaac
     
  5. I'll have to look again to see =)

    I was just lazy (relative lazy ness if you know what I mean)
     
  6. The first Beta release is almost ready! I just have to get it to properly calculate three things (I have to manually add them in right now).

    I/O features will hopefully be added in the second release.

    It should be pretty accurate and it even uses complex math (thanks Isaac!) to get accurate responces involving Le, Qa and Ql.

    I have the program set-up for 1/12th octave analysis, so you 1/6th people need to adjust=)

    You will probably see the release in a few days if all goes well.
     
  7. Stephen Dodds

    Stephen Dodds Second Unit

    Joined:
    Aug 29, 1998
    Messages:
    354
    Likes Received:
    0
    Anthony,

    Nice program. One question though. How do you get your measured measurements into it?

    I can make an FRD file with Unibox, but it would be nice to get the measurements I've made using TrueRTA. I tried using the SPL Trace program to convert my TrueRTA screen shot jpgs but get nothing but blank FRDs.

    As a suggestion, it would be kind of cool to expand the program over the whole audible frequency. Oh, and to be able to model active crossovers as well.


    Cheers

    Steve
     
  8. as long as the file is an .frd file, the program can import it. I am actually working on a revision that corrects some stupid spelling, the fine tuning spacing, and will extend out to 1k.

    This program does not, and will not model anything other than the parametric functions. THere are other programs for that.

    Enjoy
     

Share This Page