Linkwitz Transform Spreadsheet...Help!

Discussion in 'Archived Threads 2001-2004' started by Brian Bunge, Aug 3, 2001.

  1. Brian Bunge

    Brian Bunge Producer

    Joined:
    Sep 11, 2000
    Messages:
    3,716
    Likes Received:
    0
    Trophy Points:
    0
    I was playing with the Linkwitz Transform spreadsheet last night and came up with a few questions. What should the value of "k" be on the spreadsheet? I noticed that it defaults to 1 and that it says k>0 but I don't know what it's really for. There doesn't seem to be any reference to it in the notes. I did notice if I increase the value that it changed the response of the system. It seemed the response was flatter down to 20Hz but also had a steeper slope below 20Hz. Is this just something that you change to tweak your response? I did notice that changing k also changed the resistor and capacitor values as well.
    Can anybody help me?
    Thanks,
    Brian
    ------------------
     
  2. Julian Data

    Julian Data Second Unit

    Joined:
    Oct 5, 1999
    Messages:
    408
    Likes Received:
    0
    Trophy Points:
    0
    Brian,
    Who's spreadsheet are you using? I am using Paul's and I don't see a K setting.
    ------------------
     
  3. Brian Bunge

    Brian Bunge Producer

    Joined:
    Sep 11, 2000
    Messages:
    3,716
    Likes Received:
    0
    Trophy Points:
    0
  4. Julian Data

    Julian Data Second Unit

    Joined:
    Oct 5, 1999
    Messages:
    408
    Likes Received:
    0
    Trophy Points:
    0
  5. Brian Bunge

    Brian Bunge Producer

    Joined:
    Sep 11, 2000
    Messages:
    3,716
    Likes Received:
    0
    Trophy Points:
    0
    Julian,
    First of all, thank you for the link! Also, in this spreadsheet I see k! Look in the box where you enter f(0), Q(0), f(p), and Q(p). Just below those values is k. That's what I'm trying to figure out. It is changeable and it does change the response.
    Does anyone know what this is?
    Brian
    ------------------
     
  6. Greg Monfort

    Greg Monfort Supporting Actor

    Joined:
    May 30, 2000
    Messages:
    884
    Likes Received:
    0
    Trophy Points:
    0
    If you right click on the 'K' value, you'll see the formula for it in the header. It's the ratio of existing Vs desired Fb, Qtc. When the 'K' value is < 0, you get negative component values, which of course you can't have.
    GM
    ------------------
    Loud is beautiful, if it's clean
     
  7. Julian Data

    Julian Data Second Unit

    Joined:
    Oct 5, 1999
    Messages:
    408
    Likes Received:
    0
    Trophy Points:
    0
    Brian,
    Sorry for being so blind. [​IMG] Greg is correct.
    Damn.. I am tired.
    ------------------
     
  8. Brian Bunge

    Brian Bunge Producer

    Joined:
    Sep 11, 2000
    Messages:
    3,716
    Likes Received:
    0
    Trophy Points:
    0
    OK, now that I have a handle on that I have another question. Using the new spreadsheet that Julian provided the link for I have another problem. I am modeling the ACI SV18 in a 66.5L enclosure, which gives me Qtc of .707. The Qts=.227, Fs=15.5, Vas=579L. I "transform" it to fp=20Hz and Qp=.6 with a subsonic filter at 12Hz, Q=.7 and 2000W of amp power.
    Whenever I try to enter the driver parameters below (specifically PE) it tells me I have exceeded both the amp's max power and the driver's Xmax. The driver specs are Qes=.253, Xmax=20mm, PE=1000, Sd=1200cm^2. If I enter anything higher than 100W for PE it gives me this message. Shouldn't PE be the driver's RMS value?
    Also, on the schematics page, what's the purpose of multiplying the calculated R & C values by M? Why would you need to change the "impedance level"?
    OK, Jack G. and all you guys that always recommend trying a LT circuit, please help me out here!
    Thanks,
    Brian
    P.S. I'd load the file as I have it saved if someone could tell me how!
    ------------------
     
  9. Brian Bunge

    Brian Bunge Producer

    Joined:
    Sep 11, 2000
    Messages:
    3,716
    Likes Received:
    0
    Trophy Points:
    0
    Thanks anyway, Jack! Maybe I should post this over on Madisound or the Bass List?
    Brian
    ------------------
     
  10. Jack Gilvey

    Jack Gilvey Producer

    Joined:
    Mar 13, 1999
    Messages:
    4,948
    Likes Received:
    0
    Trophy Points:
    0
    I had used the older one just a bit, and I just grabbed the newer one.
    PE appears to be the actual applied power, and not the thermal/power rating of the driver. You can see the SPL change when PE is changed. In almost every instance, Xmax is reached well before the actual power limitation of the driver, so that's not too surprising.
    It does appear to figure Vd incorrectly, reporting it 100 times too large, if I entered things correctly.
    To save it, just choose "File", "Save as..." and change the name to "Briansub.xls", or something more appropriate [​IMG] .
    [Edited last by Jack Gilvey on August 07, 2001 at 08:58 AM]
     
  11. Brian Bunge

    Brian Bunge Producer

    Joined:
    Sep 11, 2000
    Messages:
    3,716
    Likes Received:
    0
    Trophy Points:
    0
    That makes more sense now! When you mention SPL are you referring to pSPL all the way to the right? Do you think this is the Max SPL and if so whould you just assume this is at the highest point on the response curve? That whould make sense to me.
    As far as Sd, I think I input that wrong. Sd of .120m^2 should have been 120cm^2 and not 1200cm^2 as I input it.
    Thanks, Jack!
    Brian
    P.S. Where's Tyler, Max, et. al.? [​IMG]
    ------------------
     
  12. Brian Bunge

    Brian Bunge Producer

    Joined:
    Sep 11, 2000
    Messages:
    3,716
    Likes Received:
    0
    Trophy Points:
    0
    Or pSPL could be at Fp, which in my case is 20Hz. It would be nice for some clarification on this.
     
  13. Jack Gilvey

    Jack Gilvey Producer

    Joined:
    Mar 13, 1999
    Messages:
    4,948
    Likes Received:
    0
    Trophy Points:
    0
    quote: Or pSPL could be at Fp, which in my case is 20Hz. It would be nice for some clarification on this.[/quote]
    If you look under the "Table 1" tab, you'll see the "excursion" table (in blue). This lists the SPL, excursion, etc. for each frequency based on the applied power you've inputted under PE. You can see that the SPL given in the pSPL box does not occur until very high, over 1000Hz in the case of my sim, and is not the SPL at Fp. Fp can be from 3 to 6db or so down, depending on Q, from the peak. A more useful peak SPL number to use would be, say, 70Hz from that table, as most subs are crossed over and rolling off by then. It should be very close, within a db or so, to the absolute peak number, anyway.
    [Edited last by Jack Gilvey on August 07, 2001 at 10:00 AM]
     
  14. Brian Bunge

    Brian Bunge Producer

    Joined:
    Sep 11, 2000
    Messages:
    3,716
    Likes Received:
    0
    Trophy Points:
    0
    OK, I see that. I think I looked at that before but forgot about it. There's a lot of info. to digest on this spreadsheet! I guess 106dB @20Hz (at least theoretical) isn't too bad for a relatively small sealed enclosure!
    Brian
    ------------------
     
  15. Brian Bunge

    Brian Bunge Producer

    Joined:
    Sep 11, 2000
    Messages:
    3,716
    Likes Received:
    0
    Trophy Points:
    0
    Jack,
    I'm really going to annoy you now! What about proper component selection? Are 1/4W 10% tolerance resistors and disk capacitors adequate or do I need something beefier?
    Anyone else feel free to pipe up too! This doesn't have to be just a dialog 'tween me and Jack!
    Brian
    ------------------
     

Share This Page