Using WinISD to completely design a cabinet?

Discussion in 'Amps and Cabs [BG]' started by Magneto, Feb 1, 2005.

  1. I've been over some of this before in my other cabinet threads.

    I'm wondering how many of you have done this and had good results the first time around. I'm wondering how accurate this program is, which of course leads me to wonder how accurate the tuning would be using their port calculations.

    WinISD's data is conflicting with just about everything I've read at the Eminence website.

    Let's take the Delta PRO 15 (8 ohm) for instance: The spec sheet on this driver "suggests" a cab volume of 2.25-3.75 cu-ft. PartsExpress suggests a volume of approx. 4.0 cu-ft.

    However, WinISD is showing that the smoothest freq plot requires a cab volume of approx. 8.4 cu-ft, with a tuning frequency of 42.37 hz. Wow, that's one big honkin' cab for a single 15 driver. Ok, sure that means that the default, smoothest plot would be obtained using this data, but that is so far off what the Eminence data suggests, that it leads me to wonder. If you were to use this particular driver for bass guitar, would you go ahead and build an 8.4 cu-ft cabinet for it? If not, what tradeoffs would you make? I'm not looking for a thunder-bass cabinet, but at the same time I would like to have a cab that would work well with my Behringer 300 watt amp.

    I talked to Avatar, and they told me that they tuned their B115H cab to approx. 49hz, which is what Eminence's engineers calculated using the Eminence Kappa Pro LF2 and a cabinet volume of approx. 4.0 cu-ft, (which by the way is significantly different than what WinISD is suggesting.) Of course this a COMPLETELY DIFFERENT driver with different characteristics.

    Surely I have alot to learn, but the more I research and model drivers in various volume/tuning frequency arrangements, the more confused I am getting.

    Could someone with experience model the Eminence Delta pro 15 using WinISD, using a box volume of 4.2 cu-ft, and let me know what they would try to tune this cabinet to, and why? I will be using these drivers in my cabinet, and I don't have much flexibility in cabinet volume. I'm not expecting miracles or perfection, and I promise I won't come back later griping about undesirable results.

    Thank you for your help with this.. again..

  2. Start with my spread sheet, get an idea of which driver to use. Plug it into WinISD.
  3. msquared


    Sep 19, 2004
    Kansas City
    What does the frequency response graph from PartsExpress look like? Is it close to what the WinISD graph for a 4.0 ft^3 box looks like?

    What one person thinks is "optimal" isn't always what another person thinks is "optimal". PartsExpress might be taking movability or car audio installation into consideration while WinISD is always going to default to whatever is the flattest frequency response on paper as an optimum.
  4. Petebass


    Dec 22, 2002
    QLD Australia
    I'll run that speaker through WinISD Pro when I'm next at home.

    As a general rule, never ever believe the speaker manufacturers when they give you recommended cabinet dimensions. The same goes for those trying to sell you a speaker.

    When you're talking maths, 1+1 is always 2. If WinISD is calculating 8cu.ft, i'd say it is correct for the "Optimal Flat Response Alignment". Bgavin's spreadsheet confirms the math is correct for that Alignment.

    However it also means you're using the old version of WinISD. The Pro version lets you choos the other alignments as well. For instance the SBB4 alignment results in a cab size of 6.19cu.ft. Tuned to 42Hz you get a -3dB point of 47Hz.

    I's sooner trust WinISD, especially the Pro version, over anything a salesman tells me.
  5. I have your spreadsheet, but am trying to learn how to interpret it. I do not know what SBB4 alignment is, or many of the abbreviations in that sheet. I'm having a hard time finding info on this.
    I assume they are referring to enclosure types, etc..
    I don't own WinISD pro. I only have the beta.
    I'm not sure how PartsExpress come about their suggested 4.0 cu-ft enclosure volume. Seems to go hand in hand with what Eminence is recommending.
    Like I probably already mentioned, I don't need a perfect box. I would really like to hear what you guys would shoot for if you were trying to tune a bass guitar cabinet :
    A: Using the Delta pro 15,
    B: Installing it in a 4.2 cu-ft enclosure.
    Might not be the best speaker or cab size choice, but it's what I've got. I'm sure it will work, and probably won't sound half-bad, I just don't want to guess at a tuning frequency.
    Thanks for helping..


  6. I plugged the Delta pro 15 into WinISD Alpha and it seems like anywhere between 40 and 45hz tuning frequency will give you the best overall flat response, power handling, and group delay. I would aim for 42hz, which happens to be the drivers fs.

    To set the tuning frequency to 42 hz you could use two 3 inch diameter ports that would need to be 5.5 inches long each.
  7. The Delta 15 is also a possible candidate for a sealed box, and it just so happens the the optimal flat response is in a 4.2 cu ft box. It would have a much better group delay which will make it sound more defined. The downside is you would be down 3db at 76 hz as opposed to 56 hz in the ported box.

    FWIW I don't think you need the extra 20hz. I would try it sealed first and simply add a port (or ports) later to see what you like better. You can always block the ports later.
  8. MuzikMan, Thanks a bunch for helping with this. I just installed the WinISD pro alpha this morning, and have been playing with it. It is quite a bit different from the beta, and has alot of helpful info attached with it.

    What's the deal with the different alignment choices? I have no idea what these are, am not finding the info in WinISD's help, and only finding vague information searching the web.

    What's the alignment for a basic bass guitar cab, vented? BB4/SBB4? QB3 Quasi-Butterworth? C4/SC4?
    The WinISD help file seems to just skip right past this option..

    Port construction: Any tips on what materials to use? PVC pipe? Glue it into the baffle board, flush with the front face of the baffle?
    What I was considering was boring the proper-sized vent holes, then glueing the vents in, but that's gonna screw me if I find I need to adjust anything, and I'll bet I'm gonna have to do just that.
    Any decent attachment methods that allow for removal and fine-tuning?
    Do the ends really need to be flared? Would it help any to bevel the inside edges of the PVC to keep port noise down?

    As always, thanks a bunch for helping..

  9. BruceWane


    Oct 31, 2002
    Houston, TX
    My guess is that the cab volume and tuning suggested by Eminence is oriented toward protection from overexcursion (i.e. maximum power throughout frequency range), rather than flattest or most extended frequency response. It's probably somewhere in between minimising excursion and ideal frequency response; probably still reasonably flat, as it wouldn't be too smart to recommend a cab design that sounds like crap.

    ....and the cab volume/porting suggested by WinISD doesn't appear to consider maximum power at all, so you'll want to look at that graph as well as the frequency response when you design your cab. Designing a cab that is -3db@33Hz is worthless if it'll only handle 20 watts below 100Hz.
  10. BruceWane


    Oct 31, 2002
    Houston, TX
    I just loaded up the specs for the Delta Pro 15 in WinISD.........and from the looks of it, I don't see any use for this driver. Even in a very small box, at bass frequencies it'll be exceeding it's x-max at very low wattages. Exceeding x-max (voice coil leaving the gap) doesn't mean the speaker is necessarily farting out (exceeding x-mech - voice coil hitting the back plate), but it does mean that it's likely to start distorting somewhat. You'd have to keep it in a small box tuned very high to keep it clean.

    When you play around with WinISD, try raising and lowering box size and tuning while keeping an eye on ALL the different graphs - especially frequency response (transfer function magnitude) and maximum power. The best frequency response curve almost always yields a very poor maximum power chart, so you'll probably want to find a design somewhere in between. Something with a pretty good frequency response that still keeps the voice coil in the gap.

    Example - try a Kappa 15LF in a 3.0 cubic foot box tuned to 50Hz. Max power shows a dip to 300 watts between ~65Hz and ~90Hz, but that's not that bad, probably a bit of distortion, might not even be noticeable........try the Delta Pro 15 in the same size box.......max power dips to 100 watts between ~55Hz and ~160Hz......that's a bit of a problem........then looking at the response curves, the kappa wins there too........
  11. BruceWane,

    I've read both of your replies. Very interesting. Do you know what the different alignment choices in WinISD pro alpha mean?
    I don't want to proceed until I can figure this out. I don't want to choose some funky subwoofer box or something like that, and end up with a car boombox?
    I asked over at the WinISD forum.. no replies.. I guess they don't know what they are either. Did a google search, came up with tons of math calculations, no explanation about these different types.

    Oh well.. I guess it's time to buy a book.

    Thanks again..
  12. If you set a port in winISD to say 2"x21" and it gives a length, it is safe to use that length for a slot port of those dimensions correct?
  13. ESP-LTD


    Sep 9, 2001
    You've decided what driver to use, and what cabinet volume to use, and that it will be a ported box.

    I would tune it to the lowest note you expect to play. As I understand it, if you play below the tuning frequency of the box, the driver will have little support against over-excursion. The trade off will be that you will sacrifice some low end between the Fs of the driver and the Fb of the box.

    I would start by deciding what you want- low, loud, or small, and then approach it from a budget perspective- how much of "low/loud/small" you can get for your price point. That will determine a speaker to use and you go on reducing variables from there.
  14. Run on faith, then.


    If you want tight bass, go SBB4.
    If you want the lowest extension, look at Flat or whatever has the lowest F3 between SBB4/QB3/Flat. I prefer SBB4.

    Next, decide what driver size you want to use.
    After that, sort the sheet by driver sizes, and Box Volume (Vb) so you can figure out what cabinet you want to live with. Look at various drivers in that size and figure out which one has a tuning (Fb) at or near your lowest note, such as 41 Hz for low E.
  15. BruceWane


    Oct 31, 2002
    Houston, TX
    As previously stated, you don't want to tune the cab much higher than the lowest note you'll be playing (~40Hz for E, ~30Hz for B) because the cabinet "unloads" the driver below tuning frequency, i.e. it offers no resistance to the motion of the cone at frequencies below tuning. This allows cone excursion to go unchecked. Check out the cone excursion graph on any design in WinISD and you'll see that excursion goes off the chart below the tuning frequency. You've got a little - just a little - elbow room where the excursion is rising but still at an acceptable level below tuning frequency, so you can usually get away with, for example, tuning a cab intended for 5 string use to 35Hz instead of 31Hz. It doesn't matter if excursion is off the chart at 20Hz when you're not going below 31Hz.

    (Note - as I recall, WinISD shows excursion on the chart as a "peak-to-peak" measurement. Driver x-max is usually listed as a "peak" measurement, so a driver with an x-max of 3mm will have a "peak-to-peak" measurement of 6mm - 3mm from resting point to maximum "push", plus 3mm from resting point to maximum "pull" of the cone. If you don't keep this in mind, the cone excursion charts in WinISD will make ALL drivers look like they're grossly exceeding x-max.)

    The different alignments just offer different balances of frequency response versus group delay. Low group delay means tighter, more defined bass. But going purely for low group delay doesn't get you the best frequency response. It's a balancing act, and the choices are up to you. Personally, I like to stick to designs that allow the driver to handle pretty close to it's max wattage throughout the frequency range, as this usually results in a cab that is smaller (more portable), will have a low group delay (tight & defined), and can be cranked to high levels if necessary without farting out. The trick is finding a driver that will have both a good frequency response in this kind of design AND enough efficiency to be loud enough for real-world use. It's a very tricky balancing act, ESPECIALLY if you're going down to B or lower.
  16. BruceWane,

    Once again, thanks for taking the time to explain this the way you did. It's making more sense.
    One thing I'm trying to figure is how to make sense of the group delay graph.
    Man, the WinISD definition doesn't say much about what a person should be looking for. I'm "assuming" that the smoother the plot on that graph, the more even the delay is. High spikes (which most drivers seem to have near low frequencies) are not desirable, right?

    I see what you mean about power handling. It really is a juggling act, and I promise I've been doing some serious jugglin'! I still haven't ordered my drivers yet. I don't want to buy something that is a waste of money, using them in cabinets that don't do them justice.

  17. Petebass


    Dec 22, 2002
    QLD Australia
    The trick is to make sure to group delay peak doesn't go over 20m/s. There will always be a peak at the low frequencies, so don't let that bother you.

    Generally speaking SBB4 will yield the best group delay of all of the vented alignments. Sealed cabinets go better again, but they have other characteristics which need to be considered before you'd consider it for a bass guitar design.

    As you're finding out, the balancing act can be quite delicate. But gee it's fun, not to mention rewarding.........