Dismiss Notice

Psst... Ready to join TalkBass and start posting, make new friends, sell your gear, and more?  Register your free account in 30 seconds.

Slot mounted 15" driver in DIY cab?

Discussion in 'Amps and Cabs [BG]' started by fdeck, Dec 2, 2005.


  1. fdeck

    fdeck Supporting Member Commercial User

    Mar 20, 2004
    Madison WI
    HPF Technology LLC
    Long ago, I saw TOA speakers where the drivers were mounted behind narrow slots. I was thinking that if I put a 15 x 10 inch baffle in front of my 15" speaker, it would give me a dispersion characteristic more like a 2x10 cab.

    There must be a downside, but I can't think of any. What will happen to the on-axis sensitivity?

    Thoughts?
     
  2. The Clap

    The Clap

    Jan 5, 2004
    Scottsdale, AZ
    It doesn't work like that, unfortunately. What you're trying to do is accomplished by changing the radiating area, IE, using smaller speakers, not just smaller openings.
     
  3. fdeck

    fdeck Supporting Member Commercial User

    Mar 20, 2004
    Madison WI
    HPF Technology LLC
    But the opening becomes the radiating area. Nothing comes out of the system that doesn't come out of the opening.

    Making a comparison to optical waves, or even water waves, the diffraction pattern from an aperture is determined by the dimensions of the aperture.

    I guess what I really want to know is more of a physics type explanation about what happens to the on-axis sensitivity.
     
  4. fdeck

    fdeck Supporting Member Commercial User

    Mar 20, 2004
    Madison WI
    HPF Technology LLC
    I may be on the way to answering my own question. Nothing happens to the on-axis sensitivity assuming that the aperture isn't small enough to form a Helmholtz resonator with the air space behind it.
     
  5. Tim__x

    Tim__x

    Aug 13, 2002
    Alberta, Canada
    Though a neat idea, I don't think it's very practical. At higher frequencies where dispersion is a problem the wavelength will be small compared to the length of the chamber which may cause problems with reflection and resonance.

    It's a difficult unintuitive problem, I don't think anyone can give you a concrete answer without experimental data.
     
  6. billfitzmaurice

    billfitzmaurice Commercial User

    Sep 15, 2004
    New Hampshire
    Owner, Bill Fitzmaurice Loudspeaker Design
    Not so fast there, voice-coil breath.

    Slot loading can be used to increase horizontal dispersion, using a vertically aligned slot. There are slot loaded tweeters, such as the Selenium ST324, which do just that. But, as fdeck noted, the width of the slot must be considerably less than a wavelength to be effective. Say you want to increase dispersion at 1kHz. A quarter-wavelength wide slot would do so, and that requires about a three-inch wide slot.

    But that three-inch wide slot also creates a chamber between the cone and the baffle. That chamber transforms a sealed box into a 4th order bandpass, and a reflex box into a 6th order bandpass, in both cases with a 2nd order lowpass function on the front wave output of the driver. Whatever you gain in dispersion is lost by the low pass filtering, and more. For that reason slot-loading with woofers is used to extend low-frequency response, not to control dispersion.
     
  7. fdeck

    fdeck Supporting Member Commercial User

    Mar 20, 2004
    Madison WI
    HPF Technology LLC
    Bill, thanks for a nice cogent reply. Also, your comments about power compression have caused me to re-think some of the concepts that I had been considering for combining various drivers and amplifiers. I won't bother gassing for a 300-Watt amp to use with my 300-Watt driver, since I can easily get whatever compression I want electronically.

    For amusement, I also looked up the resistivity of copper versus temperature. In order to have 3 dB power compression, the voice coil has to get pretty darn hot. Now I understand why the coil form has to be made of something like Kapton.
     
  8. billfitzmaurice

    billfitzmaurice Commercial User

    Sep 15, 2004
    New Hampshire
    Owner, Bill Fitzmaurice Loudspeaker Design
    Power compression sourced purely from heat in the coil is only the tip of the iceberg. Compression from mechanical sources is even more of a consideration. For instance, a 350 watt rated Delta 10 in a typical 1 cu ft VB tuned to 50 Hz runs out of xmax at 80 Hz with only 15 watts input. Even a 4x10 loaded with Delta 10s can't utilize more than 60 watts input from 80 Hz on down. One more reason why when choosing drivers almost the last thing you need be concerned about is their power handling.
     
  9. 62bass

    62bass

    Apr 3, 2005
    It makes a lot of sense Bill. I've claimed that for years but didn't have the technical expertise to back it up. And that's one of the reasons I don't care much if my 300 watt head is driving an 8 ohm load. I know from experience that my 15 Kappa Pro will break up and sound ugly by the time it reaches 150 watts. But that speaker is so efficient that's a lot of volume and more than I need for a small club gig. Bigger rooms I can add another cabinet.

    By the way, last night I heard a 3 piece group play in a local watering hole. It was a small room and the bass player was using one of the older Trace Elliot combos with a 10 in a bandpass cabinet. Plenty of volume and he was using a 5 string. They didn't play at stupid volumes but loud enough for rock.
     
  10. Tim__x

    Tim__x

    Aug 13, 2002
    Alberta, Canada
    Sometimes I'm glad to be proved wrong.

    Really? (Not an expression of disbelief, only of suprise) It seemed too obvious.
    Do ported chambers still behave near-ideally at high frequencies (frequencies where wavelength is comparable to various internal dimensions)?
     
  11. billfitzmaurice

    billfitzmaurice Commercial User

    Sep 15, 2004
    New Hampshire
    Owner, Bill Fitzmaurice Loudspeaker Design
    I did an experiment once to find out which was more influential on the lowering of system resonance in a horn loaded cab, the length of the horn or the size of the throat opening. I found throat size was much more significant until the horn was at least three feet long. You can drop the effective fs of a driver 10 Hz or more by slot loading. It's a very easy way to get deeper response, but the low pass function kills midrange response, forcing you use a midrange driver. That's why you don't see slot-loaded 2x10s.
    I'm not clear on what you're referring to there.
     
  12. Tim__x

    Tim__x

    Aug 13, 2002
    Alberta, Canada
    Let me try again, my wording isn't always very clear.

    How does a ported system behave near port resonance?
    And what happens when port resonance is near the frequency the box is tuned to (fb).
     
  13. billfitzmaurice

    billfitzmaurice Commercial User

    Sep 15, 2004
    New Hampshire
    Owner, Bill Fitzmaurice Loudspeaker Design
    The port doesn't have a resonance, it's volume and area in combination with the box volume will dictate the system resonance (fb). The two are inseparable. At that frequency there is maximum output from the port, while cone excursion is minimal. Below that frequency response rolls off at 24dB/octave.
     
  14. fdeck

    fdeck Supporting Member Commercial User

    Mar 20, 2004
    Madison WI
    HPF Technology LLC
    I have valiantly attempted to sort out the excursion limit issue. My tentative theory is based on knowing that a bass doesn't put out a pure sinewave. Indeed, the bass signal seems to model fairly well as an equal weighting of the fundamental and the first couple of overtones.

    For my analysis, I plot excursion versus frequency at some nominal input power, such as 1 Watt.

    I assume that the cone excursion amplitude for any given bass tone is the average of three points on the curve, at frequencies f, 2f, and 3f. The worst case is usually when 2f is on top of the resonant hump, and the other two harmonics on the two sides. For some cabinets that are prone to "farting out," there is another worst case point corresponding to the lowest note of the bass. I tend to avoid considering designs that have that problem.

    For the bass speaker designs that I have considered, the two sides are usually halfway down from the top. Thus, the actual excursion is roughly 67% of the peak excursion shown on the graph. You could use an "effective" excursion limit of roughly 1.5x the rated Xmax value. This would let you roughly double the maximum amplifier power at which the speaker can operate, when considering only the excursion limit.

    Anyhoo, that's my theory. It's the best I can do to analyze the behavior of a speaker for real bass signals. It doesn't take into account the phase of the harmonics. Even when this analysis is used, a lot of candidate speaker designs still have Xmax limited power handling below their thermal power rating. For instance, it doesn't offer much help to the Delta 10's in Bill's example.

    My oddball Eminence 15" has a 5 mm Xmax, and in its tiny DIY cabinet, the max power graph in WinISD is a flat line at the driver's rated power of 300 W. Thus I am assuming that power compression would actually be the limiting mechanism for this particular system.

    I am thinking of buying either a GK Backline 600 or one of the new Clarus heads to put on top of this speaker.
     
  15. billfitzmaurice

    billfitzmaurice Commercial User

    Sep 15, 2004
    New Hampshire
    Owner, Bill Fitzmaurice Loudspeaker Design
    Since you insist, xmax isn't the only issue. Xmech also comes into the picture. While many drivers have insufficient xmax to remain linear in response in the bass to even 1/4 (-6dB) of their rated wattage, most prosound drivers have an xmech (maximum excursion before the voice coil bottoms out) of at least 4 times xmax, and 10 times isn't unusual. So if you take, for instance, a Delta 10 with an xmech of 1.6mm that won't take more than 15 watts at 100 Hz, it's xmech of 18mm means you still can feed it 350 watts without hurting it. A short xmax and long xmech is like having a built in limiter. Problems arise when xmax and xmech are close to each other and the driver can be damaged by overshoot, something often seen in hi-fi subwoofers.

    But to your particular situation, yes, if you can run 300w without exceeding xmax, let along xmech, then mechanical compression isn't a concern, and you're probably getting plenty of SPL so that heat derived power compression is no concern either. Some designers get upset when xmax isn't linear all the way down to fb, but that's seldom a problem in real world applications. In the first octave the power demands of the fundamental usually run at least 10dB less than that of the second harmonic, so if xmax at half power is not exceeded only to 60 Hz or so that's usually sufficient anyway.
     
  16. fdeck

    fdeck Supporting Member Commercial User

    Mar 20, 2004
    Madison WI
    HPF Technology LLC
    Again, thanks for the useful information. Admittedly, trying to understand this stuff may be more of an obsession for me than a practical issue. You've helped out a lot.
     
  17. Tim__x

    Tim__x

    Aug 13, 2002
    Alberta, Canada
    I'm talking about the quarter-wave resonance of the port, which can be changed independantly of fb.
     
  18. billfitzmaurice

    billfitzmaurice Commercial User

    Sep 15, 2004
    New Hampshire
    Owner, Bill Fitzmaurice Loudspeaker Design
    A port does not have a quarter-wave resonance. A duct has a quarter-wave resonance, but usually the frequency at which it occurs is far too high to be of any consequence. A duct more than a foot long is a rarety, and the quarter-wavelength resonance of a duct even that long would be 282 Hz. A duct with a quarter-wave resonance of 50 Hz would have to be 5.6 feet long, and at that point you no longer would have a reflex alignment, you'd have a transmission line.
     
  19. Tim__x

    Tim__x

    Aug 13, 2002
    Alberta, Canada
    That's just semantics, any real life port has a nonzero length and a quarterwave resonance.

    In any normal vented box yes, in a midrange bandpass you might near resonance.

    Thank you, this confirms my suspicions and is the answer to the question I so ineloquently put forward.