augerpro said:
Yes, it goes way down, either run a bunch of woofers, a horn or put up with poor measured distortion
Measure an expensive pro driver it will be nearly as bad - probably the lowest distortion woofer is the JBL 2242H - That's what I use in my sub horn
Excursion equals distortion, the lower you go the higher it goes. I like to high pass my bass too (not sub bass) it cleans up the sound big time
Russell Dawkins said:
I use the little Paradigm X30, it's actually pretty good if you just use the high pass to the bass amp (it mucks up the rest if you just biamp)
Depending on where the bass panels are in the room either 50 or 70 cycles 3rd order. The 10's on the panels actually have a rise starting at 80 peaking at 60 then rolloff the same (like a camel hump) down to 45. The peak is 3-4 db. I normally have it set at 70 and get a nice smooth rolloff starting at 60 hz. 50's better if I have them further from the side walls
The Paradigm crossover has infinite low pass third order and 0 to 180 phase adjustment to lock it all in.
Here's a strangely (suspiciously?) flat 18"
http://www.partsexpress.com/pdf/264-387s.pdf
http://www.partsexpress.com/pe/pshowdetl.cfm?&PartNumber=264-387
http://www.partsexpress.com/pdf/264-387s.pdf
http://www.partsexpress.com/pe/pshowdetl.cfm?&PartNumber=264-387
mige0 said:
Shocking.
Too sad the W416Neo is the only Neo 15" I know with a medium Qts. Cantare had the Omniex 22 for half the price, but for the W416Neo that would still be 325 Euro.
http://www.olimpiaudio.com/Listini/ListinoCleveland.html
I even only know one 15" with medium Qts and cast frame, the A&D Rudolf is using. Shouldn`t there be a market for quality OB drivers?
Variac said:
Their flattest 18"ers are this
http://www.seleniumloudspeakers.com/site2004/catalogo/pdf/18ws600.pdf
and that:
http://www.seleniumloudspeakers.com/site2004/catalogo/pdf/pro_woofer_wpu1809-wpu1809-slf_new.pdf
What speaks for correct measurement is that they also show ugly curves, especially for their smaller drivers.
Latest Thoughts
Crude sketch, I know, but rather than spend all day making it look pretty in Canvas, here's what I've been thinking about.
I've used the low-Q 2nd-order crossovers since the late Seventies - the Q is low enough (0.35 to 0.5) they behave more like 1st-order filters than 2nd-order, yet an octave or two below the nominal crossover, there is much more attenuation than a basic 1st-order crossover (which reduces IM distortion from out-of-band signals, very desirable). They are also have the significant advantage of being much less sensitive to impedance peaks than even-order filters, not a small advantage when considering the cost (and sound-quality issues) of an impedance equalizer for the 6" midranges.
The quasi-Zobel networks are a little something I did in the Ariels - I started with a standard inductance compensator, and brought an adjustable pot over to the listening position. I'd adjust the pot until the sound (with pink-noise stimulus) sounded neutral and free of "hash" or roughness. Since this is more of a phase adjustment than anything else (the change in rolloff rate is not easy to see, but it is easy to hear), it's a quick way of bringing the drivers into alignment with themselves.
The composite capacitor is a techique used in professional monitors to give the best possible quality from large-value capacitors - by polarizing them to a voltage that is beyond the range of the largest possible drive signal (if this were intended for a pro application, the polarizing voltage would be higher). The small-value shunt caps are typically 100x to 1000x times smaller than the "big" caps, and may or may not be polarized, whichever sounds best. The resistors on each end prevent annoying "pops" from capacitor leakage when the driver or amplifier is connected.
The split-pole or "shaded" crossover used for the bass and mid drivers allows more precise control of the lowpass characteristic, as well as closer control of the vertical polar pattern. It also provides broader dispersion for the mids (something I want) at the nominal crossover frequency of 3 kHz. Note the series-parallel connection for the 3 drivers gives a 6 dB gain in sensitivity (for a constant voltage drive) compared to a single driver (at the lower frequencies).
The construction can be quite simple - a wood-finish flat panel with 12" deep winglets for the bass module, with a curved plexiglass/acrylic wing for the mid module that sits on top of it. If a compression driver and wavegiude is chosen for the HF, that's it, and if a flat-face tweeter like the RAAL is chosen, a smaller pair of plexiglass/acrylic wings for the tweeter would be a good idea.
The crossover can sit on the bottom plate of the bass module, with barrier terminal strips where the flying leads to the drivers (and the amplifier) attach. This method allows easy removal and transport of the separate bass, mid, and HF modules.
There are obviously lots of driver choices. I am more intolerant of resonances in the 1~5 kHz region than most, so I choose drivers based on that first, then select for other things like low distortion (although again focussing most closely on the 1~5 kHz region). Unlike Magnetar (whose advice, creativity, and inspiration are very much appreciated), I'm more of a bug about magnet types, so I'm leaning towards Alnico and neodymium where possible.
Crude sketch, I know, but rather than spend all day making it look pretty in Canvas, here's what I've been thinking about.
I've used the low-Q 2nd-order crossovers since the late Seventies - the Q is low enough (0.35 to 0.5) they behave more like 1st-order filters than 2nd-order, yet an octave or two below the nominal crossover, there is much more attenuation than a basic 1st-order crossover (which reduces IM distortion from out-of-band signals, very desirable). They are also have the significant advantage of being much less sensitive to impedance peaks than even-order filters, not a small advantage when considering the cost (and sound-quality issues) of an impedance equalizer for the 6" midranges.
The quasi-Zobel networks are a little something I did in the Ariels - I started with a standard inductance compensator, and brought an adjustable pot over to the listening position. I'd adjust the pot until the sound (with pink-noise stimulus) sounded neutral and free of "hash" or roughness. Since this is more of a phase adjustment than anything else (the change in rolloff rate is not easy to see, but it is easy to hear), it's a quick way of bringing the drivers into alignment with themselves.
The composite capacitor is a techique used in professional monitors to give the best possible quality from large-value capacitors - by polarizing them to a voltage that is beyond the range of the largest possible drive signal (if this were intended for a pro application, the polarizing voltage would be higher). The small-value shunt caps are typically 100x to 1000x times smaller than the "big" caps, and may or may not be polarized, whichever sounds best. The resistors on each end prevent annoying "pops" from capacitor leakage when the driver or amplifier is connected.
The split-pole or "shaded" crossover used for the bass and mid drivers allows more precise control of the lowpass characteristic, as well as closer control of the vertical polar pattern. It also provides broader dispersion for the mids (something I want) at the nominal crossover frequency of 3 kHz. Note the series-parallel connection for the 3 drivers gives a 6 dB gain in sensitivity (for a constant voltage drive) compared to a single driver (at the lower frequencies).
The construction can be quite simple - a wood-finish flat panel with 12" deep winglets for the bass module, with a curved plexiglass/acrylic wing for the mid module that sits on top of it. If a compression driver and wavegiude is chosen for the HF, that's it, and if a flat-face tweeter like the RAAL is chosen, a smaller pair of plexiglass/acrylic wings for the tweeter would be a good idea.
The crossover can sit on the bottom plate of the bass module, with barrier terminal strips where the flying leads to the drivers (and the amplifier) attach. This method allows easy removal and transport of the separate bass, mid, and HF modules.
There are obviously lots of driver choices. I am more intolerant of resonances in the 1~5 kHz region than most, so I choose drivers based on that first, then select for other things like low distortion (although again focussing most closely on the 1~5 kHz region). Unlike Magnetar (whose advice, creativity, and inspiration are very much appreciated), I'm more of a bug about magnet types, so I'm leaning towards Alnico and neodymium where possible.
Attachments
Nice Lynn! Glad to see it starting to take shape.
The variable R on the Zobel is slick. A friend mentioned that I should try this trick, but I never did. Guess I'll have to now, you speak so highly of it. =)
Just one question. or two.
Do you think you can hit the 103dB target in the low end? Open baffle loss is going to roll off the low end pretty fast. The staggered crossover on the bass drivers will keep it up some, but enough?
Or does it matter? You'll be going to a sub at some point anyway, right?
The variable R on the Zobel is slick. A friend mentioned that I should try this trick, but I never did. Guess I'll have to now, you speak so highly of it. =)
Just one question. or two.
Do you think you can hit the 103dB target in the low end? Open baffle loss is going to roll off the low end pretty fast. The staggered crossover on the bass drivers will keep it up some, but enough?
Or does it matter? You'll be going to a sub at some point anyway, right?
I already own a single REL subwoofer coming in at 60 Hz, and plan to replace it with a stereo pair of Rythmik 12" servo subs, with each subwoofer cube placed on the outer side of the dipole/quasicardioid speakers. I'm guessing the dipole/quasicardioid speakers will make it down to 60~80 Hz without too much trouble - the extension will mostly be a function of how much recycled cotton filling I use in the bass module and proximity effects in the room.
My intuition about commercial dipoles just not having enough radiating area was confirmed at the RMAF - actually, the situation was much worse than I thought. Commercial OB's are WAY short of effective radiating area, by a factor of several times.
There were several rooms with thin and scrawny-sounding OB's, with predictably small radiating areas and poorly chosen crossover frequencies. One room crossed 15" woofers to a very stressed-sounding 1" compression-driver tweeter, and several other exhibitors demonstrated that a pair of 10" low-efficiency audiophile drivers (in an OB) couldn't even fill a pint-sized hotel room. Not even close to what horns or pro gear can do.
My instinct now is that two or more times the radiating area of a professional studio monitor is required for reasonable dynamic range in a domestic setting. Putting the drivers next to the floor helps some, and I was surprised nobody was doing this at the show.
P.S. In terms of esthetics, a gloss-black Bass module (with a quartet of mix-n-match 12" or 15" drivers) combined with a plexiglass/acrylic Mid and HF modules would be easy to make and not as visually oppressive as a single-piece all-wood baffle taking up a large part of the room. If you like the old-school look, you could even make the Bass module look like a Marshall stack, if you're into that kind of thing.
The separate modules for each part of the frequency range has the huge advantage of allowing time alignment tuning to be done precisely, easy transport, and easy driver swaps and re-designs for each part of the spectrum. Like horns better than ribbons or an array of direct radiators? Swap the module and twiddle the crossover, all done.
My intuition about commercial dipoles just not having enough radiating area was confirmed at the RMAF - actually, the situation was much worse than I thought. Commercial OB's are WAY short of effective radiating area, by a factor of several times.
There were several rooms with thin and scrawny-sounding OB's, with predictably small radiating areas and poorly chosen crossover frequencies. One room crossed 15" woofers to a very stressed-sounding 1" compression-driver tweeter, and several other exhibitors demonstrated that a pair of 10" low-efficiency audiophile drivers (in an OB) couldn't even fill a pint-sized hotel room. Not even close to what horns or pro gear can do.
My instinct now is that two or more times the radiating area of a professional studio monitor is required for reasonable dynamic range in a domestic setting. Putting the drivers next to the floor helps some, and I was surprised nobody was doing this at the show.
P.S. In terms of esthetics, a gloss-black Bass module (with a quartet of mix-n-match 12" or 15" drivers) combined with a plexiglass/acrylic Mid and HF modules would be easy to make and not as visually oppressive as a single-piece all-wood baffle taking up a large part of the room. If you like the old-school look, you could even make the Bass module look like a Marshall stack, if you're into that kind of thing.
The separate modules for each part of the frequency range has the huge advantage of allowing time alignment tuning to be done precisely, easy transport, and easy driver swaps and re-designs for each part of the spectrum. Like horns better than ribbons or an array of direct radiators? Swap the module and twiddle the crossover, all done.
Hi Lynn,
Considering mid to upper bass radiating area and how much is enough. Can you provide a guideline for roll off points that might be useful?
For instance, an Eminence Beta 8 begins to roll off at 150Hz and is 6 dB down at 70 Hz. Will this be adequate for use as an upper bass driver, in concert with a transmission line for bass? Should I be planning on two of these drivers for the 75 Hz to 300 Hz realm?
I don't have need of the power levels you are reaching for, as 103 dB will be quite loud in my situation. What would you suggest please?
Bud
Considering mid to upper bass radiating area and how much is enough. Can you provide a guideline for roll off points that might be useful?
For instance, an Eminence Beta 8 begins to roll off at 150Hz and is 6 dB down at 70 Hz. Will this be adequate for use as an upper bass driver, in concert with a transmission line for bass? Should I be planning on two of these drivers for the 75 Hz to 300 Hz realm?
I don't have need of the power levels you are reaching for, as 103 dB will be quite loud in my situation. What would you suggest please?
Bud
OB's are brutal in their cone-area requirements. To merely keep up with their closed-box equivalents, cone area has to DOUBLE for every octave below the baffle peak. Think about that for a while.
It's not like conventional closed-box direct-radiators are doing all that well, either. Excursion goes up at a rate of 12 dB/octave - that's why they are called constant-acceleration devices (the acceleration is constant with frequency, velocity goes up 6 dB/octave, and excursion goes up 12 dB/octave). For OB's working below the baffle peak, the situation is even worse, with the required increase in excursion 18 dB/octave!
Classical music has an energy peak around 300~500 Hz, so this give us a hint that we need serious radiating area in this region, aside from other considerations. My rule-of-thumb is simple: if you can see the cone moving - ever - it is distorting. Count on it.
This is why the horns have it all over the direct-radiators. It's too bad I don't care for horn colorations, since physics are on their side, with horn-gain providing the maximum benefit towards the lower portion of the horn's working spectrum.
The only other alternative is increased cone area, and this is limited by beaming artifacts as the frequency increases. Fortunately, excursion requirements drop rapidly with increasing frequency (much faster than beaming becomes a problem), so we can stay out of trouble here.
I've heard Gary Pimm's Beta-8 based system. His listening room is fairly small, and I think the midrange dynamics are only just OK - about on a par with my Ariels, which are not really dynamics champs. I feel the same way about Lowthers used in direct-radiator or rear-horn setups - OK but not fabulous dynamics. The exact same Lowthers in an Oris or LeCleac'h front horn are another animal entirely, with dynamics that put a Avante-Garde Trio horn system to shame. Stupendous impacts and remarkably subtle and vivid tonality, a completely different sound.
It's just too bad I don't care for that residue of horn coloration, which I cannot ignore. I wish I could. I'd buy the appropriate LeCleac'h profile horn, a suitable widerange driver (without that awful whizzer!), and listen happily for many years afterward.
I can't honestly recommend any Eminence driver, except possibly for use below 200 Hz. I've yet to hear one that can match my 15-year-old Vifa polypropylene mids, which are only just OK by modern standards. This is one of those personal-taste things, though - people I respect, including Gary Pimm, like them more than I do.
The serious prosound stuff intended for studio-monitor use (TAD, JBL, Italian and German pro drivers) are really good. The 10" TAD I heard in the Audio Kinesis room was stunning, no other word for it. The best pro stuff, along with the audiophile-efficiency confections from Skaaning, are truly top-class and worth the money.
It's not like conventional closed-box direct-radiators are doing all that well, either. Excursion goes up at a rate of 12 dB/octave - that's why they are called constant-acceleration devices (the acceleration is constant with frequency, velocity goes up 6 dB/octave, and excursion goes up 12 dB/octave). For OB's working below the baffle peak, the situation is even worse, with the required increase in excursion 18 dB/octave!
Classical music has an energy peak around 300~500 Hz, so this give us a hint that we need serious radiating area in this region, aside from other considerations. My rule-of-thumb is simple: if you can see the cone moving - ever - it is distorting. Count on it.
This is why the horns have it all over the direct-radiators. It's too bad I don't care for horn colorations, since physics are on their side, with horn-gain providing the maximum benefit towards the lower portion of the horn's working spectrum.
The only other alternative is increased cone area, and this is limited by beaming artifacts as the frequency increases. Fortunately, excursion requirements drop rapidly with increasing frequency (much faster than beaming becomes a problem), so we can stay out of trouble here.
I've heard Gary Pimm's Beta-8 based system. His listening room is fairly small, and I think the midrange dynamics are only just OK - about on a par with my Ariels, which are not really dynamics champs. I feel the same way about Lowthers used in direct-radiator or rear-horn setups - OK but not fabulous dynamics. The exact same Lowthers in an Oris or LeCleac'h front horn are another animal entirely, with dynamics that put a Avante-Garde Trio horn system to shame. Stupendous impacts and remarkably subtle and vivid tonality, a completely different sound.
It's just too bad I don't care for that residue of horn coloration, which I cannot ignore. I wish I could. I'd buy the appropriate LeCleac'h profile horn, a suitable widerange driver (without that awful whizzer!), and listen happily for many years afterward.
I can't honestly recommend any Eminence driver, except possibly for use below 200 Hz. I've yet to hear one that can match my 15-year-old Vifa polypropylene mids, which are only just OK by modern standards. This is one of those personal-taste things, though - people I respect, including Gary Pimm, like them more than I do.
The serious prosound stuff intended for studio-monitor use (TAD, JBL, Italian and German pro drivers) are really good. The 10" TAD I heard in the Audio Kinesis room was stunning, no other word for it. The best pro stuff, along with the audiophile-efficiency confections from Skaaning, are truly top-class and worth the money.
Re: Latest Thoughts
Well, the Neo TAD 4002 driver sounds better than the alnico 5 4001 but the ferrite 1603 is better than the alnico 1601, but the JBL alnico 2240 sounds better than the ferrite 2245 and the NEO 2450 and so on and so on and so on........... I don't care what the magnet is made out of (RIP Harvey) I care what sounds best to me for the application.. and sometimes it comes down to what I have in the back room........
Glad to see this project coming along!
Lynn Olson said:
Well, the Neo TAD 4002 driver sounds better than the alnico 5 4001 but the ferrite 1603 is better than the alnico 1601, but the JBL alnico 2240 sounds better than the ferrite 2245 and the NEO 2450 and so on and so on and so on........... I don't care what the magnet is made out of (RIP Harvey) I care what sounds best to me for the application.. and sometimes it comes down to what I have in the back room........
Glad to see this project coming along!
agent.5 said:
They will be very difficult to get! Not that many out there now..
If you want a good 10 mid forget the TAD and get a JBL 2123, it's easy to get and probably a magnitude better through the midrange. The same guy designed it except in the JBL's case it was designed as a NO compromise midrange, not a midbass driver