all right i'm attaching some sketches to help you visualize what i have in mind ...
this design has gone through a bunch of iterations and some aspects of it keep sort of vacillating which i will address later ...
for now i will just describe what the current iteration is, as depicted in the sketches ...
5-way system consisting of a center subwoofer stack that goes between the two main 4-way speakers ...
both speakers and subwoofer stack are 80" tall ( to fit in basements ) and the width is 32" for speakers because that's 1/3 of 96" that plywood comes in ... and 2 x 32" for the subwoofer stack. each subwoofer module is 25.5 x 32 x 40 inches, with a total of four modules. 25.5" is just 24" + 2 thicknesses of plywood. the main speakers would have approximately similar depth ( about 2 feet ) but since they curve towards the listener they don't have any hard number for their depth ...
in my sketches you can also see a 75" TV mounted on the subwoofer stack ... that is indeed where the TV will go but it doesn't have to be 75" ... although the width of 75" matches the width of the subwoofer stack exactly, which may or may not be a coincidence ...
total system width would be about 12 feet which is about as narrow as i could get it but it should allow it to fit in most living rooms outside of NYC and SF.
in the current iteration of the design the driver compliment is as follows
25 hz - 80 hz:
https://www.eighteensound.it/en/products/lf-driver/21-0/4/21ntlw5000-4350 liter net box, vented, tuned to 25 hz ...
four such boxes in the center behind the screen ...
80 hz - 200 hz:
https://www.parts-express.com/JBL-2262HPL-338312-004X-12-Neo-Woofer-294-45040 liter net box, vented, tuned to 70 hz
four 8 ohm drivers per side for a total 2 ohm load
200 hz - 800 hz:
https://www.eighteensound.it/en/products/lf-driver/8-0/8/8NTLW20005 liter net box, vented, tuned to 130 hz
four 8 ohm drivers per side for a total of 2 ohm load
800 hz - 3 khz:
https://www.bmsspeakers.com/index.php-62.html?id=bms_4594nd-midon the following horn:
https://www.bcspeakers.com/en/products/horn/1-4/0/ME4648 ohms ...
3 khz - 20 khz
https://www.bmsspeakers.com/index.php-48.html?id=4526nd2 x 2 array of 8 ohm drivers in series parallel for a 8 ohm load
the horn array will consist of four identical but not symmetrical horns that must be 3D printed
that is to say the arrangement of the four horns will be symmetrical but the horns themselves will be asymmetrical
the asymmetry will be used to bring the drivers as close to each other as possible, with magnets almost touching each other
all in all 8 identical horns will have to be 3D printed from a single file
main crossover will be MiniDSP 4X10 HD and it will plug in directly into the TV via optical Toslink ... i will throw some protection caps on the compression drivers to be safe as well ... the 80 hz crossover point will be realized internally to the 4-channel DSP amplifier that will power the woofers and subs
the TV will ultimately be the source, it will probably be a 2022 Mini-LED TV from either Samsung or Sony. i had a 2021 Samsung Mini-Led that i used as a source like this via optical output and it worked really well as the TV could do YouTube, Netflix and Spotify ... unfortunately it was a first generation Mini-LED and had a flicker so i sent it back to Costco ... in 2022 Samsung is coming out with 2nd generation Mini-LED and Sony is entering the Mini-LED market as well ... it will take time for prices on new models to drop but it will also take time for me to design and build this system ...
there will be a total of 3 amplifiers:
QSC PLX 2402 ( which i already own ) will power the 8" midbass array at 2 ohm, and will have a Noctua fan mod
QSC CX 404 ( four channel ) is $400 used on Ebay and will power all compression drivers, it will have the same Noctua fan mod
JBL Crown DSI 2.0 MA4 ( four channel ) will power both subwoofers and 12" woofers. this amp is expensive but it packs a lot of power and DSP and i may need those.
now for where the design keeps vacillating ...
i sometimes wonder if it's best to use 12" or 15" JBL woofers and whether it is best to use 8" or 6" Eighteen Sound tetracoil midbass ...
i sometimes wonder if the midbass should be vented or sealed ...
i sometimes wonder whether to use BMS 4592 mid, 4594 mid or 4599 mid ... the 1.4" throat versions can be used with B&C ME464 horn while the 2" versions would need either SEOS 30 horn or a DIY horn with 3D printed throat adapter and plywood mouth
i sometimes wonder if the system should be 5-way or 6-way ... in case of 6-way it would have an array of Six Beyma CP-12 Supertweeters crossed at 8 khz ...
i sometimes wonder if for tweeters i should go with BMS 4526 or Faital HF106 ... and in case of faital whether to use 1, 2, or 4 ... or if using Supertweeters perhaps a single Faital HF1440 ...
the reason i am using an array of tweeters like this is 2 fold:
- it lower distortion by quadrupling throat area thus lowering compression and distortion ...
- it doubles the coverage angle at the extreme high frequencies where horns begin to beam ...
yes i realize there is the issue of comb filtering and that's why i keep vacillating on what is the best solution in that area ...
as for what i'm trying to achieve with this is it is intended to be my main system for the foreseeable future - it isn't intended as any sort of exercise - it is intended to be THE system i use whenever i want to listen to music or watch music videos or movies at maximum volume ... it is intended to have the absolutely maximum performance level possible within what can be achieved in a system of reasonable size and cost ... spouse approval factor is not a consideration but i want to challenge myself by keeping it as compact and cost effective as possible - i could not respect myself if i went sloppy and simply started to mindlessly pile on heaps of subwoofers using "the more the merrier" approach ...
this system will be acoustically in a large space as the house is an open concept plan with one continuous space spanning from the porch to the patio and also both first and 2nd floors ... it's basically in a "living room" but it's also the same space as an eat in kitchen and sunroom and pretty much open to the entire rest of the house except bedrooms and bathrooms. the floor is wooden as there is basement below. the ceiling is sloped from 10 to 20 foot high. all other surfaces are fairly thin drywall and lots of windows - the house is mostly acoustically transparent - i don't expect much standing wave issues as i used to have in an apartment in a concrete building in NYC. on the other hand where i could fill that apartment with a single sealed 18" TC Sounds LMS subwoofer because those cinder block walls and concrete floors and ceilings trapped the bass and kept it resonating in the space ... by contrast in the space i'm currently in the bass will simply spread through the house and then leak out to the street so i need much higher bass output even for my listening position that will be about 8 feet away from TV screen ...
i have also been on a professional sound reinforcement forum discussing hearing damage and considerations for designing very high SPL systems like this until they eventually told me to go here instead because i'm not a sound reinforcement professional ... but before i was shown the door one good dude did give me a list of AES papers on hearing damage to read ( he wrote most of them ) so i'm currently reading that ...
the system in question is definitely deep into hearing damage territory but i'm trying to quantify it more accurately so i can optimize for lowest audible distortion at highest allowable levels and do so with the most compact and affordable system ... also read recently a very nifty article about audibility of distortion at different frequencies which allowed me to reduce the size of subwoofer array since as it turns out we are essentially deaf to distortion at subwoofer frequencies ...
don't be shy - i want to hear your criticism ! i will eventually hear it one way or the other and i would much rather hear it BEFORE i start building !
to conclude this post i will throw in some sample material i enjoy so that you can have an idea of what i will be trying to reproduce with this system:
i occasionally listen to metal as well because i was a metalhead as a teen but for the most part these days i listen to electronic music with lots of deep bass.
this design has gone through a bunch of iterations and some aspects of it keep sort of vacillating which i will address later ...
for now i will just describe what the current iteration is, as depicted in the sketches ...
5-way system consisting of a center subwoofer stack that goes between the two main 4-way speakers ...
both speakers and subwoofer stack are 80" tall ( to fit in basements ) and the width is 32" for speakers because that's 1/3 of 96" that plywood comes in ... and 2 x 32" for the subwoofer stack. each subwoofer module is 25.5 x 32 x 40 inches, with a total of four modules. 25.5" is just 24" + 2 thicknesses of plywood. the main speakers would have approximately similar depth ( about 2 feet ) but since they curve towards the listener they don't have any hard number for their depth ...
in my sketches you can also see a 75" TV mounted on the subwoofer stack ... that is indeed where the TV will go but it doesn't have to be 75" ... although the width of 75" matches the width of the subwoofer stack exactly, which may or may not be a coincidence ...
total system width would be about 12 feet which is about as narrow as i could get it but it should allow it to fit in most living rooms outside of NYC and SF.
in the current iteration of the design the driver compliment is as follows
25 hz - 80 hz:
https://www.eighteensound.it/en/products/lf-driver/21-0/4/21ntlw5000-4350 liter net box, vented, tuned to 25 hz ...
four such boxes in the center behind the screen ...
80 hz - 200 hz:
https://www.parts-express.com/JBL-2262HPL-338312-004X-12-Neo-Woofer-294-45040 liter net box, vented, tuned to 70 hz
four 8 ohm drivers per side for a total 2 ohm load
200 hz - 800 hz:
https://www.eighteensound.it/en/products/lf-driver/8-0/8/8NTLW20005 liter net box, vented, tuned to 130 hz
four 8 ohm drivers per side for a total of 2 ohm load
800 hz - 3 khz:
https://www.bmsspeakers.com/index.php-62.html?id=bms_4594nd-midon the following horn:
https://www.bcspeakers.com/en/products/horn/1-4/0/ME4648 ohms ...
3 khz - 20 khz
https://www.bmsspeakers.com/index.php-48.html?id=4526nd2 x 2 array of 8 ohm drivers in series parallel for a 8 ohm load
the horn array will consist of four identical but not symmetrical horns that must be 3D printed
that is to say the arrangement of the four horns will be symmetrical but the horns themselves will be asymmetrical
the asymmetry will be used to bring the drivers as close to each other as possible, with magnets almost touching each other
all in all 8 identical horns will have to be 3D printed from a single file
main crossover will be MiniDSP 4X10 HD and it will plug in directly into the TV via optical Toslink ... i will throw some protection caps on the compression drivers to be safe as well ... the 80 hz crossover point will be realized internally to the 4-channel DSP amplifier that will power the woofers and subs
the TV will ultimately be the source, it will probably be a 2022 Mini-LED TV from either Samsung or Sony. i had a 2021 Samsung Mini-Led that i used as a source like this via optical output and it worked really well as the TV could do YouTube, Netflix and Spotify ... unfortunately it was a first generation Mini-LED and had a flicker so i sent it back to Costco ... in 2022 Samsung is coming out with 2nd generation Mini-LED and Sony is entering the Mini-LED market as well ... it will take time for prices on new models to drop but it will also take time for me to design and build this system ...
there will be a total of 3 amplifiers:
QSC PLX 2402 ( which i already own ) will power the 8" midbass array at 2 ohm, and will have a Noctua fan mod
QSC CX 404 ( four channel ) is $400 used on Ebay and will power all compression drivers, it will have the same Noctua fan mod
JBL Crown DSI 2.0 MA4 ( four channel ) will power both subwoofers and 12" woofers. this amp is expensive but it packs a lot of power and DSP and i may need those.
now for where the design keeps vacillating ...
i sometimes wonder if it's best to use 12" or 15" JBL woofers and whether it is best to use 8" or 6" Eighteen Sound tetracoil midbass ...
i sometimes wonder if the midbass should be vented or sealed ...
i sometimes wonder whether to use BMS 4592 mid, 4594 mid or 4599 mid ... the 1.4" throat versions can be used with B&C ME464 horn while the 2" versions would need either SEOS 30 horn or a DIY horn with 3D printed throat adapter and plywood mouth
i sometimes wonder if the system should be 5-way or 6-way ... in case of 6-way it would have an array of Six Beyma CP-12 Supertweeters crossed at 8 khz ...
i sometimes wonder if for tweeters i should go with BMS 4526 or Faital HF106 ... and in case of faital whether to use 1, 2, or 4 ... or if using Supertweeters perhaps a single Faital HF1440 ...
the reason i am using an array of tweeters like this is 2 fold:
- it lower distortion by quadrupling throat area thus lowering compression and distortion ...
- it doubles the coverage angle at the extreme high frequencies where horns begin to beam ...
yes i realize there is the issue of comb filtering and that's why i keep vacillating on what is the best solution in that area ...
as for what i'm trying to achieve with this is it is intended to be my main system for the foreseeable future - it isn't intended as any sort of exercise - it is intended to be THE system i use whenever i want to listen to music or watch music videos or movies at maximum volume ... it is intended to have the absolutely maximum performance level possible within what can be achieved in a system of reasonable size and cost ... spouse approval factor is not a consideration but i want to challenge myself by keeping it as compact and cost effective as possible - i could not respect myself if i went sloppy and simply started to mindlessly pile on heaps of subwoofers using "the more the merrier" approach ...
this system will be acoustically in a large space as the house is an open concept plan with one continuous space spanning from the porch to the patio and also both first and 2nd floors ... it's basically in a "living room" but it's also the same space as an eat in kitchen and sunroom and pretty much open to the entire rest of the house except bedrooms and bathrooms. the floor is wooden as there is basement below. the ceiling is sloped from 10 to 20 foot high. all other surfaces are fairly thin drywall and lots of windows - the house is mostly acoustically transparent - i don't expect much standing wave issues as i used to have in an apartment in a concrete building in NYC. on the other hand where i could fill that apartment with a single sealed 18" TC Sounds LMS subwoofer because those cinder block walls and concrete floors and ceilings trapped the bass and kept it resonating in the space ... by contrast in the space i'm currently in the bass will simply spread through the house and then leak out to the street so i need much higher bass output even for my listening position that will be about 8 feet away from TV screen ...
i have also been on a professional sound reinforcement forum discussing hearing damage and considerations for designing very high SPL systems like this until they eventually told me to go here instead because i'm not a sound reinforcement professional ... but before i was shown the door one good dude did give me a list of AES papers on hearing damage to read ( he wrote most of them ) so i'm currently reading that ...
the system in question is definitely deep into hearing damage territory but i'm trying to quantify it more accurately so i can optimize for lowest audible distortion at highest allowable levels and do so with the most compact and affordable system ... also read recently a very nifty article about audibility of distortion at different frequencies which allowed me to reduce the size of subwoofer array since as it turns out we are essentially deaf to distortion at subwoofer frequencies ...
don't be shy - i want to hear your criticism ! i will eventually hear it one way or the other and i would much rather hear it BEFORE i start building !
to conclude this post i will throw in some sample material i enjoy so that you can have an idea of what i will be trying to reproduce with this system:
i occasionally listen to metal as well because i was a metalhead as a teen but for the most part these days i listen to electronic music with lots of deep bass.
Attachments
Last edited:
The system you have envisioned is very complex. Most of the large studio control rooms and film sound stages do not have main monitors with as much complexity as you are envisioning. Designing, building, and measuring a system of this complexity, and getting a good result, is very difficult.
Are you planning a 5-way speaker because you do not believe that a 4-way will get the job done, or have you just decided you want a 5-way because you really want a 5-way? it is fine either way, but it helps us to know what your thoughts are.
Are you planning a 5-way speaker because you do not believe that a 4-way will get the job done, or have you just decided you want a 5-way because you really want a 5-way? it is fine either way, but it helps us to know what your thoughts are.
first of all thank you for trying to help me keep my sanity by not forcing me to think about this on my own for months on end
that said ... i honestly do not understand the argument that more frequency bands equals more complexity
i used to have a 5-way DIY system in the car and it worked great except the car died before i had a chance to install the subwoofers
the Harman Kardon system in my car right now is a 4-way but the subwoofers are 8" ... whereas in my design 8" is the midbass ... so actually the Harman has 3 frequency bands above the 8" while i only have 2 ... the Harman has 1" tweeters, 3" midranges, 5" bass drivers and 8" subwoofers - it sounds great ... incidentally my 5-way DIY car audio system also had 8" drivers as the 4th frequency band from the top ... i just never installed the 15" RCF subwoofers which i still have in the basement ...
besides, i didn't start out with some number like 5 or 6 - i started out by reverse engineering JBL Cinema Screen Array which is a 3-way ... but i soon realized two things ...
1 - even though it has 15" woofers the JBL Cinema Screen Array still is designed to work in conjunction with 18" subwoofers ...
2 - it's a relatively budget speaker, and in the more premium speakers JBL has moved to coaxial HF drivers ...
so working from there i kept the four 8" midranges that JBL Cinema Screen Array has but decided to spread the 15" and 18" drivers apart in size by turning 15" into 12" and 18" into 21" ...
i then replaced the 4" HF driver with a coaxial BMS or B&C driver ...
then i split the coaxial into separately Midrange driver and Tweeter ...
then i replaced a 2" tweeter with four 1" tweeters for lower distortion and better dispersion as well as better HF extension ... this idea was taken from JBL Marquis dance club series which use dual supertweeters crossed to 3" compression driver at 10 khz ... that actually turned the system into 6 way but i then was able to consolidate the tweeters and supertweeters into a single array cutting it back down to 5-way ...
in other words the JBL Marquis is already a 5-way ( fully DSP active ) system even though it doesn't use a coaxial HF ... if the compresison in JBL Marquis system is replaced with Coaxial that turns it into a 6-way ... so i was actually able to streamline that design by turning it into "only" a 5-way ...
i also rearranged the woofers and midranges into a configuration similar to what is used in JBL VTX arrays for greater compactness of overall design ( the Cinema Screen Array is too tall for regular rooms )
i'm basically combining design elements of Cinema Screen Array, Marquis Dance Club Series and VTX Array ( all from JBL ) plus the latest advancement in Coaxial HF compression drivers from BMS ( JBL actually rebadges some BMS drivers as their own ) ...
im basically building a sort of a mutant JBL system ... even the Beyma bullet supertweeters ( not shown in this sketch, but considered in other iterations of the design ) are in a certain sense rip offs of older discontinued JBL supertweeters ... which JBL itself replaced with Selenium ST400 used in the Marquis Dance Club series ...
finally i could easily turn this system from a 5 way into 4 way simply by consolidating 12" and 21" drivers ... but that would actually DOUBLE THE SIZE of the system because i wouldn't be able to hide the subwoofers behind the TV ! those recording studios you mention have the luxury of placing the speakers wherever they want because they are designed from the ground up to have the space for those speakers while i have to fit mine into a normal room. the only way i was able to do that is by using a 80hz crossover to the subs and hiding the subs behind the TV.
it's also the same how the Cinema Screen Arrays have the luxury of being something like 9 feet tall so they can put 18" woofers vertically on top of each other while i would run out of ceiling height in a basement if i did that so i have to instead use 12" woofers symmetrically left and right, which doubles the woofer count but allows me to fit into my space ...
it's really about taking proven JBL designs and on one hand modernizing them ( JBL reserves latest tech for their VTX line ) and on the other hand reworking them until they fit into the space ...
that was basically the design process ...
also keep in mind all commercial systems are designed to make profit - which means they are all simpler than they need to be because if they were not they wouldn't be profitable and profitability trumps other considerations ... if the same guys who designed all these commercial systems FOR SOME CUSTOMER were instead designing them for THEIR OWN USE they would quite likely end up with a more complex design because it wouldn't be about the bottom line ...
Last edited:
You have to be happy with the system.
I would just not be interested in 2 ohm loads.
With multiple drivers I think thermal capability is more than enough.
Not necessary, likely to vent midrange.
It has been done before in live PA systems. The reflex/tuning point of the system is below the
crossover. So the smearing or poor effect a vented cabinet can have is not in the signal.
The cabinet is just simply vented for heat removal. Not needed for home system.
The ME464 is very nice horn, performance wise and visual as well.
2 ohm loads stress the amplifier and can cause higher distortion.
Many of us have designed, measured, simulated and built our own amplifiers.
I have no interest in 2 ohm loads. Not needed in home situation.
Controlling levels / and crossover for 5 way system is complicated.
I would likely look at measurement equipment and possibly DSP
or Digital Sound Processor to dial in and EQ your system.
If not, most of your concept is has realistic reasons.
But it could be overwhelming and little difficult.
I believe you could do a Vented 18" for bass
With crossover around 200/ 250 Hz which is were you will get a 6 dB drop.
So it will be easy to adjust bass levels.
Then use a sealed midrange.
Then the ME464.
I have used actual live sound towers in the house. 2x15 and Horn.
No where close to Hi Fi standards. Typical live PA crossover.
Just a simple stereo pair 2x15 and horn really rocked the house!! with maybe 20 to 40 watts.
With 18" and good midrange, and nice horn
Think you would be surprised what these live sound drivers can do.
I would just not be interested in 2 ohm loads.
With multiple drivers I think thermal capability is more than enough.
Not necessary, likely to vent midrange.
It has been done before in live PA systems. The reflex/tuning point of the system is below the
crossover. So the smearing or poor effect a vented cabinet can have is not in the signal.
The cabinet is just simply vented for heat removal. Not needed for home system.
The ME464 is very nice horn, performance wise and visual as well.
2 ohm loads stress the amplifier and can cause higher distortion.
Many of us have designed, measured, simulated and built our own amplifiers.
I have no interest in 2 ohm loads. Not needed in home situation.
Controlling levels / and crossover for 5 way system is complicated.
I would likely look at measurement equipment and possibly DSP
or Digital Sound Processor to dial in and EQ your system.
If not, most of your concept is has realistic reasons.
But it could be overwhelming and little difficult.
I believe you could do a Vented 18" for bass
With crossover around 200/ 250 Hz which is were you will get a 6 dB drop.
So it will be easy to adjust bass levels.
Then use a sealed midrange.
Then the ME464.
I have used actual live sound towers in the house. 2x15 and Horn.
No where close to Hi Fi standards. Typical live PA crossover.
Just a simple stereo pair 2x15 and horn really rocked the house!! with maybe 20 to 40 watts.
With 18" and good midrange, and nice horn
Think you would be surprised what these live sound drivers can do.
Can you boil down your (user) requirements to a short list? Perhaps:
You've come up with a lot of "ways" to achieve what you want, but what is it you really want?
- budget
- +/- x dB on-axis? +/- x dB at 10, 20, 30 degrees off-axis? -3dB point in the treble at what frequency and off-axis angle?+/- dB power response?
- What kind of radiation "shape" are you after? It seems you've already decided to use waveguides/horns, is there a reason for that?
- Peak SPL at listening position? Average SPL?
- Maximum/threshold for each order of harmonic distortion?
- Minimal/moderate/limitless use of room treatment?
You've come up with a lot of "ways" to achieve what you want, but what is it you really want?
well B&C actually created a demo system where they crossed the ME464 horn straight to a pair of 15" subwoofers in an MTM configuration ... it's on YouTube
so yes you most certainly CAN do a simpler system
but as i mentioned the system would be 100% DSP - no passive crossovers in the passband - only protection caps against turn on thumps and DC offset and such on compression drivers
i have been previously using a DSP crossover ( behringer ) for my subwoofer and with DSP it's a different process - it's not about measuring and calculating and entering it right once - it's basically you start using it right away and then just keep tweaking it by ear ... forever
you never need to get it "right" at all ... because whenever something starts to bother you - you just change it at no cost - and you can keep changing it as many times as you want - and if it doesn't bother you - then who cares if it is "wrong" ?
simple goals:
1 - at any given frequency between 25 hz and 16 khz - maximum SPL that will not instantly cause permanent hearing damage
2 - distortion level as close to inaudible as possible at frequencies and output levels mentioned above
3 - 90 degree horizontal coverage, enough vertical coverage to be able to both stand up and lie down on a couch, which works out to about 60 degrees or so
4 - no HF rolloff AT ALL off axis to 8 khz ... not more than about 3db @ 12 khz and 6db @ 16 khz
5 - on axis response can always be equalized completely flat, is not relevant in the age of DSP
6 - horns provide combination of output, bandwidth, coverage and directivity control other approaches can't dream about
7 - harmonic distortion is irrelevant, music will always have more harmonics itself than anything your system can produce, on other hand intermodulation distortion and subharmonic generation can be an issue ...
8 - no room treatment, the space is large, won't be an issue
9 - budget is as cheap as possible to meet my goals
Correct, agree
DSP would be very helpful
And allow more control to balance such a system.
ME464 seems like very nice Horn
Should be a interesting to play around with.
Not only performance wise, but horns just fun to look at.
Certain romance with horns, almost like the glow of tubes
DSP would be very helpful
And allow more control to balance such a system.
ME464 seems like very nice Horn
Should be a interesting to play around with.
Not only performance wise, but horns just fun to look at.
Certain romance with horns, almost like the glow of tubes
a pair of the ME464 horns with the DCX464 driver it was designed to mate with will play 450Hz-18KHz at volumes that will permanently damage your hearing within minutes.
at these frequencies, less is more. IMO you want a wideband constant directivity point source, especially if you aren't going to treat the room. put them in the front corners and toe them in.
and if your space is drywall and glass you will be hearing the sound of those things resonating and rattling as much as you hear the subwoofer. there's not much remedy for that.
at these frequencies, less is more. IMO you want a wideband constant directivity point source, especially if you aren't going to treat the room. put them in the front corners and toe them in.
and if your space is drywall and glass you will be hearing the sound of those things resonating and rattling as much as you hear the subwoofer. there's not much remedy for that.
It is not just the number of "ways". Your system will have a complicated radiation pattern, and it may be difficult to simulate. For example, you have an array of 4 midrange horns crossing over to a single horn tweeter. Can we simulate this so that we know in advance that it will work well? I don't know. There are horn experts on this site and they will have to answer that.
Another complication with so many drivers is just the individual driver QC. Let's assume that after a few months, one of the four 8NTLW2000 8" drivers has a defect/failure and its distortion goes up dramatically and its FR changes. You may hear that something has changed, but isolating down to the individual driver is complicated.
Your first two goals are ambitious
Poking around in google tells me that about 130 dB SPL is enough to cause instantaneous permanent hearing damage. So 1 dB less than that is 129 dB SPL. So you want your system to be capable of 129 dB SPL from 25 Hz to 16 kHz, with distortion being inaudible. Wow. Perhaps I am misinterpreting your requirement? There was a thread about a year ago from a guy who was building a recording studio, and if I recall, the max SPL requirement was something like 117 dB @ 1 M, per speaker.
In my mind, the way to think about meeting your requirement is to select the smallest woofer array that will meet your requirement. Maybe you do not need two 21" drivers per side, maybe you do, but you need to run those calculations. The next step is to decide how high those woofers can go. Can you run them up to 100 Hz? 120 Hz? Let's assume yes, 120 Hz. So then, what is the minimum mid-bass system that will meet your need... etc, etc, etc....up through the midrange and the tweeter. you may find that you don't need a 5-way. Perhaps a 3-way plus subwoofer (i.e. a 4 way) will meet your needs. If it were me, I would run the calculations, do the simulations, and let the data drive the decisions, or at least inform the decisions.
If you have done calculations, please share them, it will help us help you.
1 - Wow. Why? I can tell you firsthand, tinnitus is really not something you want to live with if you can help it.
2 - I think this is really venturing into unknown territory. No one, to my knowledge, has done any studies to quantitatively define how much masking of distortion there is at the extreme SPLs you're talking about (or, really, at any SPL for that matter).
4 - You are aware that even the best waveguides will naturally have lower output off-axis within the "loaded" passband, even within their design angle?
5 - Yes, but not necessarily while flattening off-axis response or power response.
6 - Good luck achieving #4 with horns.
7 - That's an interesting opinion, one that I don't agree with.
8 - How large is this room? Will your first reflections be delayed beyond 10-15 ms from direct incidence?
9 - Something doesn't jive here. Good luck with your design.
If you are willing to take a step back look at all your options, try reading this thread
https://www.diyaudio.com/community/threads/asathor-a-jbl-4367-clone.367215/
Imagine this speaker, but with an active DSP crossover (which you are already planning to use). Imagine it with a pair of stereo subs based on that big 21" Eighteen sound driver, crossed at ~ 100 Hz. The Asathor is a well thought-out system, capable of high output and low distortion. It is worth consideration. With a pair of massive subs, it might meet your requirements. And even if not, it may broaden your options and give you some ideas.
j.
https://www.diyaudio.com/community/threads/asathor-a-jbl-4367-clone.367215/
Imagine this speaker, but with an active DSP crossover (which you are already planning to use). Imagine it with a pair of stereo subs based on that big 21" Eighteen sound driver, crossed at ~ 100 Hz. The Asathor is a well thought-out system, capable of high output and low distortion. It is worth consideration. With a pair of massive subs, it might meet your requirements. And even if not, it may broaden your options and give you some ideas.
j.
interesting comparison ...
horns have evolved in that they originally were used to increase efficiency when power was not readily available ... today they serve different purposes
today it's really more of waveguides than horns in that amplifiers no longer need much help with gain but higher performance of modern systems demands greater control over beam pattern ...
some of the high distortion associated with horns was actually due to slow flare rate of horns that was used to give them more low end extension ... we see this in vintage horns that were very long ...
modern horns are short and almost immediately flare out to a constant directivity waveguide ... they no longer chase low-end extension but instead focus on lower distortion and providing consistent coverage ...
so basically horns are out - waveguides are in ...
my design leverages this as it uses a deeper midrange horn and a more rapidly expanding HF horn ...
the deep midrange horn is optimized for gain at lower frequencies while the HF horn is optimized for wide coverage and low distortion ...
compression driver distortion arises at high frequencies in slow expanding horns ... but my slow expanding horn will only be used up to 3 khz before distortion gets bad ...
and above 3 khz i will use rapidly expanding waveguides that do not over-compress the air thus avoiding high distortion ...
well, in theory ...
Last edited:
excellent question.
firstly i'm attaching two new images ...
my original sketch didn't show the fact that the four vertically stacked 8" midbass drivers are actually on an arc and not in a straight line ... so i'm uploading a "zoomed in" sketch that shows that detail ...
it will actually be two identical cabinets - top and bottom - with the top flipped upside down. together they will horn a sort of a waveguide for the arc array of 8" drivers. this will hopefully launch a cylindrical wavefront giving me the vertical coverage i need.
as for why there's 4 of them in the first place it all started here:
https://jblpro.com/en-US/products/5742
this system also uses four 8" drivers ... but slightly differently. in their case the top two and bottom two are driven by separate amps with the two top running to 1.3 khz while the bottom two only run to 550 hz. this allows them to dial in vertical directivity pattern. in my case i have actually simplified this radically in that i combined them all into a single four unit group that is arc shaped and runs to 800 hz. they couldn't form an arc because their waveguide is shared with other cinema speakers which are only 18" deep so they have to keep it flat - and they couldn't cross their compression driver at 800 hz because their driver is a titanium dome with titanium suspension which has to sit almost flush against the phase plug in order to reach 20 khz while mine is a polyester driver that is a dedicated midrange and rated to go down to 300 hz ... i can also use a deeper horn for this midrange driver since my speaker isn't limited by 18" installation depth as most of their range is ( and they don't make a separate waveguide for their only 24" deep model ) my deeper horn combined with a dedicated midrange allows me to cross low enough that i can use a group of four 8" drivers all running off the same amp with the same EQ.
but still, you say, why do i need four ? if i'm not trying to control vertical directivity like JBL is then what is the purpose ? well i was wondering as well because i looked at this speaker for example ( the link auto-downloads PDF brochure to your hard drive ):
https://jblpro.com/en/site_elements/marquis-dance-club-series-brochure
and they make do with a lower midrange count ... specifically they have two 8" midranges in a system with four 18" subwoofers, whereas i have four 8" midranges with two 21" subwoofers ...
one possible explanation is that a dance club system ( Marquis ) at the time ( it's discontinued ) was supposed to emphasize bass and treble ( boom-tss kind of sound ) while Cinema system had too reproduce a more midrange oriented sound ( tires screeching etc ) ...
but then we have this system:
https://jblpro.com/en-US/products/vtx-v25-ii
which i consider possibly the most advanced loudspeaker ever made. and you can see it runs four 8" drivers per two 15" drivers. now one obvious reason for this is simply packaging - there is no way to fit any other number of drivers in there symmetrically and with no wasted space. the other reason is that in a large array like this the 15" drivers will couple to each other realizing horn-like bass boost and this may help them hang in there with all those midranges, on the other hand the midranges may get slightly attenuated over long distances by dissipation into air though it mostly affects the tweeters, which is why they have so many HF units ... each with dual 3" voice coil diaphragms, so a total of 18" of 1.5 feet of tweeter voice coil per array module / cabinet ... but that's besides the point and has nothing to do with us - we don't have to throw 20 khz over half a mile where 90% of it will dissipate into air.
finally the last reason they may have so many midranges is one that does apply to us ... it's something i read just a few days ago, here it is:
https://www.axiomaudio.com/blog/dis...ortion at levels,the masking effect of music.
basically in the upper midrange we can hear about 1% distortion but in sub-bass it takes almost 100% distortion to be bothersome ... so you can drive the subs all the way to their absolute limit and nobody is going to care, but even slightly overdriving midranges may become audible. this is why in my system the sub-bass seems to be under-sized BECAUSE IT REALLY IS. and mid-range seems over-sized ALSO BECAUSE IT REALLY IS. but it's also that way in both JBL Cinema Screen array and VTX Array and my speaker will also do Cinema duty as well.
which brings us to my other uploaded graphic - the chart !
bottom reddish line is average music level as measured by a dude on another forum from his own collection of music by running headphone output to REW ... i have scaled his curve up to 110db @ 1 khz because that's the NIOSH 1 minute exposure level ... that is a level ( measured with "slow" setting on SPL meter which averages peaks over a second ) at which you can be exposed for at most 1 minute a day to keep hearing damage to "acceptable" levels. in other words this is as loud as you would ever want to listen for even just one minute. Actual rock concerts are regulated at 100db which is 10 db lower ... but of course they can last for hours and 110db is ONE MINUTE rating. I rarely listen to music for hours and sometimes i will crank it for just about a minute on a good part of a song so i went with the 1 minute rating.
second ( light blue ) line from the bottom is my estimated PEAK levels based on the above mentioned average levels. as best i could find out music has about 6db crest factor in bass and about 15db in high frequencies so i added that much to the average curve to get my peak curve.
the little cyan curve on the left is 6db of bass boost that i will almost certainly apply to spice things up.
the top five short curves are each individual frequency band of the 5-way system with output calculated for lower 3 bands using HornResp and for the upper 2 bands using some back of the napkin calculations based on published specs ...
as you can see the loudest frequency band is actually the compression midrange, which is the only frequency band covered by a single transducer. it is also the most critical frequency band occupying the center of the vocal range.
the most under-powered frequency band is the subwoofer, which is done to keep system size and cost down and also as previously explained because we tolerate very high distortion in subwoofers.
the tweeter runs out of steam a little above 8 khz or so but keep in mind that curve is the POWER RESPONSE not the on-axis response ! my system will have more output at 20 khz averaged over the 90 x 60 degree radiation angle than ANY single driver you can get and mount to any horn. and that's of course because i simply have four drivers there instead of one. and the reason i did it that way is because a single big driver will beam above 10 khz ... but a single small driver will get fried ... so i used four small drivers
Attachments
Last edited:
what do you mean by "at these frequencies, less is more" ?
the reason i'm using drivers in the same SPL range as the DCX464 is because these types of drivers when driven to high levels will produce close to 50% distortion in the most critical frequency range where distortion is most audible ...
instead the system is DESIGNED such that in normal use i will not need to drive the compression midrange to more than 1% of power, which will keep distortion acceptably low while still hitting well over 100 decibels ...
as for rattling ... that is life LOL
yes i have. i already posted this above but i'll repost it for you here ( see attachment ). i have explained in one of the posts above what this chart means.
as for precise levels at which hearing damage occurs it's complicated and dependent on frequency.
here is what i was told a few days ago by Michael Lawrence on another site:
Looks like things have gotten a little bit off topic here, but here is an AES Technical Document that aggregates a lot of established research on sound exposure and hearing loss through the lens of live events:
https://www.aes.org/technical/documents/AESTD1007_1_20_05.pdf
For the physiological elements of hearing loss due to sound exposure also see:
https://www.who.int/pbd/deafness/Hearing_loss_due_to_recreational_exposure_to_loud_sounds.pdf
As to why almost all sound exposure metrics (OSHA, NIOSH, WHO) use A-Weighting, according to the NIOSH audiology folks, it has to do not with perceived loudness at a certain SPL, but simply that the A-Weighting curve is a reasonable approximation of the free field to eardrum reference point (DRP) weighting that allows a mic in free space to characterize the acoustic energy impinging upon the eardrum, in other words, the energy that tends to be the most damaging.
See the first Appendix in this WHO publication for information on the weighting and also how the exposure calculation is carried out:
https://www.who.int/publications/i/item/safe-listening-devices-and-systems-a-who-itu-standard
That is not to say that low frequencies (or short term impulses) cannot be damaging to hearing, they both can, which is why an increasing number of sound exposure policies specify a secondary limit in Peak C (typically around 135 or 140 dB).
Some additional measures, such as monitoring the C-A of program material for excess LF exposure hazard, are proposed in the AES paper I linked above. It bears mentioning that up until recently, much of the data on this comes from audiological / industrial / aerospace sectors and it is a newer field of study looking at things in the live events context.
The WHO is releasing a global standard for Safe Listening Venues to address this in early March, informed in part by some of the research collected in this series of AES reports:
http://www.aes.org/e-lib/browse.cfm?elib=21529
https://www.aes.org/e-lib/browse.cfm?elib=21551
https://www.aes.org/e-lib/browse.cfm?elib=21552
Additionally, we're hoping to drum up interest and dialogue surrounding an industry-wide certification scheme to help people obtain actionable knowledge about these issues, see this paper if interested:
https://www.aes.org/e-lib/browse.cfm?elib=21227
As part of our inclusion of OSHA and NIOSH dosimeters directly within Smaart, I've had a fairly extensive communication with the NIOSH audiological folks on this topic. More news on that front soon but I can't say quite yet
i'm currently reading the first paper on that list ... it is excellent.
do not be quick to assume that i can't meet my requirements of inaudible distortion at hearing damage levels. hearing damage occurs sooner than people think. and distortion is not as audible as people think. i believe my system already meets the requirement.
Attachments
Last edited:
Hello, This sounds like a solid sound system, capable of delivering many loud service years. Know too little about PA driver and horns, even though i find them fascinating, luckily this is already very well covered above anyway, but i can comment from experience on some of the other parts of the system:
- If you can get QSC CX 404 for that price, that is a very solid choice I bought a QSC DCA 1644 back in 2001, and died just a few months ago, being on more about 20 years or I also have the DCA 2422Hmm.. criticism
How does it go with the floor?
Does its configuration suit your purpose?axiperiodic said:
if you want a 15" 2-way why not just buy this box:
https://www.rcf.it/c/document_libra...93f-8e10-45cf-8714-cc2979a7ca6f&groupId=20195
it already does everything your Asathor does and its fully self contained and portable with amplification built in
breathing saw dust isn't my idea of fun. i am not interested in building a clone of something i can just buy.
frankly it wouldn't be a half bad idea to buy that RCF box and just build those 21" subs for it.
i think my reasoning for not doing that was that the RCF speaker has no digital inputs whereas both MiniDSP HD 4X10 and DBX DriveRack Venu 360 do have digital input ... which makes them a more "high end" solution while the RCF is a more "PA" solution ...
i also felt like the price maybe isn't justified ...
i mean yes you CAN build a 3-way system that works ... i just don't understand this obsession with simplicity.
you need to understand the reason companies like JBL and RCF stick with these simple designs is SO THEY CAN MAKE MORE MONEY.
if they can build the speaker for half the price and still sell it at the same price then they HAVE TO DO IT or they will be driven out of business by competition.
in DIY we don't face that pressure. we DON'T HAVE TO build the smallest, cheapest, simplest box possible. it doesn't matter !
Last edited: