yeah i had QSX PLX for 15 years only thing that happened is the volume knobs are a little rusty now i have to twist them back and forth a bit to get good contact ...
the problem is the CX i buy on eBay for $400 may already be 20 years old and die the next day LOL but it's worth the risk probably ...
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 very similar a QSC DCA 1644 back in 2001, for 3000 Euro, and it died just a few months ago, being on for more about 20 years or so, it measured exactly the same in all years, 4x240 W of continuous average power into a 8 ohm loads with low distortion. I also have the DCA 2422, and can you afford it.. it has at least 3x the power but in only two channels.
The mini DSP 10 channel output is great dac/product, it has choice on individual balanced or regular output, high signal level out, regret not buying something like that, straight away, but went through a number of HQ regular soundcards with 8 dacs onboard first. And you need a p.eq, time alignment and crossovers so yes you need a DSP, but you do not need another one in your amplifier? Do your signal shaping one place. Have you considered feeding the dac a signal from a computer instead of the tv? SPDIF with toslink works fine and has lossless sound quality for 2 channels, but Dolby and DTS will need compression to save data. This is fixed in Adat protocol and If I remember correctly the mini dsp is modules, so you might need separate processing units for crossover or Dirak or what ever.. The computer has so much more power, splitting the signal up in 10 channels, crossovers, lots of EQ , limiter , time alignment, convolution, and CPU usage is about 5% max on maybe 2-3 cores.
Having as few passive crossovers as possible will make it much easer to tune all these speakers. Aim for low level crossovers (active) as much as possible. Preferable before the DAC, and separate eq for each speaker location or even driver.
Criticism.. hmm.. you could probably save a small rainforrest by choosing 18" driver with half the VAS, I mean 4 x 350 liter.. thats a appartment in some part of the world🙂 Look and high end car stereo subs, they have same high power rating (almost) but are optimized to fit inside the trunk of a car, traded against lower efficiency due to heavier cone and voicecoil to handle the power.
Have fun and good luck... I just am finishing up something similar but for another purpose (music all over) and less PA elements, but hifi drivers will burn so I need to change. It will take longer than planned🙂But it pays off..
brg Simen
ps! dont listen to (all) the mumbo-jumbo🙂 about your ears and loss of hearing. This subject is mostly opinions and little research. Examples: yes 130 dB source between 1500 to 3000 Hz , close range (less than 2 m) are not good for your ears long term sensitivity, but 130 dB at 30 Hz is just fun, no harm, our ear has at least 30 dB less sensitivity at 30 Hz vs 2500Hz. Another point often forgotten is that sound decays or decreases by -6dB everytime you double the distance from the source. So party on!!
- If you can get QSC CX 404 for that price, that is a very solid choice I bought a very similar a QSC DCA 1644 back in 2001, for 3000 Euro, and it died just a few months ago, being on for more about 20 years or so, it measured exactly the same in all years, 4x240 W of continuous average power into a 8 ohm loads with low distortion. I also have the DCA 2422, and can you afford it.. it has at least 3x the power but in only two channels.
The mini DSP 10 channel output is great dac/product, it has choice on individual balanced or regular output, high signal level out, regret not buying something like that, straight away, but went through a number of HQ regular soundcards with 8 dacs onboard first. And you need a p.eq, time alignment and crossovers so yes you need a DSP, but you do not need another one in your amplifier? Do your signal shaping one place. Have you considered feeding the dac a signal from a computer instead of the tv? SPDIF with toslink works fine and has lossless sound quality for 2 channels, but Dolby and DTS will need compression to save data. This is fixed in Adat protocol and If I remember correctly the mini dsp is modules, so you might need separate processing units for crossover or Dirak or what ever.. The computer has so much more power, splitting the signal up in 10 channels, crossovers, lots of EQ , limiter , time alignment, convolution, and CPU usage is about 5% max on maybe 2-3 cores.
Having as few passive crossovers as possible will make it much easer to tune all these speakers. Aim for low level crossovers (active) as much as possible. Preferable before the DAC, and separate eq for each speaker location or even driver.
Criticism.. hmm.. you could probably save a small rainforrest by choosing 18" driver with half the VAS, I mean 4 x 350 liter.. thats a appartment in some part of the world🙂 Look and high end car stereo subs, they have same high power rating (almost) but are optimized to fit inside the trunk of a car, traded against lower efficiency due to heavier cone and voicecoil to handle the power.
Have fun and good luck... I just am finishing up something similar but for another purpose (music all over) and less PA elements, but hifi drivers will burn so I need to change. It will take longer than planned🙂But it pays off..
brg Simen
ps! dont listen to (all) the mumbo-jumbo🙂 about your ears and loss of hearing. This subject is mostly opinions and little research. Examples: yes 130 dB source between 1500 to 3000 Hz , close range (less than 2 m) are not good for your ears long term sensitivity, but 130 dB at 30 Hz is just fun, no harm, our ear has at least 30 dB less sensitivity at 30 Hz vs 2500Hz. Another point often forgotten is that sound decays or decreases by -6dB everytime you double the distance from the source. So party on!!
so think of the floor as an acoustical mirror ... the arc goes into that mirror at 90 degree angle and appears to continue in the reflection ... on it's top side it has to curve back just enough to cover my ears when i'm standing up at a listening distance ... that angle will have to be calculated ...
a complete VTX system costs more than a house. as such i haven't even asked myself the question whether it would fit my needs or not.
the big box for subwoofer isn't because it can't use a smaller box. it can. it's because larger box has more output from the bass reflex port, resulting in louder, deeper bass. with a sealed enclosure there is no benefit in going bigger than manufacturer recommended box because you won't get any extra output. with a vented box you can actually get more output from same driver by going bigger because the port will be moving more air. and of course you can get even more output with a folded horn subwoofer, but it will again be much larger. so it's just about deciding what size of box you're comfortable with. in my case as long as the subwoofer can fit through the door and down the stairs and into the basement it's not too big.
Your drawing does not do this.
Then maybe it is wrong and you are aspiring to something unnecessary?
my drawings could use some work ...
as for the VTX i'm only copying some aspects of it just as i'm only copying some aspects of Screen Array and Marquis
and it isn't so much copying as just borrowing ideas ...
actually they made a service bulletin for those faulty pots, so if your amp has serial number so and so. I received mine for free, but that many years ago.
If you want another example of good design, look at Lab-Gruppen, my Lab-1000 has been running since I got it in 2001, before that it had 10 years of hard labour in NRK (our BBC). Use it everyday still, and I have measured 88 volt unclipped output in 8 ohm bridge, decent for a 2x350W
Attachments
In my opinion, crossing over three times within the critical lower midrange to lower HF region, and using ported midrange enclosures with all the energy storage that necessarily results will do nothing whatsoever to help reproduction accuracy. As the sub enclosure becomes larger, so does the stored energy, and the bass becomes ever more sloppy and tuneless. In the HF, arraying drivers is impossible due to the required spacing at these wavelengths unless phase coherent, and trying it with horns will produce considerable comb filtering/lobing. There is certainly a case to be made where more is not more if sound quality is a key factor. If you just want a loud system, simply build it as you see fit and I am certain that you will enjoy it, but heed the warnings regarding hearing loss - there is good reason for industrial noise level exposure limits...
it is possible that you are right
but i don't want to live in fear 🙂
LOL
anyway i think i'm going to sign off for the day ...
talk to you guys later.
Trust me, you would rather want to live in fear of tinnitus than with actual tinnitus. It is so much more than just hearing loss. There is a reason there is a correlation between suicide and tinnitus. It's just really really unpleasant, and it's forever.
I would listen to the advice you are given in this thread, people are just trying to coach you onto a track that leads to a system you more likely will be happy with. Designing for 110db average volume levels just does not make sense. Those large commercial arrays are designed for very large rooms, and would sound terrible in a home environment.
I would listen to the advice you are given in this thread, people are just trying to coach you onto a track that leads to a system you more likely will be happy with. Designing for 110db average volume levels just does not make sense. Those large commercial arrays are designed for very large rooms, and would sound terrible in a home environment.
the ideal situation is to have just one acoustic source per passband, each spaced at 1/4 wavelength of the crossover frequency to the next and matched in directivity/polar response. not an easy thing to do, and many times impossible. but the more you compromise this for high SPL/low distortion - the more lobing, comb filtering and boundary reflection interference "hash" you will create in the room. and the sound you hear in your room is 50% or more reflected energy. multiplying acoustic sources just to get a low calculated distortion number without regard for radiation patterns in the room is a recipe for disaster.
erratic polar response, comb filtering, rattling windows and structural energy storage - this is "distortion" too. the system you are contemplating is the equivalent of randomly welding six Ferraris worth of parts together and driving it offroad.
Last edited:
I am trying to understand for which size of a room the system is meant to be.
I think hundreds of square meters...
Or thousands of square feet...
I think hundreds of square meters...
Or thousands of square feet...
Strongly agree with this post. One acoustic source per passband, and keep adjoining acoustic centers within 1/4 WL if at all possible.
Steep xovers really help achieve this, but you'd have to get linear phase ones.
A few equipment specific comments:
I'd recommend staying with either the BMS or B&C coaxs that run all the way up to 20kHz. Should be better than splitting it as proposed.
Ime, one of those CDs on a horn like the xt1464 will sound better than using the ME464. (ive played with both a fair amount.)
The PLX amps do not really like 2 ohms. If you don't have a completely stiff 120VAC line, they can momentarily dropout when running hard.
A few room vs current proposal comments:
From the sounds of the room environment/construction, the room is going to talk back much more than you may anticipate. Due to subs of course.
It will be the weakest link in high SPL clean clear sound.
I've pursused high SPL, high dynamics, with vanishing acoustic distortion, for a half-dozen years now.
My current 5-way setup uses these amps: subs PL380, lows PL340, mids bridged CX404, highs bridged CX-168, VHFs bridged CX-168.
I figure from your post you'll know what that infers. 🙂
Last thing...a system like i've put together.....like you're proposing can sound so clean, it's often easy to be listening to far louder levels than realized.
Danger Will Robinson!
The 215-DCX mentioned above must be awesome.
I personally would go for 3/4 way unity/synergy horn and mono subwoofers. That is my ultimate plan anyway🙂 1.4" driver for the top end (I will get some Beyma CP755Nd drivers soon), then 4 x 4" BC, 2 x 12" BC, all neodym magnets for weight. Yes, Mark's Syn9 was a great inspiration for that. And some large bad *** mono subwoofer below 100 Hz. I am not an expert at all, but your suggestion seems to introduce issues that will make the sound worse.
On the other hand, I was really surprised how well multicell horns work - but the cells must be small enough. The 4 horn array does not seem like a good idea to me - if not using a coaxial or good 1.4" driver, I would use just one great 1" driver in a proper waveguide/horn.
I personally would go for 3/4 way unity/synergy horn and mono subwoofers. That is my ultimate plan anyway🙂 1.4" driver for the top end (I will get some Beyma CP755Nd drivers soon), then 4 x 4" BC, 2 x 12" BC, all neodym magnets for weight. Yes, Mark's Syn9 was a great inspiration for that. And some large bad *** mono subwoofer below 100 Hz. I am not an expert at all, but your suggestion seems to introduce issues that will make the sound worse.
On the other hand, I was really surprised how well multicell horns work - but the cells must be small enough. The 4 horn array does not seem like a good idea to me - if not using a coaxial or good 1.4" driver, I would use just one great 1" driver in a proper waveguide/horn.
I think you try to overcomplicate things, that DCX464 in the ME464 horn is very clean and low distortion to about 400Hz, and even lower is still ok to most standards. Combine that with a good woofer (Faital 12PR300) and subwoofer (FaitalPRO 18XL1800) and you got all you want i think. crossovers at 500 and 150Hz higher order with a dsp is what i would do in your case. Even at extreme volume this will sound clean if you design it right. 5 Way is overkill and overcomplicating things here, so more room for problems and more costs that are not needed.
All right bro let's see it !
Post the pics !
I don't care if it's just a prototype that's still held together by clamps or something - i want to see the speakers, the amps the DSP ...
what is your signal chain like and how do you control the volume ?
as i mentioned i want to use TV as a source because i want the system to be user friendly. it's not something i want for some basement den but for the living room - though i won't let anybody touch my speakers or electronics i want them to still be able to do things like choose content they want to listen to / watch and adjust the volume or mute. the problem is ( at least in case of Samsung ) the TV will not allow volume adjustment when the output is optical toslink ... and in fact new Samsungs don't have analog outputs at all ... and separately adjusting gain on a dozen different amplifier gain knobs is not an option ...
my hope was to be able to use a remote control with MiniDSP to adjust the volume ...
the other reason i don't want to use a PC as a source is that i noticed if i play the same video from the PC using TV as monitor and i play that same video on the TV directly ... it looks better when played directly on TV. it is NOT a subtle difference at all - it is day and night. basically what happens is the computer applies some processing to the video before sending it over HDMI and then TV applies its own processing and the two clash and cause artifacts ... on other hand when TV does all the processing from start to finish everything just looks right.
and besides it's just inefficient to have to run a PC when the TV already has all functionality i need already built in ... it's just not a clean design
bottom line ... the TV is the source but blocks the volume control on Toslink output ... i need to somehow control the volume, preferably using some kind of remote ...
i meant the fear of the system not coming out right not the fear of tinnitus.
system coming out right is a gamble. hearing damage is a certainty.
however yesterday i was thinking about it and actually i may NEED this system to save my hearing.
because without it i'm just listening in headphones which is WORSE.
ultimately i want to feel the bass in my chest but with headphones that is NEVER going to happen no matter how much i turn the volume up, so i'm probably going to lose hearing sooner WITHOUT the system than with it !
all right let's look at this in detail ...
i understand where 1/4 wavelength is coming from but this only applies to region of overlap between drivers ... in case of array that would be the entire passband ... in case of crossover a narrow region of frequency range the width of which depends on crossover order ... both MiniDSP and Crown DSI 2.0 offer up to 48 db / oct crossover slopes ... so any off-axis null would be at most 1/2 octave wide or so or in a 5-way system with 4 crossover points the combined width of all four nulls would be two octaves out of 10 octaves of audible spectrum and this is worst case scenario ...
narrow nulls are not particularly audible ... and 48 db / oct is not destiny ... i expect that in the future DSP technology will improve and steeper slopes will become available ... indeed with FIR functionality in the MiniDSP some people are saying up to 192 db / oct is possible but i don't know how to use FIR so i am not counting on this ... though i see no reason why 10 years down the line the DSP can't be swapped for something that DOES offer 192 db / oct slopes with no phase issues ... and if this never becomes available it means it isn't necessary because 48 db / oct already keeps the null narrow enough to not be bothersome ...
in most cases it is not realistic to try to keep two drivers getting crossed over close enough ... i have some JBL EON 1500 speakers with a foot between woofer and tweeter which are crossed at 2.7 khz ... and the slope is maybe at best 18db/oct and it works fine ...
now in ARRAY situation where the overlap is the entire passband now center to center spacing becomes a real issue ... i may yet reconsider that 4 X 8" array because it certainly doesn't meet the 1/4 wavelength criteria ... still, i think you're being alarmist because in practice no arrays meet this criteria - all those VTX arrays run those 8" drivers to at least 1 khz and if JBL would do this on their most advanced speaker it means it is acceptable compromise ... i think what is happening actually is that the more drivers you have in an array the more the response begins to even out ... so if you only had 2 drivers with a relative delay of more than 1/4 wavelength you would get strong comb filtering ... but with 4 drivers or more it would begin to even out ...
as for my high frequency array yes that is giving me a lot of pain to think about ... those drivers would be at least a wavelength away from each other at ALL frequencies in the passband ... however in that case we have the ability to control the dispersion pattern such that basically we split it using horizontal and vertical planes and each of the four drivers handles its quarter of radiation pattern ... of course it won't be a perfect split but in most cases some drivers will be stronger than others and thus the nulls shouldn't get too deep - but the entire passband will be covered with these irregularities ... which does bother me ...
most headphones have very uneven response on the top end and people find it acceptable ... my 2 X 2 HF array would have a similarly uneven response in the top end ... unfortunately i'm one of the few people who do NOT find the uneven response of most headphones acceptable ... with comb filtering starting at 3 khz that is a little too close for comfort to the vocal range ... i am still contemplating a 6-way solution that would push the supertweeter array out to 8 khz for that reason ... the main issue with that solution is simply packaging - it is hard to fit all these drivers on the system and have it look clean ... maybe impossible
the sketch i posted for the 5-way still would look pretty clean if built well ... while i also had a sketch for the 6-way which i will post now ( see attached ) that simply isn't as clean of a design ... in the 6-way setup there is additional crossover point at 8 khz with a single driver going up to 8 khz then switching to six Beyma CP12 Bullet Supertweeters each with 40 degree conical dispersion adding up to 60 X 90 overall coverage ... that older concept was also 40" wide but i later decided it didn't need to be wider than 32" ... even though wider is better in terms of being able to control directivity it would limit applications of the system as it simply would not fit in some rooms ...
Attachments
There are members here that can produce equipment that is better than is commercially available. Why limit yourself.