New Project with JBL's 4647 and 2445

[nestawasright]

That's a VERY important question, actually.

The more drivers there are in a system, the more difficult it is to get them to integrate with one another. The ideal system would have but one driver, and it would cover the entire audio spectrum. It doesn't exist in practice, of course, largely because of the laws of physics in sound reproduction.

Recognizing this, I believe everyone would agree that two-way is the next reasonable choice, and a worthy pursuit. Because of the same limitations of technology, however, it's not easily achieved. There are compromises that must be made in putting together a good sounding system with just two drivers.

Getting full high-frequency extension is one of the problems. The simple answer is to add a tweeter on top of a competent two-way, and some would argue, since the tweeter is "supplemental," it's still a two-way system. That's a stretch, I'd say. On the other hand, using a sub at the other end IS, in my mind, merely an enhancement of a two-way system.

To most accurately reproduce the entire audio spectrum, more drivers, four or even five, often, are required. There are many advocates of such multi-way systems here, and they dedicate themselves to overcoming the difficulties of having the multiple drivers play together accurately and cohesively. Many threads in this forum document that endeavor.

Your system is a competent two-way. I am encouraging you to acquire the proper crossover for it and to see if you like the way it sounds. You may have to do some reconditioning of the drivers, as well.

I have no experience with your horn and mid/high driver, but considering what the system was designed for, I assume it's worth keeping together. I suspect there are other members here who know the 2445 and 2380/A well. They can better advise you on achieving optimum performance with that particular combination.

I DO know the 2225/6H in that box, and you'll likely find you'll want to add a subwoofer to your system for best performance, or replace the woofer and retune the box as I'll be describing in the Q&D thread. 2226H will not provide extended bass, but that which it DOES provide is highly accurate and articulate. Read punchy and dynamic here.

Bottom line, know what you have first before proceeding. Get the proper crossover, and play it for a couple of months. Who CARES what it looks like for now? That is easily remedied, once you determine it is worthy of the effort. As you say, the MUSIC is the prime consideration here, and THAT is yet to be experienced....

Zilch's point are well explained and noted. (If he ever wishes to teach, please make a go at it. Myself I relish the thought of spending a few weeks in your comapny) Zilch's answer however, in my young mind on this issue, raise some questions I hope the abled forum member might answer for the benefit of us all.

Does it not suffice that a crossover would deal with the eminent problems the presence of a lot of drivers would have on a system? All the drivers, from my understanding, reproduce particular output in the spectral length, and in areas where this reproduction overlaps, a crossover is brought in to redress that problem.

I accept that one driver would not and cannot reproduce the entire audio spectrum. However, may I point forum members to this fellow. Those who spend time designing cabs would find it of particular interest.
http://www.danleysoundlabs.com/DTS20.html

I struggle with the logic that because for scientific reasons we cannot have one driver that would produce the entire audio spectrum, therefore, we must subsequently settle for the next figure to one, which is two. The problem it seems to me is accurate representation and since a lot of drivers cannot do that, why could it be held then that two would or should?

The need to add a subwoofer to this system, which I would, for I take much of what Zilch says with much respect and read with close attention, shows that except if the drivers of any system is designed to reproduce most, since all can't be the case, of the audio spectrum, then there would always be the need for "supplemental" cabs on the system. The point here is one, is not possible, two is often found to be inadequate. Thus, where does that lead us? Does anyone here has a two ways system that reproduces at least 90% of the audio spectrum? Would they care to share?

Zilch pointed out, rightlyfully, that two-ways systems could be found to be lacking, ususally he says around the HF areas. This I assume is generally, because most drivers are made to produce mid to sub woofer frequencies, thus when you use them to create two-ways, it's almost only natural that you would need a tweeter or horn to compensate for the HF spectrum. Such is not the case on this particular system, as is well noted by Zilch. Thus, I would be looking for a sub-woofer. Please forward suggestions.

On from theory to more practical issues about this system. If I may ask some more questions.

Would anyone with particular knoweldge of the horns discussed here PLEASE pitch in with what they know. In particular, advise on achieving optimum performance with this particular combination. Please.

Zilch mentioned that I may need to "reconditioned" my drivers. This is my first time going pro on a home system. That naturally means I have never heard of "reconditioning" drivers. Any and all help would be GREATLY appreciated. Or how to recondition the cab.

Truly, the music is yet to be enjoyed. And that raises yet another question. I know there are equipements used to measure sound production from speakers. Would someone please say how this is done, and say why one must do such measurements? Are they really accurate, given that they measure one thing and the music, the song itself, is made to sound in another way-not sure if that makes sense written down.

Zilch is right, I will keep my eyes on that Q&D thread and proceed to set this system up and listen to it. I am still stuck on finding proper equipments that would compliment these drivers.

(P.S Please forgive all erros in spellings. Currently baby-sitting twins. Distractions abound!)

[Zilch]

I'm workin' up a "tactful" response.

[duaneage]

I struggle with the logic that because for scientific reasons we cannot have one driver that would produce the entire audio spectrum, therefore, we must subsequently settle for the next figure to one, which is two. The problem it seems to me is accurate representation and since a lot of drivers cannot do that, why could it be held then that two would or should?

A 30 hz bass wave is about 60 feet long and a 15000 hz wave is about 1/2 inch. Frankly it is a miracle that 2 drivers can reproduce the range required let alone one. The real problem is directivity and at 1000 hx a 15 inch driver begins to "beam" it's output like a laser, instead of a wide pattern like a spotlight.

Dainty tweeter diaphrams are better than heavy cones at reproducing fast , high frequencies.

The need to add a subwoofer to this system, which I would, for I take much of what Zilch says with much respect and read with close attention, shows that except if the drivers of any system is designed to reproduce most, since all can't be the case, of the audio spectrum, then there would always be the need for "supplemental" cabs on the system. The point here is one, is not possible, two is often found to be inadequate. Thus, where does that lead us? Does anyone here has a two ways system that reproduces at least 90% of the audio spectrum? Would they care to share?

A smaller room reduces the need for driver area because the bass notes are reinforced by the walls. The more air in a room the more driver area for bass is required. As Mark Gander pointed out years ago, "You must move air to make noise"




Truly, the music is yet to be enjoyed. And that raises yet another question. I know there are equipements used to measure sound production from speakers. Would someone please say how this is done, and say why one must do such measurements? Are they really accurate, given that they measure one thing and the music, the song itself, is made to sound in another way-not sure if that makes sense written down.
Sound level meters are not too expensive and give an idea of what is going on. Other equipment like signal generators, voltmeters, impedance bridges, test discs, etc are necessary when your creating a speaker and do not yet know what the outcome will be.

having built a few dozen over the years I can tell you that some were sucessfull and some were not initially very good. If you follow Zilch's recommendations as well as others on this site you will probably be happy with the results.



(P.S Please forgive all erros in spellings. Currently baby-sitting twins.
Does this mean the budget is tight for massive audio projects?

Distractions abound!)[/QUOTE]

[Mr. Widget]

"I struggle with the logic that because for scientific reasons we cannot have one driver that would produce the entire audio spectrum, therefore, we must subsequently settle for the next figure to one..."


The primary issue is with quality. You can take an 8" driver like the JBL LE8T and get it to reproduce the full spectrum with a bit of EQ, but it won't have the dynamics or ultimate SPL of a larger more sophisticated system. It will also have greater harmonic and intermodulation distortion as the high frequencies are being physically tortured by the bass notes... to a lesser extent this is true of two-ways also.

You need to decide what are the important issues for you and pursue a design that satisfies that goal. Don't forget room size which may also affect your decision... it isn't as simple as should I build a two-way or a ? The components that you mentioned at the beginning of this thread are not the easiest components to work with to create a home system. If you add a sub and a tweeter they can be outstanding, or you can sell them, get a pair of 2235Hs, and some 1" drivers and horns and follow Zilch down his 4430 path.

Personally I would love to build a two-way along the lines of the 4430 or K2-S9500, but I can't stand the compromises... the highs sound hard and the lower mids are just not as detailed or articulate as they are with a well designed four-way. There are issues with large four-ways too though and some prefer the simpler sound of a good two-way...

Good luck!

Widget

[toddalin]

A 30 hz bass wave is about 60 feet long and a 15000 hz wave is about 1/2 inch.

Hate to be nitpicky, but please do the math before quoting misleading values.

speed = wavelength x frequency

Speed of sound at sea level is 1,116 ft/sec

1,116 ft/sec / 30 cycles/sec = 37 feet (not 60 feet)

1,116 ft/sec / 15,000 cycles/sec x 12 in/ ft = 0.9 inches (not 1/2 inch)

(Sorry, I write technical noise and air quality studies for a living.)

[spkrman57]

wavelength(inches) = frequency / 13,500

I use this for aligning voice coils in a multiple driver system when vertical alignment of the voice coils is not possible.

I come up with 1.1" for 15khz

Not saying mine is right, just what I have used and it works fairly close for me.

Ron

[Zilch]

http://www.drewdaniels.com/FREQ.pdf

[jbl]

Zilch,
I have the exact system as the 4647 except that I bought the components separatly. The cabinet is the 4507. The woofer, the 2225H. I also thought about the style of the Altec 19 with the horn box sitting on top. Instead, I still don't know where I got the guts to do it, cut out the baffle to accept the 2370 horn. I took the 4671 OK catalog, it's the 4671 with the the horn installed into the cabinet and used the picture to scale where the cutout should be. The tricky part was to clear the wood brace which goes from the baffle to the rear of the cabinet. Cut too low or too high and you'll have enough fire wood for the winter.
Walnut venner and a plastic "glass" top, mitered front edges and a very nice looking system.
Don't underestimate the low end response. Room gain and or loss will determine your opinion. For me, the right side of the room extends the usable low end to around 33 Hz while room loss on the left side is noticable.
I stood on the right side (stronger side) and played the left channel with my tone generator. When I stand by the right speaker with only the left playing, the low bass is as strong as when the right speaker alone is playing. The room has a tremendious effect on the low friquency response.
With a good DC amplifier and preamp (Marantz 300DC and 3650) I never felt the need for a sub woofer. For me, a sub woofer would be in the range of an 8 ft3 cabinet with the 2245. Not enough real estate at the moment.

Ron

[Zilch]

Cool, Ron: Do you have a pic or two for us?

Also, what'd you do for the UHF?

You're using 2425/6J on the 2370?

And which crossover? 3110A?

[Zilch]

There's an entire discipline and a substantial literature of building speaker systems. Read through the Audio Engineering Society anthologies to develop familiarity with the complexity of the science involved. Colleges offer degrees in it, too.

If one thing is clear in my experience in the Q&D project, it's that I can't just put any two drivers together with a crossover and expect them to work worth a whit as a system. The components interact with each other and the box that houses or mounts them in a complex fashion.

Two drivers is difficult. Add more, and the difficulties compound. All must complement each other in a very precise way for a successful outcome. It's VERY easy to fail at this. Zilch knows.

Thus, I (and others) look to the work and experience of the engineers at JBL for guidance, in the products they have produced, and in the literature they have published. I can virtually guarantee that your system was not assembled on a whim, and that's why I encourage you to complete it and give it a good listen in your environment according to your musical tastes as a baseline for any "improvements" you may subsequently wish to consider or accomplish.

Does it not suffice that a crossover would deal with the eminent problems the presence of a lot of drivers would have on a system? All the drivers, from my understanding, reproduce particular output in the spectral length, and in areas where this reproduction overlaps, a crossover is brought in to redress that problem.

No crossover can compensate for the variables among drivers. Every one of them behaves differently, integrates differently with others, and has a different "voice," or sonic character. Even drivers with identical or similar T/S parameters sound different. Every driver combination requires a specific crossover for optimum performance.

There ARE, howerver, active sytem "controllers" at the leading edge of this technology that can compensate and balance many of the variables at once. They actually "listen" to the system performance to automatically optimize it. Nonetheless, the outcome cannot surpass what is possible with any particular chosen driver combination, which can still sound like poop.

I struggle with the logic that because for scientific reasons we cannot have one driver that would produce the entire audio spectrum, therefore, we must subsequently settle for the next figure to one, which is two. The problem it seems to me is accurate representation and since a lot of drivers cannot do that, why could it be held then that two would or should?

No doubt, a well chosen assembly of multiple drivers can accurately reproduce the spectrum. Overcoming the complexity of getting them to work together to accomplish that is the problem. It's not, in my view, that a two-way can do it better, rather, that what it CAN do, is achieved with less difficulty, and thus, often, with greater success. If what you suppose could in fact be easily accomplished, none of us would be doing this.

The need to add a subwoofer to this system, which I would, for I take much of what Zilch says with much respect and read with close attention, shows that except if the drivers of any system is designed to reproduce most, since all can't be the case, of the audio spectrum, then there would always be the need for "supplemental" cabs on the system. The point here is one, is not possible, two is often found to be inadequate. Thus, where does that lead us? Does anyone here has a two ways system that reproduces at least 90% of the audio spectrum? Would they care to share?

4430 comes mighty close, and that's why I have dedicated so much effort to it. There are tradeoffs in that design, however. Some would say the woofers you have reproduce a much more articulate, punchy mid-bass, and consider the 2235 in the 4430 to be sluggish and "tubby" by comparison.

Asking the same driver to concurrently reproduce the lower midrange from 500 Hz to 1 kHz is also a stretch, though earlier "vintage" designs relied upon high-mass cones and weaker motors for even higher frequencies. I think we now mostly agree those systems do not sound very good in comparison to better designs. The tradeoffs are better understood, now, and the design parameters as well.

Further, the 4430 incorporates HF "boost" circuitry in the crossover to force the compression driver/Biradial horn combination to reproduce frequencies near the limits of their capabilities. It's a delicate balance that just works, but to some tastes, the quality of that HF leaves something to be desired in comparison to that produced by a true UHF driver.

You'll have to listen to a pair of 4430's to decide for yourself how successfuly they accomplish the objective. I can confidently state, though, they are mighty good at it, and very pleasant listening.

Would anyone with particular knoweldge of the horns discussed here PLEASE pitch in with what they know. In particular, advise on achieving optimum performance with this particular combination. Please.

2380 is a Biradial constant directivity horn, not unlike that used in 4430. Being a larger format, it will reproduce lower frequencies with competence, highly desireable since the woofer is no longer required to produce them. Your system crosses at 500 Hz, a BIG plus.

Also, it has a 90° (horizontal) x 50° (vertical) dispersion pattern, which JBL touts as optimum for home use, both for music and home theater. Read and learn about the specifications in the data sheet and the JBL Pro website technical library. I am just now beginning to work with the similar, next-generation 2352 horn. It's playing in the background here right now.

Truly, the music is yet to be enjoyed. And that raises yet another question. I know there are equipements used to measure sound production from speakers. Would someone please say how this is done, and say why one must do such measurements? Are they really accurate, given that they measure one thing and the music, the song itself, is made to sound in another way-not sure if that makes sense written down.

I have plastered RTA (Real Time Analyzer) curves all over the forums since I first acquired the capability. The device plays full-spectrum pink noise through the complete system, and "listens" to the reproduction, displaying the response at the various frequencies, every 1/3 or 1/6 octave from 20 Hz to 20 kHz.

There are more sophisticated and more accurate ways to measure system performance, to be sure, but an inexpensive ($150 - $300) RTA provides substantial information instantly. I can tell you from my own experience that it is quite impossible for me to know what is going on with a sound system by ear alone. I cannot assess comparative system component performance, either, without, at minimum, an RTA.

It's certainly true that no measurement, or combination of measurements, for that matter, defines how the system sounds playing music. However, certain fundamentals regarding the frequency response must be in place or the system is GUARANTEED not to sound good. That's what an RTA measures, and I see many more forum members acquiring the capability to objectively evaluate and improve their sound systems using one.

[johnaec]

2380 is a Biradial constant directivity horn, not unlike that used in 4430. Being a larger format, it will reproduce lower frequencies with competence, highly desireable since the woofer is no longer required to produce them. Your system crosses at 500 Hz, a BIG plus.

Also, it has a 90° (horizontal) x 50° (vertical) dispersion pattern, which JBL touts as optimum for home use, both for music and home theater. Read and learn about the specifications in the data sheet and the JBL Pro website technical library.

Actually, the 2380 is spec'd as a 90 x 40 horn. And while the 2380 may provide full loading down to 500 hz, pushing either the 2445/2446 or 2380 down to 500hz is a real stretch - both are about 10db down at 500hz, and the power rating is also lowered down there.

Edit: I went back and looked at the specs, and the 2445/2446 actually do 500hz pretty well on a plane wave tube. It's with the 2380 that it's about 10db down at 500hz.

On another note - I just BIN'd a pair of 3110A's, (800hz), to try with my 2206H/2226H + 2446J/2380A/2385A combos...

John

[jbl]

Cool, Ron: Do you have a pic or two for us?

Also, what'd you do for the UHF?

You're using 2425/6J on the 2370?

And which crossover? 3110A?

Hi Zilch,
Sorry, no pictures as of yet. I'll be getting a Digital Camera soon. Will post pictures.
I'm using the 2425 and 2370 for the HF response. A 2 way system. The HF response is very life like. Almost 3D! You can see the sharpe edges of the cymbal in your mind with the right recordings.
Yes. The crossover is the 3110A. Set to its MEDIUM position. The crossover is seamless between the 2225 and 2425.
Great sound!

Ron

[Zilch]

Thank you, Ron. We'll watch for the pics!

I'm bettin' nestawasright has some good listening ahead once his system is operating properly....

[Zilch]

On another note - I just BIN'd a pair of 3110A's, (800hz), to try with my 2206H/2226H + 2446J/2380A/2385A combos....

That sounds like a good option to try. 2225/2226 can play 800 Hz easily in that system, I'd think....

[duaneage]

Hate to be nitpicky, but please do the math before quoting misleading values.

speed = wavelength x frequency

Speed of sound at sea level is 1,116 ft/sec

1,116 ft/sec / 30 cycles/sec = 37 feet (not 60 feet)

1,116 ft/sec / 15,000 cycles/sec x 12 in/ ft = 0.9 inches (not 1/2 inch)

(Sorry, I write technical noise and air quality studies for a living.)

wavelength(inches) = frequency / 13,500

I use this for aligning voice coils in a multiple driver system when vertical alignment of the voice coils is not possible.

I come up with 1.1" for 15khz

Not saying mine is right, just what I have used and it works fairly close for me.

Ron

Whatever.

Someone come up with the right exact formula and settle it.

My Point is still made.