New driver and horn for my 'speakers: JBL 2450J + Truextent BeX4016 + Arai horn

[marco_gea]

I'm starting this thread to document the journey I've embarked on, namely replacing the mid-range section of my current 3-way loudspeakers:
http://www.homebuilthifi.com/project/9409

Intrigued by a post that I read here a while back:

An interesting JBL driver candidate for diaphragm replacement with Be would be the 2450. This driver was originally developed to accommodate a Be diaphragm as JBL was considering going head to head with TAD with that design. It was the reason for the development of the "Coherent Wave" phase plug. It allows for output from the diaphragm to be recombined in-phase at the driver throat in comparison to the out-of-phase output from a traditional phasing plug with variable length paths. According to JBL, it really only works with a Be diaphragm. The Al and Ti diaphragms are in breakup mode over most of their bandwidth, and as one designer said, are so "phasey" that there is limited benefit from the "Coherent Wave". However, a Be diaphragm, being pistonic for most of its bandwidth, would theoretically benefit the most.

This theoretically in-phase output would be a unique driver attribute compared to the TAD drivers and even JBL's current Be drivers. It would be interesting to have feedback if this two decade old technology actually results in a superior driver with a Be diaphragm.

I decided to purchase a pair of second-hand 2450J drivers and send them to Guido Behringer to have Truextent BeX4016 diaphragms installed in them.

I intend to pair these with Yuichi Arai's A320FL radial horns, which were originally documented in the May 1992 issue of the Japanese magazine "MJ":

7563075631

This is an interesting design that adopts a Hypex expansion with a cut-off frequency Fc = 320Hz and T = 0.6, resulting in a very linear on-axis frequency response down to Fc when paired with JBL 2441 drivers, as well as good horizontal directivity control to approx. 10 kHz (I intend to use a supertweeter for the last octave or so):

75632

I had the horns made out of baltic birch plywood by a French company specializing in bespoke audio designs:

75633

Another interesting feature of this design is the fact that the throat adaptor operates a very smooth transition from the circular throat of the driver to the rectangular throat of the radial horn, while respecting the same throat exit angle of the "classic" JBL 2" drivers (244x and 2450) in the horizontal plane:

75634

I will post again in the next few days/weeks as I proceed to make the first measurements using my set-up (calibrated Beyerdynamic MM1 microphone + HolmImpulse software)...

Marco

[aertex]

Very interesting Marco. Thanks for posting. I will certainly be following up to see what you do and find. Great craftmanship on those horns!

[gibber]

Another interesting feature of this design is the fact that the throat adaptor operates a very smooth transition from the circular throat of the driver to the rectangular throat of the radial horn, while respecting the same throat exit angle of the "classic" JBL 2" drivers (244x and 2450) in the horizontal plane:

I will post again in the next few days/weeks as I proceed to make the first measurements using my set-up (calibrated Beyerdynamic MM1 microphone + HolmImpulse software)...

Marco


Hi Marco,

i tried this 'plug & Be fram with the single piece of JBL476 that i own (offers for a second one welcome...)
Used on a white whale, i found the "throatless" design of the 2450SL 1.5" core (476Be is very close to it) doesn't harmonise so well with short horns. Guess the length of throat-in-driver the EV HR series was designed for plays a role. I got better results using 2447 instead of 2450SL/476Be. It offers some 12mm or so length of throat and the same phase plug.

Why do i write this? Your horn is also short even if it has a bit more of its own throat section than the EVs. But if you bump into the same issue as me and can't realize the 300Hz that Yuichi-san did, maybe try the 2446 core -- the "Coherent Wave" design was used as well, and maybe the 2446 also is longer than it's Neodymium sibling.

Good luck with the project, eager to follow your progress here !

Ralph

[JeffW]

Hi Marco,

i tried this 'plug & Be fram with the single piece of JBL476 that i own (offers for a second one welcome...)
Used on a white whale, i found the "throatless" design of the 2450SL 1.5" core (476Be is very close to it) doesn't harmonise so well with short horns. Guess the length of throat-in-driver the EV HR series was designed for plays a role. I got better results using 2447 instead of 2450SL/476Be. It offers some 12mm or so length of throat and the same phase plug.

Why do i write this? Your horn is also short even if it has a bit more of its own throat section than the EVs. But if you bump into the same issue as me and can't realize the 300Hz that Yuichi-san did, maybe try the 2446 core -- the "Coherent Wave" design was used as well, and maybe the 2446 also is longer than it's Neodymium sibling.

Good luck with the project, eager to follow your progress here !

Ralph



JBLs model numbers are a bit confusing. Marco has the 2450, it is a 2" exit driver with a throat. The 2450SL is the 1.5" exit throatless driver. The 2450 would be closer in geometry to the 2446. Apologies if I misunderstood your post.

[bubbleboy76]

What is the current price of these Truextent-diaphragms?

[Odd]

Behringer-Electric (http://behringer-electric.de/index.php?option=com_content&view=article&id=9&Itemid=118)

[Dr.db]

Very nice horns, thanks for sharing! :)

[ivica]

I'm starting this thread to document the journey I've embarked on, namely replacing the mid-range section of my current 3-way loudspeakers: http://www.homebuilthifi.com/project/9409 Intrigued by a post that I read here a while back: An interesting JBL driver candidate for diaphragm replacement with Be would be the 2450. This driver was originally developed to accommodate a Be diaphragm as JBL was considering going head to head with TAD with that design. It was the reason for the development of the "Coherent Wave" phase plug. It allows for output from the diaphragm to be recombined in-phase at the driver throat in comparison to the out-of-phase output from a traditional phasing plug with variable length paths. According to JBL, it really only works with a Be diaphragm. The Al and Ti diaphragms are in breakup mode over most of their bandwidth, and as one designer said, are so "phasey" that there is limited benefit from the "Coherent Wave". However, a Be diaphragm, being pistonic for most of its bandwidth, would theoretically benefit the most. This theoretically in-phase output would be a unique driver attribute compared to the TAD drivers and even JBL's current Be drivers. It would be interesting to have feedback if this two decade old technology actually results in a superior driver with a Be diaphragm. I decided to purchase a pair of second-hand 2450J drivers and send them to Guido Behringer to have Truextent BeX4016 diaphragms installed in them. I intend to pair these with Yuichi Arai's A320FL radial horns, which were originally documented in the May 1992 issue of the Japanese magazine "MJ": 7563075631 This is an interesting design that adopts a Hypex expansion with a cut-off frequency Fc = 320Hz and T = 0.6, resulting in a very linear on-axis frequency response down to Fc when paired with JBL 2441 drivers, as well as good horizontal directivity control to approx. 10 kHz (I intend to use a supertweeter for the last octave or so): 75632 I had the horns made out of baltic birch plywood by a French company specializing in bespoke audio designs: 75633 Another interesting feature of this design is the fact that the throat adaptor operates a very smooth transition from the circular throat of the driver to the rectangular throat of the radial horn, while respecting the same throat exit angle of the "classic" JBL 2" drivers (244x and 2450) in the horizontal plane: 75634 I will post again in the next few days/weeks as I proceed to make the first measurements using my set-up (calibrated Beyerdynamic MM1 microphone + HolmImpulse software)... Marco Hi Marco, Very, very nice design and realization. So You have adopted FL design, so , may be horizontal dispersion would be a problem. Regards Ivica

[marco_gea]

Hi Marco, Very, very nice design and realization. So You have adopted FL design, so , may be horizontal dispersion would be a problem. Regards Ivica

Hello, I'm aware that the Yuichi's other designs that have "fins" (aka "vanes") manage to control horizontal directivity up to higher frequencies. This is to be expected.

However, the fins are also problematic in terms of increased internal reflections and diffraction.

In fact, TAD themselves also went from a horn with fins (the TH4001) to a "fin-less" (FL) one with the later TH4003 model.

As always, there's no free lunch.

But, according to the polar measurements published in MJ, the horizontal dispersion of the A320FL horn appears to be just fine up to approximately 10kHz. And I intend to use a supertweeter crossed over at around 7kHz anyway (yes, despite the Be diaphragms. I'll post more on the reason why soon).

Marco

[gibber]

JBLs model numbers are a bit confusing. Marco has the 2450, it is a 2" exit driver with a throat. The 2450SL is the 1.5" exit throatless driver. The 2450 would be closer in geometry to the 2446. Apologies if I misunderstood your post.

Hi Jeff, actually i am not sure if you misunderstood my post or not.

I was careful to point out that my trials were based on 1.5" drivers. There, the 2447 fitted with 476Be fram did better than the 476/2450Sl core on a short-throat horn (i believe it was HR6040A, but maybe it was HR9040A). These EV horns offer only 13mm of throat length before rapidly expanding.

In the 2nd paragraph of my post i tried to recommend 2446 (the 2" equivalent of 2447) in case the experiment with 2450 does show odd results (the horn Marco uses has 50mm of throat length before serious expansion occurs, so maybe this is a non-issue).
I was also careful in only hinting at a possibly longer throat section inside 2446 (compared to 2450). I actually never measured throat length of these myself, but if 2446 is mechanically done the same way like 2447, then it might offer a longer throat than 2450.

The issue of combined driver&horn throat length being too short was brought to my attention first when the JBL presenter on an AES convention here in Munich hinted at differences he got using 2447 and 2251 drivers on 2352 horns.
Ralph

[marco_gea]

I was also careful in only hinting at a possibly longer throat section inside 2446 (compared to 2450). I actually never measured throat length of these myself, but if 2446 is mechanically done the same way like 2447, then it might offer a longer throat than 2450.


This is interesting.

Based on what I could find and read, I thought the internal structure of the 2446 and 2450 drivers was identical (apart of course the former's Ferrite vs. the latter's Neodymium magnet).
In fact, I was under the impression that the internal conical throat section had remained the same all the way from the original 375 (=2440) to the 2441, 2445, 2446 and 2450.

If anyone has reliable information to the contrary, I would be most interested in reading it.

Marco

[marco_gea]

OK, here are the first measurements (MLS gated at 200Hz) performed outdoors.

First, the on-axis IR and frequency and phase response:

75712

+/- 2dB from 600Hz to 9kHz, not bad! Also, a very 'clean' IR with very little overshoot and ringing. I guess this is one of the main benefits of the Be diaphragm, but it also says something of the horn in terms of lack of reflections etc.

Incidentally, to my eyes this response curve looks nicer than that of JBL's 476Mg + H9900 horn as used in the K2 S9900 system:
http://www.audioheritage.org/vbulletin/attachment.php?attachmentid=48630&stc=1&d=1290270292
(http://www.audioheritage.org/vbulletin/attachment.php?attachmentid=48630&stc=1&d=1290270292)
(http://www.audioheritage.org/vbulletin/attachment.php?attachmentid=48630&stc=1&d=1290270292)It will also allow for a much simpler electrical high-pass, without all the parallel notches that have become so common in JBL's recent systems.
I'm not sure, but I tend to think that the fewer acoustic issues to 'solve' in the electrical domain, the better.

Then, the 30º off-axis response:

75713

This is less smooth, with some minor peaking at ~2kHz and then some beaming after 7kHz or so.
I was expecting this, hence my intention to use a supertweeter.

More to come...
Marco.

(http://www.audioheritage.org/vbulletin/attachment.php?attachmentid=48630&stc=1&d=1290270292)

[ivica]

OK, here are the first measurements (MLS gated at 200Hz) performed outdoors.

First, the on-axis IR and frequency and phase response:

+/- 2dB from 600Hz to 9kHz, not bad! Also, a very 'clean' IR with very little overshoot and ringing. I guess this is one of the main benefits of the Be diaphragm, but it also says something of the horn in terms of lack of reflections etc.

......
(http://www.audioheritage.org/vbulletin/attachment.php?attachmentid=48630&stc=1&d=1290270292)
(http://www.audioheritage.org/vbulletin/attachment.php?attachmentid=48630&stc=1&d=1290270292)It will also allow for a much simpler electrical high-pass, without all the parallel notches that have become so common in JBL's recent systems.
I'm not sure, but I tend to think that the fewer acoustic issues to 'solve' in the electrical domain, the better.

Then, the 30º off-axis response:

75713

This is less smooth, with some minor peaking at ~2kHz and then some beaming after 7kHz or so.
I was expecting this, hence my intention to use a supertweeter.

More to come...
Marco.

(http://www.audioheritage.org/vbulletin/attachment.php?attachmentid=48630&stc=1&d=1290270292)

Hi Marco,

Very nice result, especially very good off axis response relative to large 2inch horn throat.
It would be nice to explain us total sound picture when all drivers applied.

Regards
Ivica

[gibber]

I agree that the 0° amplitude response looks very good, indeed, in the range that interests first&foremost.
A loss of -20dB @19 kHz is however, more than i observed with Be frams. Not that it is of much consequence.
It could be the the fram is too far from the phase plug. Does the driver have something close to a "0", or does it have a larger number felt-pen written on the back plate just outside the fram mounting area?

on-axis phase response doesn't look like it's going to allow use below 800 Hz. Actually group delay display would be nicer for judging this.

How about K2/3/4 -- can you bring that into the plot ? -- would be interesting to see how the BW-fram fares mounted to this 2" horn.
On his site, Guido published comparisons between 476Be and Radian 951PB mod'ed by himself to BW-4016 fram and it looked like same-same to me on this 9800(?) 1.5" horn clone.

Ralph

[marco_gea]

on-axis phase response doesn't look like it's going to allow use below 800 Hz.

One must be mindful that a gated MLS measurement has a resolution equal to the gating frequency (in this case, 200Hz). This makes it unsuitable to discuss the low end of the response.

By way of comparison, here's the same measurement but with no gating, and with 1/9th octave smoothing:

75727

Marco

[jerv]

....used on a white whale, i found the "throatless" design of the 2450SL 1.5" core....

Ralph

The EV HR9040 is a real interesting horn, and I have alwas wanted to try a pair - but the documentation

(http://www.google.no/url?sa=t&rct=j&q=&esrc=s&source=web&cd=11&cad=rja&uact=8&ved=0ahUKEwi8rILGqOzRAhVFEpoKHW6BBYQQFghXMAo&url=http%3A%2F%2Fwww.electrovoice.com%2Fdownloadfi le.php%3Fi%3D971351&usg=AFQjCNF1qzUDg07fIjDuSmdQ04Hqn28j8A&sig2=nSKPWV4JPZDIVMP5yb2DGw)

states that the horn throat is only 1.3".
How does this match up with the JBL 1,5 inch throat?
Have you done any modifications?

Best,
Espen

[ivica]

I agree that the 0° amplitude response looks very good, indeed, in the range that interests first&foremost.
A loss of -20dB @19 kHz is however, more than i observed with Be frams. Not that it is of much consequence.
It could be the the fram is too far from the phase plug. Does the driver have something close to a "0", or does it have a larger number felt-pen written on the back plate just outside the fram mounting area?

on-axis phase response doesn't look like it's going to allow use below 800 Hz. Actually group delay display would be nicer for judging this.

How about K2/3/4 -- can you bring that into the plot ? -- would be interesting to see how the BW-fram fares mounted to this 2" horn.
On his site, Guido published comparisons between 476Be and Radian 951PB mod'ed by himself to BW-4016 fram and it looked like same-same to me on this 9800(?) 1.5" horn clone.

Ralph

Hi Ralph,

But pay attention that this is 2inch throat, without any stencil in the horn throat, so 10kHz is very good, from my experience.
Using the horn driver combo from say 700Hz and up, I think is OK, especially if 4th order hi-pass is applied.
Regards
Ivica

[marco_gea]

Hello,

here's some more info on this project as it unfolds.

The Woofer will be the same one that I was previously using, i.e., the Fostex FW405N (http://www.fostex.jp/products/fw405n/)

This is a modern woofer of recent design, with a hybrid cellulose+carbon fibre cone and a newly developed aluminum die-cast frame based on FEA.

Magnet is ferrite, and sensitivity is medium-high at ~93 dB/W(m).

The features that attracted me were:

(i) Medium Efficiency Bandwidth Product (EBP) = Fs/Qes = 56.

This allows the use of an overdamped bass-reflex box characterized by an almost 12dB/oct initial roll-off and associated low Group Delay, while still obtaining an F-3 of approximately 40 Hz in anechoic conditions (which leads to a nice almost flat response in room, given the typical Room Gain over the last octave).

Also note how the excursion is contained under Xmax down to 20Hz with 130Wpc in input.

Vb = 2 * VAS * Qt^2 = 180 L
Fb = 0.39 * Fs/Qt = 23 Hz

7579775798

(ii) Very low mechanical damping = Fs/Qms = 4.0.

This, in my experience, is a recipe for excellent low-level detail retrieval. In other words, the woofer sounds good and does not mask or smear micro-details also at very low SPLs.

By way of comparison, this is roughly the same mechanical damping as that of the TAD TL-1601a (Fs/Qms = 4.1), while only slightly higher than that of the JBL 1501AL-2 used in the flagship Everest DD67000 system (Fs/Qms = 3.7).

Instead, it is HALF of that of the more PA-oriented JBL 2226H (Fs/Qms = 8.0) and of that of the venerable JBL 2235H (Fs/Qms = 7.9).

(iii) Extremely low distortion.

The levels of 2nd and 3rd harmonic distortion are amongst the lowest ever measured by the German magazine "Hobby-Hi-Fi":
< 0.1% at 90dB(1m) all the way down to 20Hz (measured on an infinite baffle).

75799

[marco_gea]

In my current speakers, I am using the Fostex T925A ring radiators (http://www.fostex.jp/products/t925a/).

For this new endeavour, however, I decided that I wanted to try and stay with Beryllium diaphragms all the way to the very top of the frequency range.

I therefore located and purchased a pair of Pioneer PT-R7 III
(http://audio-database.com/PIONEER-EXCLUSIVE/unit/pt-r7iii-e.html)

This is an extraordinary ribbon tweeter which is unfortunately no longer manufactured.

It features a Beryllium diaphragm loaded by a short front horn machined from solid aluminium, and a massive AlNiCo magnet, resulting a sensitivity of 97dB/W(m) and an outstanding frequency response that reaches out to 120 kHz (!).

75800

The associated Impulse Response, calculated by way of reverse Fourier transform using the published frequency response (unfortunately, I do not have access to a microphone that is linear beyond 20 kHz) is very clean and "fast", as expected:

75801

[marco_gea]

For the Woofer-Mid crossover, I got inspiration from the design used by Pioneer in their TAD/Exclusive monitors

(e.g., http://audio-database.com/PIONEER-EXCLUSIVE/speaker/model2402-e.html),

and later reprised by Shozo Kinoshita in his Rey Audio monitors

(http://www.reyaudio.com/large-e.html),

which is based on an asymmetrical 6th-order electrical low pass / 2nd order electrical high pass, with both drivers down -6dB at Fx.

In my case, the chosen crossover frequency is 600 Hz, so as to stay approximately one octave above the horn's cut-off.

When both the Woofer and Mid are connected with positive polarity and the Mid is physically set back so as to create a suitable offset between the two acoustic centres, this crossover results in the emissions of both drivers to be in phase over a relatively wide frequency range around the crossover frequency, as can be seen in the figure provided. As further proof of the excellend phase match, when reversing the polarity of the Mid, a deep notch at Fx is observed.

75802

This type of crossover is handy in several ways:

(i) the front-to-back offset allows the convenient positioning of the horn-loaded mid atop the Woofer box without requiring any form of delay (be it analogue or digital);

(ii) the same offset also results in the two impulse responses to be almost perfectly 'time aligned', i.e. both traces leave the horizontal line simultaneously;

(iii) the 6th-order low pass effectively does away with any unwanted resonances at the top of the woofer's operating range.

Two woofer section will adopt a balanced topology to reduce crosstalk with the mid and tweeter sections, and the compression driver will be attenuated with the Fostex R100T autoformer (http://www.fostex.jp/products/r80br82br100t/), whose 8 Ohm shunt resistor also handily leads to an almost perfectly level 8 Ohm impedence.

[marco_gea]

For the second crossover between Mid and (super)-Tweeter, I opted for a simple symmetrical 2nd-order "Constant Power" design at 7kHz, with both drivers down -3dB at Fx.

Inevitably, given the short wavelengths at play and the physical separation between the acoutst centres, the super-Tweeter will not be time-aligned.

However,

(i) this was not deemed important at such high frequencies, where the human auditory system's sensitivity to phase is reduced;

(ii) the resulting comb filtering between the Mid and Tweeter outputs falls within the Equivalent Rectangular Bandwith (ERB), and is thus essentially inaudible.

Greg Timbers also seems to agree on this, as he said:
"The comb filter effect is normal and cannot be eliminated, even with the use of electronic delay. [...] In spite of the curve, we have had no negative comments about the 045Be and the choppy response through the crossover range"

Additionally, having an intentionally misaligned constant power crossover at such high frequencies results in a 1/6th octave-smoothed summed response that is essentially invariant over a +/- 30 degree listening angle, since any additional misalignment introduced by moving the head laterally is swamped by the >360 degree original misalignment.

75803

Finally, the resulting modelled impedance curve of the complete system is very flat and with only moderate phase angles throughout:

75804

...MORE TO COME! :-)

[gibber]

The EV HR9040 is a real interesting horn, and I have alwas wanted to try a pair - but the documentation
...
How does this match up with the JBL 1,5 inch throat?
Have you done any modifications?

Hi Espen,

it almost doesn't matter if you insert an adaptor or not. The HR series even has ex-factory that three hole mounting arrangement for the Altec 1.4" drivers of the time. Invitation for users to create a 2.5mm mismatch.
For fear of nasty visitor comment i wouldn't dare a permanent install of a 1.5" without a conical flange piece, but truth be told i never saw a significant effect, just a bit of ripple here or there, sometimes even smoothening a response taken with the "correct" transition fitted :confused:.

EV published measurements of even more gross throat size mismatch in one of the "PA Bible" articles as a quick fix guide for roadies during perfomances. Check it out.

Ralph

[gibber]

Hi Ralph,

But pay attention that this is 2inch throat, without any stencil in the horn throat, so 10kHz is very good, from my experience.
Using the horn driver combo from say 700Hz and up, I think is OK, especially if 4th order hi-pass is applied.
Regards
Ivica

Yes, agree this will work from 700 if x/o steeply, and "upstairs" the Be ribbon will help to offset matters.
Have you ever compared "same" motor in two exit versions, say 1.5 and 2.0 inches for >10 kHz Response?

My guess would be that say, a 2450SL 1.5" should perform the same way than a 2450 2". After all, both drivers would see further expansion into the horn they are bolted to. With the 2450 having part of that expansion bolted to it's magnet ex-JBL already.

In that sense, the smaller 1,5" exit could only possibly bring a benefit in extension >10kHz, if the horn accepts 1,5" and then rapidly expands (such as the PT Family or the 2352/2353 horns).

What do you think?

[ivica]

Yes, agree this will work from 700 if x/o steeply, and "upstairs" the Be ribbon will help to offset matters.
Have you ever compared "same" motor in two exit versions, say 1.5 and 2.0 inches for >10 kHz Response?

My guess would be that say, a 2450SL 1.5" should perform the same way than a 2450 2". After all, both drivers would see further expansion into the horn they are bolted to. With the 2450 having part of that expansion bolted to it's magnet ex-JBL already.

In that sense, the smaller 1,5" exit could only possibly bring a benefit in extension >10kHz, if the horn accepts 1,5" and then rapidly expands (such as the PT Family or the 2352/2353 horns).

What do you think?

Hi gibber,

From my experience with 2450-2" and 2450-1.5", applied to the almost the same horn, but with proper horn throat relative to the used driver ( 2" or 1.5") have shown that I have got +/-30 deg.off-axis dispersion up to the 10kHz with 2" driver, but about 14kHz when 1.5" driver is used.

In the horn throat is have been applied pi/4*D ( D=2" or 1.5") vertical size reduction ( like DD66000 horn as an example, or 2328 adapter).
From the mentioned experiments I can say that +/-30 deg.off-axis dispersion can be get up to the frequency calculated as: 34500/(D*pi/4)*1.2, where D is in [cm] is driver output diameter.

regards
ivica

[gibber]

...proper horn throat relative to the used driver ( 2" or 1.5") have shown that I have got +/-30 deg.off-axis dispersion up to the 10kHz with 2" driver, but about 14kHz when 1.5" driver is used ... From the mentioned experiments I can say that +/-30 deg.off-axis dispersion can be get up to the frequency calculated as: 34500/(D*pi/4)*1.2, where D is in [cm] is driver output diameter

So Ivica,

doesn't that basically bring down the argument to that of slot size (rectangular or circular "slot") for being the source of the better dispersion?



In the horn throat is have been applied pi/4*D ( D=2" or 1.5") vertical size reduction ( like DD66000 horn as an example, or 2328 adapter)...

The EV HR9040 is a real interesting horn ... but the documentation ... states that the horn throat is only 1.3".
How does this match up with the JBL 1,5 inch throat?

The above arrangement of size reduction is perhaps also relevant to Espen's concern: it happens all the time! Look at how the original HR9040 throat reduces the 33mm entry to just 28mm as a matter of a few millimetres of pathlength into the horn (now oval-looking):

758337584175831
(Source: http://www.diy-audio.narod.ru/litr/Keele_1975-05_AES_Preprint_-_Whats_So_Sacred_Exp_Horns.pdf)
75832

The above Keele reference lists a P.W.Klipsch 1945 (filed) patent that was commented on here (http://www.suiter.com/patent-of-the-day-loud-speaker-horn/) to the effect that (i paraphrase) First, expansion is only in at single plane for a part of the length to bring radiation angle of the horn up to desired target in this plane; 2nd, expansion takes place at right angles to said 1st plane while continuing to expand in the angle reached in the first expansion. Expansion stops where targeted mouth area is reached.

So, all horn designers do what they please and place "obstacles" at will in order to get their designed pattern
Ralph

[ivica]

So Ivica,

doesn't that basically bring down the argument to that of slot size (rectangular or circular "slot") for being the source of the better dispersion?





The above arrangement of size reduction is perhaps also relevant to Espen's concern: it happens all the time! Look at how the original HR9040 throat reduces the 33mm entry to just 28mm as a matter of a few millimetres
(Source: http://www.diy-audio.narod.ru/litr/Keele_1975-05_AES_Preprint_-_Whats_So_Sacred_Exp_Horns.pdf)
......
The above Keele reference lists a P.W.Klipsch 1945 (filed) patent that was commented on here (http://www.suiter.com/patent-of-the-day-loud-speaker-horn/) to the effect of (i parphrase) ... first, expansion is only in at single plane for a part of the horn length to bring radiation angle up to desired target in this plane; 2nd, expansion takes place at right angles to said 1st plane while continuing to expand in the angle reached in the 1st expansion. Expansion stops where targeted mouth area is reached.

So, all horn designers do what they please and place "obstacles" at will in order to get their designed pattern
Ralph

Hi Ralph,

Such throat size reduction can be seen on M2 horn too, as 39mm throat has been reduced (partially ) to round about 19mm in order to involve diffraction in the horn throat so HF off axis dispersion woild be enhanced over +/- 45 degs.
Regards
Ivica

[marco_gea]

At last I found the time to work on the crossovers, and to assemble the system.
A couple of pictures below.
Keen observers will notice that I reverted to my tried-and-trusted Fostex T925A "bullet" tweeters.
In-room measurements and listening impressions will follow...

7594675947

[marco_gea]

Here's the system's FR measured from the listening position, superimposed on the target described in Olive et al., 2013 (http://www.aes.org/e-lib/browse.cfm?elib=17042).

75948

The L/R differences below the room's Schroeder frequency (Fc) are mostly due to the fact that the left speaker is closer to a side wall, and hence gets more boundary reinforcement. But notice the excellent linearity of +/-dB all the way from ~300Hz to 20kHz.

Cheers,
M.

[gibber]

Hi Marco

the setup looks good! Kids (2..4yrs), or a cat in the home? Would look even better (i think) without the grille in front of the 38cm unit, but with cats...

Change of plan: why did you choose to go for the Fostex T925 instead of the TAD/Pioneer Be-Ribbon? I have 3 different pairs of them. The PTR version you have is less simply made and even more powerful than the high-reputation PTR-9. Yours has 4 of the 7 Alnico cylinders used in early TAD TL-1601B (later 1601B has very different magnet). I'd like to see any comments on sound/usable range/etc before putting in my €0.02. Posts/comments on T825 (which i believe uses the same fram as T925) at the end of this (http://www.audioheritage.org/vbulletin/showthread.php?6368-Ring-Radiator-Comparisons/page1) very interesting (one of my fav) threads.

Ralph

[marco_gea]

Hi Marco

the setup looks good! Kids (2..4yrs), or a cat in the home? Would look even better (i think) without the grille in front of the 38cm unit, but with cats...

Change of plan: why did you choose to go for the Fostex T925 instead of the TAD/Pioneer Be-Ribbon? I have 3 different pairs of them. The PTR version you have is less simply made and even more powerful than the high-reputation PTR-9. Yours has 4 of the 7 Alnico cylinders used in early TAD TL-1601B (later 1601B has very different magnet). I'd like to see any comments on sound/usable range/etc before putting in my €0.02. Posts/comments on T825 (which i believe uses the same fram as T925) at the end of this (http://www.audioheritage.org/vbulletin/showthread.php?6368-Ring-Radiator-Comparisons/page1) very interesting (one of my fav) threads.

Ralph

Hi Ralph,

yes, there's a cute 2.5-yr old at home... ;-)

As for the tweeter, the Pioneer ribbon is surely a great product in and of itself. However, it seems as if it was really primarily meant as a SUPER-tweeter, to be used as an "add-on" to an already "complete" system, to supplement the ultrasonics (>20kHz).

When instead using it as a tweeter crossed over lower (7kHz in my case), I found that its frequency response is a bit ragged in the stop band (which causes issues with the passive crossover), and then it slopes down beyond 10kHz, becoming approximately -5dB at 20kHz (the upper frequency limit that I can measure - but the Pioneer then goes on and on... till over 100kHz!).

Instead, I measured the Fostex tweeter as almost perfectly flat from 3kHz to 20kHz, which makes filtering it a breeze, and then it is also much more efficient, which subjectively makes for a better dynamic match for the compression driver below.

Oh well, you live and learn...

Cheers,
Marco

[Dr.db]

The Woofer will be the same one that I was previously using, i.e., the Fostex FW405N (http://www.fostex.jp/products/fw405n/)


This Fostex has a Rated Input of 50w, that seems very weak to me for a bassdriver. Musical power is 150w, but again that´s only half of that what the JBL 2235h is capable of.
Does Fostex has different power-ratings or are these drivers just not intended for driving high levels?

Besides that, all the other parameters look very temping.

[gibber]

Hi Ralph,

yes, there's a cute 2.5-yr old at home... ;-)

I remember the time... Enjoy it despite the occasional drawback it entails ;)



...using it as a tweeter crossed over lower (7kHz in my case), I found that its frequency response is a bit ragged in the stop band (which causes issues with the passive crossover), and then it slopes down beyond 10kHz, becoming approximately -5dB at 20kHz (the upper frequency limit that I can measure - but the Pioneer then goes on and on... till over 100kHz!)

You have cheap 3D printing now, here's your solution to raggedness from the year of Quincy Jones "Thriller" with that flat-as-a-witches guest singer (Jones or any brass considered OTT today, poly-anything will do nicely)75976759777597875979

[marco_gea]

This Fostex has a Rated Input of 50w, that seems very weak to me for a bassdriver. Musical power is 150w, but again that´s only half of that what the JBL 2235h is capable of.
Does Fostex has different power-ratings or are these drivers just not intended for driving high levels?

Besides that, all the other parameters look very temping.

Yes, the Fostex woofer is not intended for PA-high levels. It is designed first and foremost as a top-notch home hi-fi transducer, which entails a different set of design trade-offs. It sounds wonderfully detailed even at very low listening levels, and can still comfortably reach 115dB peaks at 1m, which translates to almost 110dB for a stereo pair at a typical domestic listening distance. Plenty loud enough for listening at home at realistic levels, even with very dynamic (20dB crest factor) recordings.

[marco_gea]

You have cheap 3D printing now, here's your solution to raggedness from the year of Quincy Jones "Thriller" with that flat-as-a-witches guest singer (Jones or any brass considered OTT today, poly-anything will do nicely)

Hi,

I'm not entirely sure what I'm looking at there (nor what Q. Jones has to do with any of it!).
Is that a 3D-printed replacement waveguide for the PT-Rx ribbon tweeters?

Marco

[ivica]

I'm starting this thread to document the journey I've embarked on, namely replacing the mid-range section of my current 3-way loudspeakers:
http://www.homebuilthifi.com/project/9409

Intrigued by a post that I read here a while back:

An interesting JBL driver candidate for diaphragm replacement with Be would be the 2450. This driver was originally developed to accommodate a Be diaphragm as JBL was considering going head to head with TAD with that design. It was the reason for the development of the "Coherent Wave" phase plug. It allows for output from the diaphragm to be recombined in-phase at the driver throat in comparison to the out-of-phase output from a traditional phasing plug with variable length paths. According to JBL, it really only works with a Be diaphragm. The Al and Ti diaphragms are in breakup mode over most of their bandwidth, and as one designer said, are so "phasey" that there is limited benefit from the "Coherent Wave". However, a Be diaphragm, being pistonic for most of its bandwidth, would theoretically benefit the most.

This theoretically in-phase output would be a unique driver attribute compared to the TAD drivers and even JBL's current Be drivers. It would be interesting to have feedback if this two decade old technology actually results in a superior driver with a Be diaphragm.

I decided to purchase a pair of second-hand 2450J drivers and send them to Guido Behringer to have Truextent BeX4016 diaphragms installed in them.

I intend to pair these with Yuichi Arai's A320FL radial horns, which were originally documented in the May 1992 issue of the Japanese magazine "MJ":

7563075631

This is an interesting design that adopts a Hypex expansion with a cut-off frequency Fc = 320Hz and T = 0.6, resulting in a very linear on-axis frequency response down to Fc when paired with JBL 2441 drivers, as well as good horizontal directivity control to approx. 10 kHz (I intend to use a supertweeter for the last octave or so):

75632

I had the horns made out of baltic birch plywood by a French company specializing in bespoke audio designs:

75633

Another interesting feature of this design is the fact that the throat adaptor operates a very smooth transition from the circular throat of the driver to the rectangular throat of the radial horn, while respecting the same throat exit angle of the "classic" JBL 2" drivers (244x and 2450) in the horizontal plane:

75634

I will post again in the next few days/weeks as I proceed to make the first measurements using my set-up (calibrated Beyerdynamic MM1 microphone + HolmImpulse software)...

Marco

HI MARCO,

As I know JBL Be diaphragm differs from The BeX4016, so the results would be different, especially over 10kHz, where phasing is important.
On some old papers it has be mentioned that a lot of resonances can be established in the CD driver in the diaphragm and phase plug interaction, so I believe tha at JBL they have payed their attention on all of that when they have constructed their Be diphragm, may be BeX has applied the same results, but as BeX suspension is of different type then JBL, I would expect different interaction too.
May be much cheaper (5 times) good results can be get with the JBL Tl SL diaphragms.

Regards
Ivica

[gibber]

Hi,

I'm not entirely sure what I'm looking at there (nor what Q. Jones has to do with any of it!).
Is that a 3D-printed replacement waveguide for the PT-Rx ribbon tweeters?

Marco

You were looking at a PTR7/III. Had you never opened yours? 3D? -- no.
It is brass from the year Quincy recorded that brass-laden killer of a track.
Which might serve as a blueprint for your 3D printed version, if of interest to avoid what you called "rugged".

All the best,
Ralph


N.B.: brass = non-magnetic, be sure to have no spoons, no screwdrivers, no coins, etc anywhere near if you haven't opened the thing before.

[marco_gea]

HI MARCO,

As I know JBL Be diaphragm differs from The BeX4016, so the results would be different, especially over 10kHz, where phasing is important.
On some old papers it has be mentioned that a lot of resonances can be established in the CD driver in the diaphragm and phase plug interaction, so I believe tha at JBL they have payed their attention on all of that when they have constructed their Be diphragm, may be BeX has applied the same results, but as BeX suspension is of different type then JBL, I would expect different interaction too.
May be much cheaper (5 times) good results can be get with the JBL Tl SL diaphragms.

Regards
Ivica

Hi Ivica,

in reply to the points you made:

- yes, of course the JBL and Truextent dias are different, especially in terms of the suspension used;

- the 2450 driver has a "Coherent Wave" phase plug that was reportedly developed with the specific intention to use a Be diaphragm (alas, then the marketing guys decided against it), so I would tend to believe that it ought to be well-behaved in terms of phasing;

- the 2450 driver also differs from the modern Be-equipped JBL drivers in that it has an internal conical throat - this one may regard as a good thing (it helps load the diaphragm to lower frequencies) or a bad thing (it causes beaming in the top octave);

- yes, I'm sure the Ti SL dias as originally used in the 475Nd (K2 S9500 & K2 M9500 systems) are also a good option. Based on the FR for the 475Nd in the M9500 horn, though, it appears that they still exhibit some nasty ringing, which the Be dias don't (at least up to 20kHz).

Regards,
Marco

[marco_gea]

You were looking at a PTR7/III. Had you never opened yours? 3D? -- no.
It is brass from the year Quincy recorded that brass-laden killer of a track.
Which might serve as a blueprint for your 3D printed version, if of interest to avoid what you called "rugged".

All the best,
Ralph


N.B.: brass = non-magnetic, be sure to have no spoons, no screwdrivers, no coins, etc anywhere near if you haven't opened the thing before.

OK, cool thanks.
But I don't think I'll embark on any such customization of the Pioneer tweeters any time soon.

Marco

[marco_gea]

...after a few minor tweaks to the passive crossover.

76042

The general trend is now very similar to the one recommended in the JBL synthesis installation manual:

76043

Sounds good ;-)