Beyond the Ariel - Page 344 - diyAudio
Go Back   Home > Forums > Loudspeakers > Multi-Way

Multi-Way Conventional loudspeakers with crossovers

Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Reply
 
Thread Tools Search this Thread
Old 21st April 2008, 12:45 AM   #3431
diyAudio Member
 
Lynn Olson's Avatar
 
Join Date: Jul 2004
Location: Northern Colorado
Quote:
Originally posted by Robh3606
Hello Lynn

The 1952 Reverse Flair horn I don't believe is the starting point for the modern CD horn. If I am not mistaken it goes back to 1975 and D.B. Keele's work at Electro Voice.

http://www.xlrtechs.com/dbkeele.com/PDF/Keele%20(1975-05%20AES%20Preprint)%20-%20Whats%20So%20Sacred%20Exp%20Horns.pdf
Well, the design goals of the University "Reverse Flare" are similar, although certainly poorly argued, and it does precede Keele's work by 23 years. Just as the AR-1 loudspeaker preceded the publication of Neville Thiele and Richard Small by nearly twenty years, I believe prior art, no matter how crudely realized, needs to be given credit. I believe the tendency in audio to minimize, or simply forget, the past has been extremely destructive to continued forward progress, since it leads to re-inventing the wheel over and over again.

Quote:

"This is the Engineering Report describing the JBL "diamond suspension". Apparently, it was developed as a response to quality-control problems with the tangential suspension in the JBL 2420 compression-driver diaphragm."

How did you come to that conclusion?

From page 723:

Shifting back to a 200-200 000-Hz range, Fig. 8 shows the motional impedance of a rejected transducer. The second pole is just as high (amplitudewise) as it was in Fig. 6, but now it appears at 22 kHz, where it is swamped by the difference inductance mentioned above. Fig. 9 shows the response and impedance curves of this "bad" transducer. Because the second pole is at a higher frequency, it is unable to participate in lifting the frequency response in the 15-kHz region, and the transducer tends to follow a much more "theoretical" curve based on a simple parallel tank circuit for motional impedance. There is some increased output above 19 kHz, resulting from this high-frequency pole, but it comes too late to be of use at audio frequencies. The response of the "good" transducer is shown for comparison. It is well to note the dimensional differences between these two transducers are of the order of 0.003 mm and cannot be readily determined before the transducer is assembled.

6 A New Transducer

Because the tangential surround is so highly sensitive to small dimensional differences, the diamond-shaped pattern of Fig. 3(b) is incorporated into the 45-mm transducer. This pattern seems to be less sensitive, and uniformity is easier to achieve.

The impedance curves for this new transducer are shown in Fig. 10. This shows a well-defined pole at 16 kHz and some additional activity between 30 and 50 kHz. This additional activity will be investigated at a later date. The audiofrequency response range is illustrated in Fig. 11 and shows smooth response to the 16-kHz second pole.


I've worked in manufacturing at Audionics and Tektronix. The problem with the tangential surround could be described as a quality-control issue, a product that is difficult to manufacture consistently, or a fundamental engineering problem. All are valid points of view. Altec continued to manufacture the tangential surround, JBL changed over to the diamond surround, and Emilar & Radian made plastic surrounds.

The argument in the original paper is a bit of a straw-man argument, since we are never shown measurement data from a "bad", out-of-spec diaphragm using the new technology. Maybe the new diaphragms are all perfect? This seems unlikely in any manufacturing environment - there's always a certain percentage of duds.

We are shown "good" old technology, "bad" old technology, a "good" example of the new technology, and no statistical data on the rejection rate of old and new. The paper would be a little stronger if this had been shown.


Quote:

"The only two horns that were never identified as horns by any of the listeners,"

That not what it says. There are contradictory statements in the text. Here's an example page 10:

"Only one short horn, sample 11, was ever identified as a horn and then only by one "golden eared" professional sound engineer."

From that statement I would conclude that none of the short horns were identified as horns

I can see why it was infamous. You can interpret it either way. The main thing is to keep the horns short and obstruction free.

Rob
From page 32 of Holland and Newell:

These two hypotheses agree with the observation that of the two horns in the test that produce negligible mouth reflections, samples 8 and 13, neither was ever identified as a horn, and the short horn, sample 8, did not sound like the direct-radiating reference B.

From Appendix 1:

Reference B: Son Audax PR17/HR100/1AK7. Midrange direct-radiating paper-cone loudspeaker of nominal 6 1/2-in (165-mm) diameter.
...
Sample 8: AX2 horn/Emilar EK175 driver. Short axisymmetric horn of glass-fiber construction with rapid flare terminating in a medium-size mouth. Compression driver as sample 1.
...
Sample 13: Altec 806C horn/Emilar EK175 driver. Large multicellular horn with eight individual flares of sheet aluminum construction joined to a single throat via a cast aluminum manifold. Compression driver as sample 1.


I agree this paper is problematic, with the much larger Altec 806 multicell grouped with the much smaller AX2 horn. The Altec is 600mm long yet is never identified as a horn, is claimed (without evidence) that the mouth produces "negligible" reflections, and obviously has internal reflections in the area where the transition from the round-throat compression driver is made to a square-section throat which changes again to the square entrances of the eight smaller sub-horns. On the basis of diffraction alone, that would seem to put the multicell in the worst possible category.

Yet I am forced to agree, at least in purely subjective terms, with the conclusion. I've heard large Altec multicells for myself, and was surprised at how low the horn coloration seemed to be - certainly in a different class than Altec sectorals or JBL Bi-Radials, which to me have a fair amount of horn coloration.
  Reply With Quote
Old 21st April 2008, 01:12 AM   #3432
diyAudio Member
 
Lynn Olson's Avatar
 
Join Date: Jul 2004
Location: Northern Colorado
Default More


1966, JK Hilliard & JA Renkus, Development of Horn-Type Moving Coil Driver Unit. The predecessor of the Emilar and Radian family of aluminum-diaphragm, plastic-surround compression drivers. I am not sure if Altec ever produced this 4-inch diaphragm unit.

2002, Voishvillo, Nonlinearity in Horn Drivers. An interesting mathematical model of the sources of compression-driver distortion, compared against measurements of a real-world driver and horn. Note the 12V input to the test system (page 20) results in 120 dB SPL at one meter - a level high enough to destroy audiophile-efficiency loudspeakers, where it would correspond to a kilowatt (or more) of amplifier power.
  Reply With Quote
Old 21st April 2008, 01:44 AM   #3433
diyAudio Member
 
Lynn Olson's Avatar
 
Join Date: Jul 2004
Location: Northern Colorado
Quote:
Originally posted by caninus80
so, I make wrong decison with 399 maybe in some next project.

I was trying single raal 14-150 with altec biflex and raal is too fast. The blend is not seamless. I have to look for another drivers.
Some advice from DIY friends?

Best
C
You can look at the problem of blending the ribbon (any ribbon) with the Altec Biflex in terms of moving masses - never mind the frequency-response data, it's not going to easy to match moving masses that are more than 100X different. The mass of a ribbon is measured in milligrams, and due to its low mass and relatively large area, receives almost all of its damping from the air-load. True ribbon loudspeakers are little more than scaled-up microphones - more elegant control of the shape of the magnetic field in the gap, clever damping patterns on the ribbon, different tensioning systems, etc. etc. but they really are just big microphones.

The inverse is true of a cone driver working at the top of its usable range. The mass is anywhere from 20 to 70 grams for a typical midbass driver, and is strongly decoupled from the voice coil at the highest frequencies. Any damping that is present is a side-effect of losses in the mechanical construction of the diaphragm and details of the surround damping. With the big diaphragm going so many ways all at once, the dispersion pattern is going to be very level and frequency-dependent.

These dissimilarities appear in measurements of the distortion spectrum, energy storage in the CSD, and "clutter" in the tail of the impulse response. These measurements are only indirect indicators of very different acoustic radiators - and the ear is very good at identifying what acoustic radiators are made of. Heavy masses sound heavy, light masses sound light.

In subjective terms, all that comes to mind to blend the woofer and the ribbon are low-mass aluminum-cone drivers, like a 4X, 8X, or 16X array of the small Jordan-Watts (or clones), or (again) a compression driver with a well-behaved horn (and those aren't common). That gets the diaphragm masses down to a few grams, and extends HF respond well into the ribbon's natural territory.

Ideally, the array of little metal-cone drivers, or the compression driver, should have good enough HF response the ribbon isn't really necessary, and is just a nice add-on to give the speaker a bit more "air" and openness at the top of the range.

The Altec Biflex, although a very nice speaker, isn't really full-range. Frankly, neither are Fostex, Lowther, or Feastrex whizzer-cone drivers. Yes, they have output beyond 10 kHz, but the extreme-HF quality isn't there - the whizzer cones are so thoroughly decoupled at those frequencies there is almost no damping at all, and impulse and polar responses are chaotic.
  Reply With Quote
Old 21st April 2008, 03:19 PM   #3434
Salas is offline Salas  Greece
diyAudio Chief Moderator
 
Salas's Avatar
 
Join Date: Oct 2002
Location: Athens-Greece
Quote:
Originally posted by Lynn Olson
The Altec is 600mm long yet is never identified as a horn, is claimed (without evidence) that the mouth produces "negligible" reflections, and obviously has internal reflections in the area where the transition from the round-throat compression driver is made to a square-section throat which changes again to the square entrances of the eight smaller sub-horns. On the basis of diffraction alone, that would seem to put the multicell in the worst possible category.

Yet I am forced to agree, at least in purely subjective terms, with the conclusion. I've heard large Altec multicells for myself, and was surprised at how low the horn coloration seemed to be - certainly in a different class than Altec sectorals or JBL Bi-Radials, which to me have a fair amount of horn coloration.
Maybe there is some diffraction cancellation going on over the frontal arc between cell outputs?

Maybe Dr. Earl Geddes had analyzed the Altec multicell objectively and can shed some light to its particularities. Earl?
  Reply With Quote
Old 21st April 2008, 05:16 PM   #3435
diyAudio Member
 
Join Date: Aug 2006
Default Single WB to meet the RAAL ?

Quote:
Originally posted by Lynn Olson
You can look at the problem of blending the ribbon (any ribbon) with the Altec Biflex in terms of moving masses - never mind the frequency-response data, it's not going to easy to match moving masses that are more than 100X different. The mass of a ribbon is measured in milligrams, and due to its low mass and relatively large area, receives almost all of its damping from the air-load. True ribbon loudspeakers are little more than scaled-up microphones - more elegant control of the shape of the magnetic field in the gap, clever damping patterns on the ribbon, different tensioning systems, etc. etc. but they really are just big microphones.

How about this: To quote this post by yourself:

Quote:
Originally posted by Lynn Olson
The specs of the H 21 LB 15 and the more pricey H 21 LB 15 SAG look reasonable enough, and I guess the only difference between the two is the silver voice coil with silk insulation. Dunno about the merits of silver vs increased VC mass. They're not cheap at 596 and 736 euros respectively, but that's not quite as much as a Lowther, AER, or even more stratospheric Feastrex.

They do look like a good match for the ultrafast and dynamic RAAL tweeters, with a moving mass of 5.9 grams and extension out to 10 kHz, giving a wide range of potential crossover frequencies, anywhere from 2 to 10 kHz.
Frankly, that's what I'm considering -- a PHY H 21 LB 15 XOed @ 4.5kHz first order passive with the RAAL. Just need some nice fast bass to take below @ 250 Hz....
  Reply With Quote
Old 22nd April 2008, 05:34 AM   #3436
diyAudio Member
 
Lynn Olson's Avatar
 
Join Date: Jul 2004
Location: Northern Colorado
Well, I've never heard the PHY's, so any endorsement of them is cautious at best. On the other hand, Alexander, designer and builder of the RAAL ribbons, has suggested to me (more than once) the PHY midbass drivers, as well as vertical arrays of small metal-cone woofers, so you can take that for what it's worth.
  Reply With Quote
Old 22nd April 2008, 07:16 AM   #3437
diyAudio Member
 
Brett's Avatar
 
Join Date: Jan 2002
Lynn, I haven't had time to thoroughly read all of the papers you've linked, but it seems they all deal with metal and phenolic diaphrams. Any particular reason you're not mentioning the mylar/polyester diaphrammed units such as the Beyma CP380 and 385? Meant more theoretically than neccessarily relating tho those particular drivers. I have experience with the 380 and I beleive Duke (audiokinesis) uses the 385.
  Reply With Quote
Old 23rd April 2008, 06:43 AM   #3438
diyAudio Member
 
Join Date: Aug 2006
Quote:
Originally posted by Lynn Olson
Well, I've never heard the PHY's, so any endorsement of them is cautious at best. On the other hand, Alexander, designer and builder of the RAAL ribbons, has suggested to me (more than once) the PHY midbass drivers, as well as vertical arrays of small metal-cone woofers, so you can take that for what it's worth.

Lynn,

Gracious thanks for that. I understand your position (and have to agree) that PHY -- and the likes of Supravox, Fertin, PHL etc -- might prove to be nothing more than a "costly experiment" if things do not blend nicely.

In the mean time I established contact Alexander and I'm waiting to get his impressions and experiences with those drivers. We'll see.

Thanks again,

Florian
  Reply With Quote
Old 25th April 2008, 02:42 AM   #3439
diyAudio Member
 
Join Date: Aug 2006
Default Midbass + mid alternative to single WB

Lynn,

Did you get a chance to listen to (and measure) the 12NDA520 and 6ND410 you received some time ago ?

Any further thoughts on this XO topology ?

Regards,

Florian
  Reply With Quote
Old 25th April 2008, 04:20 AM   #3440
ScottG is offline ScottG  United States
diyAudio Member
 
Join Date: Feb 2003
Location: US
Quote:
Originally posted by caninus80


I was trying single raal 14-150 with altec biflex and raal is too fast. The blend is not seamless
I have to look for another drivers.
Some advice from DIY friends?
Lynn sorry for disturb Your thraed...
And sorry for me english.

Best
C

Try the altec driver inverted and radial, like a "walsh" driver (or DDD bending wave driver).
__________________
perspective is everything
  Reply With Quote

Reply


Hide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off



New To Site? Need Help?

All times are GMT. The time now is 07:00 PM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright 1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2