Ring Radiator Comparisons

[ivica]

I found that the 9kHz hole is probably a product of the ring radiator design and it shows in the bullets (the worst), butt cheeks (about as bad as the bullets with 2402 phragms but better with 2405 phragms, and slots (the best).

Even JBL recognizes this and the crossovers are designed with this dip in mind to augment this area. That's why they have the 1 mfd/0.18 mH choke/1.5 mfd. The combination of the two caps creates a peak in this area.

Yes, your words about JBL crossover networks really explain the reason for presence of some overshoot in UHF network section (alone), and some time this 'overshoot' has to be lowered by putting the resistor 'in parallel' with the choke (N3145), especially if planed other ring radiator to be used instead of 240x (such as Beyma CP21F for example, or some "good" sample of 2405)

[maxwedge]

Need a few more stickies here.

[Steve Schell]

Here's an interesting aside to this essential old thread. My Cogent speaker partner Rich and I have recently made what we think are significant improvements to our field coil compression drivers, and are considering jump-starting our company Cogent Loudspeakers again after several years of minimal activity. We are keeping mum about the nature of these improvements, at least until we get a little further along. I will hint that they are due to the little-known contributions of a genius engineer from RCA's past. The newly modified drivers have brought new life to our systems, and we are once again stoked and in hot pursuit of the most accurate sound possible. One side effect of our driver mods is reduced top end in our DS-1428 mid/high frequency driver, so I tried adding a pair of JBL 075 tweeters crossed in with simple 2 mfd. oil caps to cover the top couple of octaves. We have actually been experimenting with various horn tweeters in our systems in recent months, with some degree of success, after years of attempting to extract adequate top octave performance from our 1428 mid driver.

Anyway, I have come away with new found respect for the 075! As a way to help evaluate our driver progress, I fetched an oldie but goodie 1970s Tama drum set from the closet and set it up in my living room, right in front of the big folded 15Hz. flare horn sub. I set up my usual recording rig in the position of my listening chair and made some recordings while flailing away on the drum kit ( I am a former mediocre drummer from elementary and first year high school). The recording rig is a pair of English Reslo ribbon mics in a classic Blumlein array, feeding a line stage hi fi tube preamp pressed into service as a mic pre with gain wide open, feeding an Alesis Masterlink hard disc recorder set at 24 bit, 96 kHz. On the recordings I speak as well as bang on the drums and cymbals, and the spoken word is a good way of establishing the correct in-room playback level. Jeez, talk about source material, the drum kit is the sh**, both in terms of frequency and dynamic range.

The fun part of this experiment is that the playback sounds remarkably similar to the live sound! The 075s play an important part in this as they handle the 5kHz. and up range remarkably well. Big snare, tom tom, kickdrum and cymbal hits sound very much (I'd say 90% +) like the real thing in a direct comparison to live drumkit debauchery. I had thought that we were still miles away from verbatim reproduction (as JBL used to call it), but now I think that we may occasionally come closer than that. The 075s hold up their end of the frequency range extremely well, and are the only tweeters on hand that keep up with the Cogent drivers in sensitivity. I'm really impressed by them at the moment, and will have to study this thread and others here to come up to speed on these fine JBL ring radiator drivers that I really don't know much about.

Any LH dudes or dudettes in so. CA that would like to have a listen to my "live vs. reproduced" demo are welcome to contact me at (562)421-5145.

[Predrag Dukic]

Could You say something about DS-1428.. I am interested in TS parameters: Mmd, BL Le etc... Could You reveal at least that?

[ivica]

Here's an interesting aside to this essential old thread. My Cogent speaker partner Rich and I have recently made what we think are significant improvements to our field coil compression drivers, and are considering jump-starting our company Cogent Loudspeakers again after several years of minimal activity. We are keeping mum about the nature of these improvements, at least until we get a little further along. I will hint that they are due to the little-known contributions of a genius engineer from RCA's past. The newly modified drivers have brought new life to our systems, and we are once again stoked and in hot pursuit of the most accurate sound possible. One side effect of our driver mods is reduced top end in our DS-1428 mid/high frequency driver, so I tried adding a pair of JBL 075 tweeters crossed in with simple 2 mfd. oil caps to cover the top couple of octaves. We have actually been experimenting with various horn tweeters in our systems in recent months, with some degree of success, after years of attempting to extract adequate top octave performance from our 1428 mid driver.......

Does your improvements being applicable to the 'ordinary' -"DOME-type Diaphragm" (ferrite, alnico, Neodymium) permanent magnet (such as JBL 244x~245x) drivers ?

http://www.oswaldsmillaudio.com/steve.html


Regards
Ivica

[pos]

Here we have heard the story about the lady working in JBL who had "repaired" rejected diaphragms for 2405/077 with "blue ball-pointing pen"....so it seems that even small change in the membrane shape has great influence on the driver response.
May be only myth
http://www.audioheritage.org/vbullet...l=1#post187592
http://www.audioheritage.org/vbullet...l=1#post188524

Some part of the pen story is true. I have several D16R2405 diaphragms that have the "rolled rings" in them with the ink color still in them. The proof that it was done during manufacture is evidenced by the fact that the radius and ink is visible under the voice coil lead-ins. I will try to get a good picture of some of them.

Did the measure better? I can't remember.

This was confirmed by David Smith here: http://www.diyaudio.com/forums/multi...ml#post3710662

Those curves are from a really good one. The ring radiators vary all over the place. The 3 peaks that make up the response can be quite nasty. They did get somewhat better when they trained a lady to go around the perimeter with a ball point pen to crease the edge just inside the suspension point. (I kid you not. You can usually see the blue ink line.)

[4313B]

Yeah, G.T. once said that the various model numbers didn't matter so much as finding a matched pair. In other words, if you had to run an 075 on the left channel and an 077 on the right channel to end up with a matched pair then so be it. I love his sense of humor.



Very cool that David is so active on that forum. Great guy!

How does he feel about the whole DSP thing? I noticed that he mentioned it in that post. I consider the M2 a great advancement of his 4430/4435 design.

[pos]

He is not that much into constant directivity these days it seems. He preaches for smooth power response, but no necessarily flat. I think he tends to prefer slowly increasing directivity with frequency, similar to what multiway direct radiator speakers do.

Here is what he has to say about the M2 waveguide, compared to the 2344:

My sense is that the JBL designers today are much less obsessed with polar curves and achieve a reasonable overall compromise in their horn designs. Less good in the polars but better in frequency response and with a flaster flare to hold the distortion down.

Probably the smartest approach.

Their current trend is to forego the diffraction slot and simply do a blend of fairly straight side walls and end flaring. As mentioned a few posts back, this will give less perfect polars but better frequency response. They are quite far from classic exponential so I would like to see some raw horn curves to see what the low end looks like.

[Odd]

Yeah, G.T. once said that the various model numbers didn't matter so much as finding a matched pair. In other words, if you had to run an 075 on the left channel and an 077 on the right channel to end up with a matched pair then so be it. I love his sense of humor.



Very cool that David is so active on that forum. Great guy!

How does he feel about the whole DSP thing? I noticed that he mentioned it in that post. I consider the M2 a great advancement of his 4430/4435 design.

I bought 6 units 2405 and found two that matched well!

[4313B]

I bought 6 units 2405 and found two that matched well!

Excellent!

[grumpy]

Excellent!

... if you're a "the glass is 1/3 -full-" kind of person

[toddalin]

When I last obtained three 2405s (for L/C/R), two had consecutive serial numbers and the other was maybe a few hundred away. Logic would put the two consecutive numbers in the L/R and the other in the C. But the one that was a few hundred away was much closer in response to one of the consecutive numbered one, and the other consecutive numbered unit actually had the best response (smoothest and most extended). So the one with the best response (but most different) became the center channel.

[Dr.db]

BTW; how do 2405 and 2403 compare in terms of efficiency

2405 has 105db/1w, but how about the cat-eye...???

[ratitifb]

BTW; how do 2405 and 2403 compare in terms of efficiency

2405 has 105db/1w, but how about the cat-eye...???

seems to be specified at the same value @1m

from http://www.audio-heritage.jp/JBL/unit/2403h.html

2405H FR


2403H FR

[pos]

Here are my own measurements, including 30° off axis horizontal and vertical (this is where the 2403 shines).
It is quite similar to the measurements published by ratitifb (where did you find those?), especially when you consider unit to unit differences (even when new). According to these measurements, -10dB would be around 110dB/1w/1m...

2405H:


2403: