Beyond the Ariel

[Lynn Olson]

Originally posted by jzagaja
Lynn - I don't understand people who looks for a reproduced sound that involves emotionally. It's a typical audiophile mania to chase the rabbit. For me life is simple - I'm doing my personal recordings of live concerts and looking for close reproduction. For me Jordan drivers do the job perfectly for the money. If a person is not interested in live music then it is a good idea of using headphones to hear how a perfect speaker without resonances should work. Surprisingly most people won't chose this reproductors - they prefer colored sexy sound. Why? because they don't know how the musical instruments sounds in nature!

Have you seen Summa impulse and step?
Have you seen impulses before and after HOM reduction in Geddes OS waveguide?

I can send them private (they are too big for the forum).

Yes, please, very interested in seeing how HOM can be reduced via Geddes' method. I have 5 mbit broadband and a Mac Dual G5 with a 250 GB disk, file size is not an issue.

The Jordan's are interesting drivers, but suffer from very low efficiency and power-handling, and require notch-filter equalization to remove the upper-midband peaks from the aluminum driver. This means they have to be used in a large array if high SPLs at low distortion are desired - and one problem with a large vertical array is non-coincident arrival times. There are two solutions to this, both a bit on the awkward side:

1) Curving the driver array towards the listener. This designs-in a fixed listening distance, and I have reservations about intentionally creating a concave wavefront, something that almost never occurs in nature, where spherical, plane, or transitional waves are the norm. About the only thing I can imagine creating a concave wavefront would be to sit inside a drum, or maybe the focal point of a parabolic reflector. Neither sounds appealing.

2) Using the Quad ESL63 approach of a time-delayed array, using the same principles as a phased-array electronically steered radar. The first sound emits from the innermost driver, and the sound from the outer layers are delayed just enough to synthesize a point-source array. This is do-able, as witnessed by the ESL63, but is extremely complex in the passive or active-equalization domain.

The issues I'm talking about are objective, not subjective. Low efficiency and low headroom are easily measured and easily audible. Similarly, non-coincident arrival is easily measured, and audible to some listeners (but not others). There are side effects that Solutions 1 and 2 create, and the audibility of these probably vary with the listener as well. As true everywhere in audio, TANSTAAFL applies: There Ain't There No Such Thing As A Free Lunch (credit to Heinlein). Improve performance in one area, it goes away somewhere else.

The real task is system optimization for musical perception, which will differ to some degree for every listener. Differences between individual perception can be surprisingly large - I was told about 10 years that Canadian researchers working on audibility of different lossy-compression codecs for cell-phones made the astonishing discovery that lossy-compression audibility depended on the native language of the listener! THAT little finding never made it into the AES literature pushing the latest glamorous lossy-compression scheme. The dirty little secret of psychology is that the research population (that stands in for all of humanity) is in fact composed of graduate students - at the same school as the published research paper.

My overall goal is to design a speaker system with studio-level peak-SPL capability while retaining the spatial qualities of the best omni or dipole systems. I don't know at this time whether this is possible or not, just as I didn't know when I began the Amity amplifier in 1997 that a 1930's textbook all-transformer, all-triode, fully-balanced circuit would be any good or not. (At the time, I got zero support from the Northwest SET crowd, who were all using RC-coupled SET circuits drawn from Sound Practices magazine, and thought that transformer coupling and balanced circuits were for dummies.)

Keeping the two goals of high-SPL and wide-open image quality in mind, my last choice for a driver is one with low efficiency and low peak SPL capability, regardless of how good it sounds on audition. If I was willing to discard the peak-SPL design requirement, I'd just go out and buy a top-class full-range electrostatic and use it with the Karna amplifier, just plug it in, no design work needed.

But unlike many DIYers, I don't design things to save money, cloning this or that well-known speaker or amplifier. That doesn't interest me, because by and large, I don't care for the limitations of commercial products. That's why I designed the Ariel, Amity, Aurora, and Karna. There are no commercial equivalents to these designs.

Now I'm exploring whether I can combine all-horn dynamics with dipole/omni spatial qualities - and without going in the direction of a horn/omni system (a design challenge of a completely different sort). The initial exploration is showing promising results for hemp-composite cones and interesting possibilities for EnABL and Mamboni edge-damping techniques - which by the way would apply to aluminum-cone Jordan drivers as well. There is tremendous room for improvement in all drivers, especially in the time domain.

I'm not interested in "what is". I'm interested in "what can be". Buying high-end hifi doesn't give me much satisfaction - designing it, and hearing new things for the first time, does.

[jzagaja]

How can I send you the files? I've used Nutshellhifii addresses but I'm not sure if they are right now?

Jordan's equalisation is trivial with DSP - see appended picture.

[Lynn Olson]

Originally posted by jzagaja
How can I send you the files? I've used Nutshellhifii addresses but I'm not sure if they are right now?

Jordan's equalisation is trivial with DSP - see appended picture.

As for the Jordan, umm, what if you don't want to pass the signal through an ADC/EQ/DAC conversion step (i.e. analog source, such as LP's and mag-tape). Being the triode nut I am, I am not in the all-electronics-sound-the-same camp - I've gone to some trouble to re-engineer commercial DACs so the analog stage immediately following the DAC has a minimum of a 1000V/uSec slew rate (the 5532/5534 used in many high-end commercial CD, SACD, and DVD-A players is 70 times slower than that). Re-engineering a commercial ADC just to cater to digital equalization is taking speaker design a little too far - I'd rather use drivers that can equalized passively, or better yet, don't need equalization at all.

Still hard to see how the Jordan gets to 115 dB without destruction of the voice coil. The Ariel makes it to 105dB, but I wouldn't dare push it past that.

[ScottG]

Lynn.. for future reference with emails..

the forum has a private mail feature - typically (PM) that person and give them your email account through that (PM).. Usually you look through the members button above to select that person for a (PM).

In any event I'd ask a moderator to remove the last post (or edit it out - if you no longer can), obviously we are all crazy here - but you might get someone sending you emails that is/should be "instituted" .


..and yeah, Sven's adaptation is a thing of beauty (well to an audio nut anyway). I just wish that supravox's frame design was more like PHY.

Its interesting - that driver, incredible decay at higher freq.s until about 4.5kHz:

http://www.supravox.fr/anglais/mesures/mes400EXC4.htm

..most of supravox's drivers though have some serious vc (chamber) resonances that crop up around 900-1000 Hz. Here is Troels's semi solution to that: http://www.troelsgravesen.dk/Supravox215GMF.htm
obviously different driver with different results though. Makes me wonder what BudP's results would be with the "pattern" - and using a higher "reactant to change" lamination boundary via some aluminum gilding leafs.

[jzagaja]

If you like the analogue sources and tube amps (I do also like them) then you will get a lot of additional colourations you must accept. Driver linearisation is then senseless.

I remember I've used strong MOSFET 200W amplifier with popular Dannish speaker with PP+tex cone. The sound was harsh and unpleasant ("quick" amplifier and "slow" drivers). KT88 push-pull typical design changed everything. Second harmonic distortions masked the higher one (if present I'm not sure). It is fine typical audiophile speaker but has no clarity and transparency of Jordans.

So you like perfect square wave DACs without ringing?

I don't know Jordans maximum SPL because I'm not interested - I'm living in student's house and can't play loud. I've found moderate levels ok for me. I remember Ted Jordan made DC tests and they should be able to disperse the heat well (all aluminium inside).

In USA you should be able to find big linear arrays made of Jordans or Bandors. One Arabic have ordered 1,5kW array!

Horn loading for Jordans appended. Phasing plug is a bit tricky for them. I think you can build "Quad ESL57 sound" by using horns as showed by Holland 15 years ago by using axisimetric "short" horn and good compression driver plus a big woofer - typical studio monitor. Nobody will distinguish in blind test...

[DougL]

If you like the analogue sources and tube amps (I do also like them) then you will get a lot of additional colourations you must accept. Driver linearisation is then senseless.

Can we debate tubes vs Solid State another time?

In USA you should be able to find big linear arrays made of Jordans or Bandors. One Arabic have ordered 1,5kW array!

There are trade-offs to Line arrays. There is some magic in good single driver implementations, but single drivers don't do everything well. Pay your money, take your choice.

I would suggest that you check out this page
The Art of Speaker Design

While you are at it, check out .The Amity, Raven, and Aurora

I have been experimenting with OB speakers for a while, and while I am sure I am not sensitive to all the subtleties that Lynn is (I think its a blessing and a curse) I will be very interested in what Lynn comes up with.

Lynn you may get a kick out of this thread. JBL L100 rebuild

Doug

[chrismercurio]

Hemp acoustics has updated the website with pdf's for many of the new drivers. It is exciting stuff!

Chris

[chrisb]

Originally posted by chrismercurio
Hemp acoustics has updated the website with pdf's for many of the new drivers. It is exciting stuff!

Chris

How's the "updated" distribution network and delivery schedule?

It would appear to be easier to design and demonstrate prototypes of "awesome" sounding drivers for the DIY market than it is to actually get any quantity of them in to the hands of builders.

The politics and gamesmanship surrounding this particular maker, of which many are aware and too polite to address publicly, are getting a bit tiresome.

Happy to be proven wrong on this.........

[Lynn Olson]

Many thanks for the Geddes PowerPoint presentation - good to know the licensed version of the Geddes horn is going to manufactured in Thailand in the near future. I'm a little dismayed to hear that people are having trouble with availability of the Hemp Acoustics products - I might end up combining the big Chinese ribbon tweeter with the 12" Alnico Tone Tubby after all - at least it'll be simple to equalize, and will have that wonderful Alnico sound.

If the moderator is reading this thread, PLEASE PLEASE PLEASE remove that e-mail address from my earlier posting. My crude e-mail concealment method won't resist the spambots for long, and the sooner it disappears from the forum, the less spam I'll get.

By the way, part of the reason for posting all this stuff is to get people having fun with dipoles, not just my ideas, but dipoles made with Lowther/AERs, Jordans, Eminence, and whatever driver tickles your fancy. Dipoles being the low-coloration enclosures they, you'll hear your favorite driver in all its glory, both good and bad.

It will be a while until I have access to my audio-design stuff, and can get back into the swing of things again. Rather than stew about my situation and feel sorry for myself (something I've had a lot of practice in since January 7th) I want to share my thoughts about hi-fi. I do have an ulterior motive - I'm kind of hoping people try out these ideas and can give me a bit of feedback when I'm ready to jump back into audio again.

In terms of gizmos I'm researching in the immediate future, I'm going to be mentioning this and that bone-regrowth devices to my surgeon on May 1st. These aren't made by New Age electronic-gizmo companies - they also make this and that charming implements that are in my leg right now. So my attention isn't 100% on hifi right now.

(Warning to the easily disturbed - don't click those last two links while you're eating dinner. I got the stainless-steel versions and a $45,000 hospital bill, fortunately paid-for by Kaiser Permanente. Next year I'm paying someone else to shovel the snow.)

[fiacono]

My first post here…

Good to hear that your might be considering going back to the TT.
I’ve been playing around with the TT Alnico's for over the last 6 months (OB) crossed with NeoPro5i ribbon. The problem I have is physically off setting the ribbon back from the baffle board. I have a few crazy ideas in my head but can’t progress any further (broken hip from a cycling accident, so yes I also have lots of brackets and screws holding my hip together).

Currently I’m using two TT per side (as you mentioned a while ago), one as a bass helper. I’m still waiting on the 15” TT bass driver specs.

Frank.