Beyond the Ariel

[Pano]

Originally posted by badman
I sure hope his is a king of crossovers- because going from a 12" to a dome will be a royal pain in the you know what!

Don't forget the Beta LTA has a whizzer cone.
And he's been mating that drver to tweeters for years. That's not to say it will be easy. I sure don't want to do it!

I'll let you know after RMAF.

[Salas]

Why they don't take the back cap out of a compression driver and expose the dome at the back so to have a dipole HF like Bastanis does, I don't know. Seems perfect for OBs.

[JohnL]

Originally posted by Lynn Olson
Here's the latest thoughts about the open baffle - will construct the baffle, and measure and audition drivers after the RMAF.

Lynn, do you see any issues with having the tweeter up this high? Sitting on top of 2 - 15s and a 12 puts things at least 45" to the top of the mid/bass module. From my seating position I try not to put my tweeters up past about 30-32" or else funny things happen to the soundstage. That's why I was initally wondering about a solution with an 18 or 21 mated to the 12 mid. 2 -15s and a 12 is one thing if your 12 is a co-ax, but putting a tweeter on top of that seems a bit high.

Please accept this all as a simple question, my expertise is nowhere near yours (or T. Danley, E. Geddes, or John K.'s and probably a few others). I am mearly basing this on some observations of my own system and some others I have listened to.

[mige0]

Hi

All I'm asking for is better, and more useful, published specs. I've been pestering driver manufacturers for any kind of time information for more than 30 years now, and most of the time, all I get is a runaround and lot of hand-waving and excuses why it's "not possible". I usually have to surreptitiously find out who the engineer is and wheedle the data out of him..

Worse! Even from really big and well regarded PRO manufacturers you can get the answer

"CSD? I am afraid I do not know what it is. Sorry."

when asking for a CSD / waterfall plot – as happened to me recently.

-----------------------




Impulse response is a good thing to look at but even more intuitive is CSD IMO. No way to equalise any resonance that shows up. And usually there are several MORE than the prominent one at cone break up – quite often in the most sensitive frequency area.

Some divers that find the way to DIY magazines are measured more extensively. Plots of harmonics at 90dB up to 5-7 order and a plot of harmonics against SPL are very revealing and easy and intuitively to compare.


--------------------





Nice draft on post 1923, Lynn.

I know you love the mirrored bass placed close to the floor and though it is against all physics I would suggest you to lift the OB a rough inch and listen again.

Bass becomes much more " open " and " swingin' " this way.


--------------------------------

I am eying a pair of Audax 6.5" PR 17HR70 midrange drivers, capable of 99 dB sensitivity, I just drug out of my stash, from back in 92 I believe, maybe even earlier than that. Mille N said they were the best midrange drivers he had heard, period.

BudP, There was a PRO version of that driver with the beautiful name:

" Audax PRD 17 HR 37 TSM KAC "






I happened to find it right at the same time than yours in my loft!

Haven't measured nor listened to it ( sadly don't have a second one.), but telling from ancient plots, FR is even smother throughout and slightly more well behaved at the roll off region.

Surprisingly the flat foam surround is still like new . Compare that to old Dynaudio foam surrounds !



Greetings
Michael

[Lynn Olson]

Originally posted by JohnL


Lynn, do you see any issues with having the tweeter up this high? Sitting on top of 2 - 15s and a 12 puts things at least 45" to the top of the mid/bass module. From my seating position I try not to put my tweeters up past about 30-32" or else funny things happen to the soundstage. That's why I was initally wondering about a solution with an 18 or 21 mated to the 12 mid. 2 -15s and a 12 is one thing if your 12 is a co-ax, but putting a tweeter on top of that seems a bit high.

Please accept this all as a simple question, my expertise is nowhere near yours (or T. Danley, E. Geddes, or John K.'s and probably a few others). I am mearly basing this on some observations of my own system and some others I have listened to.

Well, I like to have the tweeter/midrange crossover region at or above listening level, but never below. In essence, the tweeter/midrange are on or slightly above the horizon line, if there are any windows behind the speakers (the arrangement I had in Portland).

This corresponds to real-life experience in a concert hall or jazz club - we sit down, the musicians stand and do their thing. That to me is a real-life perspective, with the sound coming from close to the floor to a level just above listening height, and bouncing all around a lively-sounding room.

If a genuine real-life perspective sounds "wrong" on a loudspeaker, something is very wrong with the speaker - most probably, vertical polar patterns aimed in the wrong direction. This is actually pretty common with big audiophile speakers. I've heard quite a few expensive speakers that didn't sound coherent unless you laid down on the floor - and sounded utterly unnatural and disjointed when standing. If the "sounds good" zone is 30~32" off the floor, that's a pretty weird listening position - all I can think of is lying down on the couch half-asleep or lying on a bed (and maybe doing other things).

I sit pretty high - the listening height for me is about 40", unless I slouch way down, or sit in a chair that is impossible to get out of since it is so low.

Listening from a height of 40" or more, 30~32" tweeter height sounds squashed and dwarf-sized, and severely compresses the soundstage, giving a bizarre "tabletop" miniaturized sound, like a stereo table radio. Maybe some people like this, to me it sounds extremely unnatural.

Maybe you sit at 30~32" listening height - dunno about your physical size and household furniture - but to me, that tweeter height is very compressed-sounding. I like speakers that sound big and spacious - even the Ariels sound pretty large, far bigger than their physical size would indicate.

[AJinFLA]

Originally posted by Tom Danley
Hi Lynn
Its been years since I sampled any hifi drivers while I look through Madisound frequently. Frankly, it would appear that many drivers are designed for appearance and technical appeal as a well as function.
I do sample a lot of pro sound drivers however and since what I do is try to design speakers with hopes of the hifi experience for a crowd, have some thoughts which might be applicable here.

Hi Tom,

Wouldn't it be a bit more objective to subject the hifi drivers to similar testing as you do for the pro drivers, than cast judgement leafing through a catalog? I don't totally disagree with you about looks being important for hifi drivers, as they are aimed at audiophiles, who hear with their eyes as much (if not more?) than their ears. But with no data, your statement seems rather subjective. Plus sound quality that is blaring loud enough for public address won't always cut it within the confines of a living room. Different audience requirements.

Originally posted by Tom Danley
Although to be honest, I have never built an open baffle speaker before other than electrostatic panels.

I, for one, would be interested in what you could come up with. For domestic acoustic spaces, of course. Not open space public address.

Originally posted by Tom Danley
A bad sign (regardless of cone material) for a direct radiator is a large peak(s) associated with its hf roll off. It is assumed that a steep or low crossover fixes this problem but it is not gone. So, picture a driver with a big mound in its response associated with breakup.
Lets say that mound is 10 dB high above the “flat” response zone, this represents an frequency dependent acoustic gain element AFTER the voice coil.
So, here is where it is not fixed even with a brickwall crossover.
The driver’s motor produces harmonic distortion, that acoustic gain amplifies it by 10X (10dB) for harmonics when the fundamental is N fraction of the peak.
In other words, if the mound were at 1200Hz, the third harmonic of a fundamental at 400Hz is amplified by the acoustic gain even if one had a 120,000dB / octave digimatic crossover at 500Hz. Here it is better to fix the source.
Exotic materials are more commonly known for “those” kinds of peaks because the focus is usually on strength to weight and not damping.
The best radiator material is the best trade off of all the requirements and it is still very hard to beat paper processed fibers. You inclination to look there is I think a good one.

What about the low Q resonances within the soft cone drivers passband? What about the flexing and bending of the cone? None pistonic operation? Especially the ultra low mass ones required for HE (pro drivers) attached to powerful motors?
The cone breaking up inside it's passband? A lossy medium is best for signal preservation? Or hiding things?
What if the goal is reproduction of acoustic instruments, not paper cone-compression driver-amplified box speaker-live sound rock concert? Tom, isn't it best to give both sides of the story? Soft, flexy paper cones are without sin?
Is Dr. Toole's research (CMMD) just a marketing gimmick? Thiel?
Last time I checked, there were no paper processed fiber compression drivers. Has that changed?

Originally posted by john k...


I couldn't agree with you more. This is one rerason I particualrly don't like metal cones. The SEAS Excel metal cones, for example, have very smooth and very linear frequency response up to the point where the breakup comes in. And to me the audibility of the effect of the breakup, distortion or otherwise, are clearly audible with the crossover in place. Door bells and tuning forks should ring, not driver cones.

This is the original NaO correct? As you use to sell it after extensive design, testing and listening? With a clearly audible breakup? Was this audible breakup per customer request? Did they have to pay extra?



cheers,

AJ

BTW, Tom, what ever happened to the impulse data for the SH100?

[john k...]

Hi AJ,

No, this is the original NaO:



The original NaO used Scan Speak 8545 mid, as does the current NaO II. Sure the SS have their problem but not like the W18. The picture you are showing is of the NaO AEP (All Excel Panel) which was a short lived (for the reasons I mensioned), one off design variant I made at the time the W18 was very popular. I tried one other variant with the W18 and a sealed box woofer (below) which was enthusiastic received at a DIY event but I quickly grew tired of it too. let me say it again, I don't like metal cones or domes. (Maybe a metal cone woofer? I don't know about that as I haven't looked an any due to my bias.)



We all make mistakes. I certainly not immune.

The original NaO is still supported and has evolved into the current NaO II.

[john k...]

Originally posted by mige0


... No way to equalise any resonance that shows up. And usually there are several MORE than the prominent one at cone break up – quite often in the most sensitive frequency area.

I have to disagree with you here. The resonance in most drivers are a linear phenomenon and as such can be equalized. The problem is as Tom said, that while they can be equalized to response correctly to input signals that doesn't solve the problem of them being excited by distortion components generated as harmonis of lower input frequencies. For example, here is a CSD of a naked Seas W18



Now, here is a CSD of a 2nd order band pass active filter;



Lastly, here is a CSD of the W18 equalized to have the same 2nd order band pass response;



It isn't perfect but clearly the resonance at 500, 800, and 5k Hz are not present. The point is that linear systems can always be corrected by response shaping. It is the nonlinearities that cause the problems.

Now, some one asked aboy Floyd Tool. All I can say about that is, for example, a 1st order band pass response has a transfer function given by

TF = (s/Q)/ (1 + S/Q + s^2).

[Variac]

Hey AJ,

Don't you think it is better to use a less accusatory tone until all the facts are known?

Variac

[jholtz]

Originally posted by AJinFLA



This is the original NaO correct? As you use to sell it after extensive design, testing and listening? With a clearly audible breakup? Was this audible breakup per customer request? Did they have to pay extra?



cheers,

AJ

Hi AJ,

I believe that speaker pictured belongs to John Pastuck and was present at the 2004 Iowa DIY Event. It and Salk Sound HT3's were a few tenths apart in the voting for best sound in the unlimited class. I did not hear the ringing John speaks of but I like metal drivers and don't care for the 8545's. They are very nice speakers.

Jim