Yes, the Emerald is now very similar to Lynn's thoughts it seems, but using a 1" instead of a 1.4 or 2" CD driver and 15" drivers. After our research I'm not totally surprised by their using 15" drivers. IT appears that some of them have response as high as 12" drivers...
Lynn,
here is the Solen price list. Is the driver you are discussing on here?
http://www.solen.ca/pdf/raacad.pdf
Possibly the 70-20 XR?
I propose that you also, in the future, consider a third model with a BMS or similar 1" CD crossed over about the same place as the RAAL driver. It might sound very good, and would cost a lot less than the first options, as the proposed 12" drivers- either 18 Sound or Hemp of some sort aren't extremely expensive, and I suspect that the bass drivers won't be either..
Lynn,
here is the Solen price list. Is the driver you are discussing on here?
http://www.solen.ca/pdf/raacad.pdf
Possibly the 70-20 XR?
I propose that you also, in the future, consider a third model with a BMS or similar 1" CD crossed over about the same place as the RAAL driver. It might sound very good, and would cost a lot less than the first options, as the proposed 12" drivers- either 18 Sound or Hemp of some sort aren't extremely expensive, and I suspect that the bass drivers won't be either..
Variac said:
Interesting 50:1 price ratio between the Steinway and the Emerald - but I'm sure both sound good, underlying design principles seem quite valid to me, and it all probably comes down to voicing and driver selection. Great minds think alike!
The 745PB 1.4" Radian is $279 at USSpeakers, the BMS 4552ND is $149, and BMS 4540ND is $99 at Assistance Audio. The 4552ND and 4540ND 1" ring-radiator plastic-film CD's appear to be identical except the 4540 has a lighter single-layer voice coil, thus the more extended HF response and somewhat lower power ratings. The Fs appears to be the same, as far as I can tell.
Of all the Radians, close inspection of the Radian curves shows the 745PB 1.4" aluminum CD with the smoothest response. All three drivers look good for a crossover between 1.2 and 2 kHz, and have excellent specs by any standard. My instinct is they could come closer to the $2000 TAD beryllium CD than you might expect.
The AD-550 Azurahorn is the most luxurious choice, along with whatever current model "salad bowl" Tractrix made by Edgarhorn. There also seem to be several inexpensive plastic horn 12" mouth with modern oblate-spheroid profiles - the other OB thread had some detailed measurements of the BMS CD and one of these horns.
I'd say there are quite a few good-sounding options here, and not necessarily in the RAAL price range.
Thanks for the referral to ScottG's offering.
I'm frustrated in my search for a workable description of the LeCleac'h profile. If someone would point me in the right direction I'd be much obliged.
Maybe the better question is: will it be used here?
Answer my own question:
http://ndaviden.club.fr/pavillon/lecleah.htm
click on: "Telechargement" http://ndaviden.club.fr/outils/axial.zip
I'm frustrated in my search for a workable description of the LeCleac'h profile. If someone would point me in the right direction I'd be much obliged.
Maybe the better question is: will it be used here?
Answer my own question:
http://ndaviden.club.fr/pavillon/lecleah.htm
click on: "Telechargement" http://ndaviden.club.fr/outils/axial.zip
Going back to the issue of baffle shape. I'm not of the "I don't care how ugly it looks, just listen" persuasion. If it goes in my living room it has to be aestheticaly acceptable, I'm thinking SF Stradivari is a good thing here. It would be a challenge to do that with the aircraft rudders IMHO, I won't be going down that route. I'm sorry Lynn!
Your idea of acoustically softening the edge seems to me an excellent thing to aim for. I've spent a little time with Edge trying to defeat those aberations in the response inevitably caused by the finite size of the baffle.
It seems to me that hard edges equals significant lumps somewhere no matter what you do. Earlier on in the thread you mentioned acoustically "softening" the edge with hole patterns to reduce ripple,
Are my DIY skills up to drilling six hundred 1/4" holes around the edges? with no mistakes?....moving swiftly on.
It ocurred to me that ignoring what happens in the less critical vertical plane, using vertcal slots might actually be practical. Unfortunately it looks like Edge won't go round corners so I can't simulate it, any thoughts?
Your idea of acoustically softening the edge seems to me an excellent thing to aim for. I've spent a little time with Edge trying to defeat those aberations in the response inevitably caused by the finite size of the baffle.
It seems to me that hard edges equals significant lumps somewhere no matter what you do. Earlier on in the thread you mentioned acoustically "softening" the edge with hole patterns to reduce ripple,
Are my DIY skills up to drilling six hundred 1/4" holes around the edges? with no mistakes?....moving swiftly on.
It ocurred to me that ignoring what happens in the less critical vertical plane, using vertcal slots might actually be practical. Unfortunately it looks like Edge won't go round corners so I can't simulate it, any thoughts?
Attachments
Simon,
Come over to my thread on EnABL and Mamboni and learn how to minimize those edge reflections and diffractions to the point that you can not hear them, at all. I turn hard edged speaker boxes sonically invisible on a regular basis. Plus treating those HiV speakers, with their irregular responses is a specialty. We just ran through a Lowther Dx 4 treatment sequence and I can assure you the results are sensational.
This will free you to shape the baffle to your personal delight, although moving too far from correct distances can get you into trouble, even with this amount of aid. But, the trouble will be between drivers, not drivers and the baffle edges.
Bud
Come over to my thread on EnABL and Mamboni and learn how to minimize those edge reflections and diffractions to the point that you can not hear them, at all. I turn hard edged speaker boxes sonically invisible on a regular basis. Plus treating those HiV speakers, with their irregular responses is a specialty. We just ran through a Lowther Dx 4 treatment sequence and I can assure you the results are sensational.
This will free you to shape the baffle to your personal delight, although moving too far from correct distances can get you into trouble, even with this amount of aid. But, the trouble will be between drivers, not drivers and the baffle edges.
Bud
a couple of quick questions if I may, firstly being a complete dunce when it comes to the intricacies under discussion here, can s/one simply explain the difference between 'normal' drivers and compression drivers?? and why they would be preferred.
Regarding the 'worth' of speakers that fail when played at power volumes (that could be a poor choice of word BTW), how much of that phenomena could we attribute to the lowered sensibility of the ear to low and high frequencies at lower volume?? I use a deqx in my system, and have found good results when listening at lower volumes to gently tweak (lift) the bottom end. That of course could be cheating in this context, but it did get me wondering.
Regarding the 'worth' of speakers that fail when played at power volumes (that could be a poor choice of word BTW), how much of that phenomena could we attribute to the lowered sensibility of the ear to low and high frequencies at lower volume?? I use a deqx in my system, and have found good results when listening at lower volumes to gently tweak (lift) the bottom end. That of course could be cheating in this context, but it did get me wondering.
hyper-vivid tonality
Hi
I experienced the same as well.
At a listening distance of 15 m a step of 40 – 50 cm to the side was the difference between locked or unlocked to the sound. This translates to a deviation of around 2 degrees only.
There are two mechanisms that contribute differently IMO.
- First the reduced reflected sound due to the high directivity of horns which allows us to discriminate much more precisely to any phase differences (around +/- 12 cm difference in distance to the speakers in the example above). Omni's don't "suffer" from that in comparison.
- And second the nonlinearity of air under pressure.
A speaker that makes use of this specific nonlinear behaviour of air can be seen here:
http://www.sennheiser.com/sennheiser/icm_eng.nsf/root/09859
"http://www.sennheiser.com/sennheiser/icm_eng.nsf/root/09859"
Basically the speed of sound (= mach number) depends on the pressure of the air.
When the membrane moves forward the air gets compressed and every overlaid frequencies during this outgoing movement will be modulated differently from when the membrane goes backwards. I am NOT referring to the Doppler effect, its quite a different mechanism that moulds the sound here!
(By the way, this also might be a reason why absolute phase matters and easily can be heard when trained on that a little.)
With CD's and horns, compressing the air stronger than any other speaker, this more or less MUST result in kind of sound coloration like listening to a loud source even at moderate listening levels = sense of enhanced dynamics and bringing you subjectively closer to the acoustic event.
I'd really like to try this on my test baffle! Can you give me some further advice?
Greetings
Michael
Hi
I experienced the same as well.
At a listening distance of 15 m a step of 40 – 50 cm to the side was the difference between locked or unlocked to the sound. This translates to a deviation of around 2 degrees only.
There are two mechanisms that contribute differently IMO.
- First the reduced reflected sound due to the high directivity of horns which allows us to discriminate much more precisely to any phase differences (around +/- 12 cm difference in distance to the speakers in the example above). Omni's don't "suffer" from that in comparison.
- And second the nonlinearity of air under pressure.
A speaker that makes use of this specific nonlinear behaviour of air can be seen here:
An externally hosted image should be here but it was not working when we last tested it.
http://www.sennheiser.com/sennheiser/icm_eng.nsf/root/09859
"http://www.sennheiser.com/sennheiser/icm_eng.nsf/root/09859"
Basically the speed of sound (= mach number) depends on the pressure of the air.
When the membrane moves forward the air gets compressed and every overlaid frequencies during this outgoing movement will be modulated differently from when the membrane goes backwards. I am NOT referring to the Doppler effect, its quite a different mechanism that moulds the sound here!
(By the way, this also might be a reason why absolute phase matters and easily can be heard when trained on that a little.)
With CD's and horns, compressing the air stronger than any other speaker, this more or less MUST result in kind of sound coloration like listening to a loud source even at moderate listening levels = sense of enhanced dynamics and bringing you subjectively closer to the acoustic event.
I'd really like to try this on my test baffle! Can you give me some further advice?
Greetings
Michael
Hi Michael,
Were I doing it I would apply EnABL to everything on the front surface, drivers and all. Mamboni on the back of the big drivers, 8 inch and up and EnABL to the smaller stuff on both sides. Then I would Mamboni the back edges of the baffle and EnABL the front edges. I guarantee you that the entire speaker will just disappear.
As for sizes and spacings, I think you should bring those questions with to the EnABL Mamboni site and we will see if Dr Mamboni can be coaxed into commenting, while I jabber a mile a minute.
Then we can do the simple math and discuss what you need to have on hand to do this madness. That way we keep this thread on topic, but, I think all of you are getting close enough to actual hardware that we should address with what and how you can accomplish your goals. Besides, you can do it all for under $30, speakers and baffles. And, the EnABL and probably Mamboni treatments can be removed with alcohol.
Come visit and we will all learn from each other. Besides, Lynn approves, so how can you go wrong?
Don't miss the Lowther treatment sections, they may apply to your drivers quite nicely.
Bud
Were I doing it I would apply EnABL to everything on the front surface, drivers and all. Mamboni on the back of the big drivers, 8 inch and up and EnABL to the smaller stuff on both sides. Then I would Mamboni the back edges of the baffle and EnABL the front edges. I guarantee you that the entire speaker will just disappear.
As for sizes and spacings, I think you should bring those questions with to the EnABL Mamboni site and we will see if Dr Mamboni can be coaxed into commenting, while I jabber a mile a minute.
Then we can do the simple math and discuss what you need to have on hand to do this madness. That way we keep this thread on topic, but, I think all of you are getting close enough to actual hardware that we should address with what and how you can accomplish your goals. Besides, you can do it all for under $30, speakers and baffles. And, the EnABL and probably Mamboni treatments can be removed with alcohol.
Come visit and we will all learn from each other. Besides, Lynn approves, so how can you go wrong?
Don't miss the Lowther treatment sections, they may apply to your drivers quite nicely.
Bud
Hi
No surprise as HSS (hyper sonic sound) device were developed as weapons first.
I don't recomend them either, just wanted to put clear my point about the nonlinearity of air.
You mean this thread I guess?
http://www.diyaudio.com/forums/showthread.php?s=&threadid=100399&perpage=10&pagenumber=22
"http://www.diyaudio.com/forums/showthread.php?s=&threadid=100399&perpage=10&pagenumber=22 "
I followed to around post 80 earlier and have to read to the end first.
Greetings
Michael
No surprise as HSS (hyper sonic sound) device were developed as weapons first.
I don't recomend them either, just wanted to put clear my point about the nonlinearity of air.
You mean this thread I guess?
http://www.diyaudio.com/forums/showthread.php?s=&threadid=100399&perpage=10&pagenumber=22
"http://www.diyaudio.com/forums/showthread.php?s=&threadid=100399&perpage=10&pagenumber=22 "
I followed to around post 80 earlier and have to read to the end first.
Greetings
Michael
I did not understand what good power response can do for more realism until I used Manger drivers in various rooms. Manger does not really have linear power response and the timbre and dynamics can be excellent, natural. But it depends much on the room.
Some say that directional speakers excite the room a less. Some say that omnidirectional speakers excite the room uniformly. Some say that constant but narrower directivity or dipole octal radiation is ideal. Common box speakers are some kind of compromise. Manger is different compromise. Still it works great. So I agree with Lynn that directivity is less important. At most it is useful to match directivity of particular speaker to particular room. Nothing more.
Some say that directional speakers excite the room a less. Some say that omnidirectional speakers excite the room uniformly. Some say that constant but narrower directivity or dipole octal radiation is ideal. Common box speakers are some kind of compromise. Manger is different compromise. Still it works great. So I agree with Lynn that directivity is less important. At most it is useful to match directivity of particular speaker to particular room. Nothing more.
As always, thanks for the response Lynn. FWIW, I have no strong dogma's on this subject, I simply lack the experience to firm any strong opinions. I was just under the impression that one needed to pay attention to the polar response in the xo region in order that driver integration become seemless.
I like many of your goals and I think I too have some of your leanings.
I like many of your goals and I think I too have some of your leanings.
Hi Lynn,
Firstly, thanks for kicking off such a fantastic thread!
I was thinking why not go for dipole, ESL horns... perfect loading for the diaphragm on each side, dipole radiation, horn reflections not a problem. Would also work beautifully with direct drive from your PP amps...
Best,
Chris
Firstly, thanks for kicking off such a fantastic thread!
I was thinking why not go for dipole, ESL horns... perfect loading for the diaphragm on each side, dipole radiation, horn reflections not a problem. Would also work beautifully with direct drive from your PP amps...
Best,
Chris
BudP said:
I will take you up on your invitation Bud, but I dont think EnABL is going to tackle the step caused by the back wave interfering with the front wave- this is fundamental to the operating range of an open baffle.
My thinking was to acoustically fudge the edge of the baffle so the front/back interfence is spread over a wider range and thus has a less strong acoustic signature.
Simon
simon dart said:
I suggested a few posts back (OK, maybe several hundred) that if you access to a NC drilling machine, drilling a series of random-sized holes close to the edge, with the hole size varying from 12 to 1mm, and the largest holes close to the baffle edge. I think I gave it a corny name like Progressive Loss Mesh, but the concept is simple: a random structure with progressively increasing loss, and all of the holes less than the minimum wavelength of audio frequencies.
The same principle is used for vented speakers with intentionally lossy vents - instead of a conventional duct, the vent is filled with closely-packed drinking straws, all of which have diameters much smaller than audio wavelengths, and act as acoustic resistors, not a duct.
Yes, if you were crazy enough, you could trim a zillion drinking straws to the depth of the Baltic Birch ply of the baffle, and pack them into a vertical slot, so they'd form an acoustic resistor between front and back. That would look truly strange, and I suspect would not work as well as the random array of holes.
In terms of the visibility of the holes, there have been some very attractive OB's using veneered wood only for the driver mounting area, and perspex/acrylic for the outer wings. (Some people have even used glass for the outer wings, although that is very heavy. I imagine the commercial versions of OB's use tempered glass to address safety concerns.)
I'd vote for perspex outer wings, with the computer-drilled hole pattern on the outer edge. It would have considerable visual interest, and would sound good as well. If you want a Museum of Modern Art look, you can trans-illuminate the perspex with an array of white and/or multicolor LEDs at the base, which would light up the outer edges, and make the perspex seem to glow from within. Now that would make a visual statement about OB speakers!
JoshK said:
Actually, driver integration is mostly a matter of precise phase transitions between drivers in the crossover region. This is commonly overlooked, particularly in minimalist-crossover high-dollar audiophile systems, in the interest of low parts count in the crossover. It also gets overlooked in active-crossover systems using out-of-the-box LR4 slopes.
Ripples in the amplitude response of the driver create narrowband phase variations, and the phase angle between the drivers steers the vertical polar pattern. It is the narrowband phase variations that create a feeling of discontinuity between the drivers - it only sounds like one virtual driver filling the space between the two when the phase angle is well under 20~30 degrees.
I aim for less than 10 degrees of phase spread, which can be seen by intentionally phase-reversing the tweeter and looking for a cancellation null - which won't appear unless you have good phase tracking between drivers.
Returning to problems created by polar-pattern directivity variations, when the direct-arrival and total room spectra depart from each other (especially if there are deep nulls in the room spectra) the ambient impression is degraded, and the system can be hard to adjust for subjective flatness.
Note that I used the phrase "total room spectra", not polar pattern. This means the total power radiated into a sphere, and summed by all the reflections within the room. In terms of ambient impression, it's the reflections that fall within 5~50 milliseconds that are important, and that amounts to several hundred reflections from all directions.
Note also the phrase "subjective flatness", which is not the same as measured flatness. The whole point of subjective fine-tuning isn't to make the speaker sound great on Patricia Barber or Jazz at the Pawnshop (although that's what many audiophile mfg's do), but to get the speaker subjectively flat.
That's why I use primarily use pink-noise for assessments of flatness and coloration, although it has to be kept in mind that holes in the response are not audible with pink-noise. This is where music can be used as a secondary reference, noting the dynamic and spatial impression, where pink-noise tells you little. So it takes a round-robin cycle of measurement, audition with pink-noise, audition with music, making small adjustments, and repeating the cycle as necessary. This part takes me about six months of tedium, which is the pretty much the industry average to "dial-in" a speaker. If it takes longer than that, the drivers or the basic design concept may have problems.
I've found that impression of timbre is dominated by driver coloration, with polar-pattern variation coming in a distant second. This is easily confirmed by direct audition of widerange drivers on open baffles, something I recommend for everyone to try before settling on a favorite midbass or midrange driver. You'll discover through direct audition (in stereo, with a high-quality electronics) that drivers all sound radically different from each other. If you're an experienced designer, you'll know that crossovers and equalization only have a limited corrective power - the basic character of the driver is always there, no matter what you do with fancy EQ, so you might as well select a driver you like, and equalize around that.
Thought this might be a fun post for some..
http://www.6moons.com/audioreviews/davinci/davinci.html
I could be wrong.. but the drivers look like the 285 2000 and 400 2000 from Supravox:
http://www.supravox.fr/anglais/les_hp.htm
Field coil version? (certainly more options with bass response tailoring with them.. AND both units have considerably cleaner decays than their permanent magnet versions)
Presumably the 285 is full range - dominating the bandwidth between 800 Hz and 7 kHz. The 400 likely supports the 285 below 800 Hz.
Possibly of interest - because the 285 has high freq. loss on an off axis, it appears that the loudspeakers Linnman has being assembled in his room have some circular "object" above the 285 that *might* be a super tweeter (perhaps from fostex). On the other hand that object could just be related to shipping, considering that some of the high freq. loss at least on-axis will be compensated for via its inductive rise and the interaction with high output impedance amplifiers.
Sure - its pure speculation.. But I find that entertaining!
Hmm, on second thought - that upper driver looks quite a bit smaller, perhaps its the 165 GMF?
http://www.6moons.com/audioreviews/davinci/davinci.html
I could be wrong.. but the drivers look like the 285 2000 and 400 2000 from Supravox:
http://www.supravox.fr/anglais/les_hp.htm
Field coil version? (certainly more options with bass response tailoring with them.. AND both units have considerably cleaner decays than their permanent magnet versions)
Presumably the 285 is full range - dominating the bandwidth between 800 Hz and 7 kHz. The 400 likely supports the 285 below 800 Hz.
Possibly of interest - because the 285 has high freq. loss on an off axis, it appears that the loudspeakers Linnman has being assembled in his room have some circular "object" above the 285 that *might* be a super tweeter (perhaps from fostex). On the other hand that object could just be related to shipping, considering that some of the high freq. loss at least on-axis will be compensated for via its inductive rise and the interaction with high output impedance amplifiers.
Sure - its pure speculation.. But I find that entertaining!
Hmm, on second thought - that upper driver looks quite a bit smaller, perhaps its the 165 GMF?