This is the best 10" we've been able to find. The 6NSystem 10".
With a little EQ it will play flat to 35hz at volumes that would satisfy most all of us in normal home applications.
Lots of tonal information in low and mid base (base notes and drum kicks are fun to listen to, different harmonics) with a full sounding mid range.
We use it in a smallish transmission line cabinet up to 1530 hz.
Haven't played with many 15 and 18s. The Aura Systems 18" I have back in Canada are nice, and even took them up to 300 hz, but can't imagine there are many out there that could compare to these 10s for tone, punch and accuracy. We have not experimented with them in a front loaded horn either but don't see why they wouldn't work well.
I think I was able to attach there sheet here.
With a little EQ it will play flat to 35hz at volumes that would satisfy most all of us in normal home applications.
Lots of tonal information in low and mid base (base notes and drum kicks are fun to listen to, different harmonics) with a full sounding mid range.
We use it in a smallish transmission line cabinet up to 1530 hz.
Haven't played with many 15 and 18s. The Aura Systems 18" I have back in Canada are nice, and even took them up to 300 hz, but can't imagine there are many out there that could compare to these 10s for tone, punch and accuracy. We have not experimented with them in a front loaded horn either but don't see why they wouldn't work well.
I think I was able to attach there sheet here.
De gustibus et coloribus non est disputandum. But this is not about (music) taste. The perception of the quality parameters shows little variation between persons and over time. But there are inter-person differences in weight factors for the range of quality parameters involved, which do change over time and with circumstances/conditions.
First my apologies to all of you for following lengthy contribution, it will be my last one for a while.
Since the start of my professional career I have done a lot of research that involved image quality and on numerous occasions I have run perception experiments. In my current job as a scientist at a university, we run a lot of perception experiments a.o. to define new metrics. It is always a goal to turn perceived quality steps into quantities that can be easily measured and that have a scale that corresponds to perception steps or quality steps.
There are many biases that influence perception experiments that we have to take into account. For a good experiment you need at least 20 test persons, but it has been frequently demonstrated that you can go a long way with a reduced number (approx. 5) of trained test persons. (I take the same approach for audio and all of the comments I post about perceived quality is the average result of a group of trained listeners. This rarely deviates from my own perception, so for text simplicity I might state that it is my perception result. When the results have a too large inter-person variability I would not mention the average result or would make it clear that there is a large variability). Although there is very little deviation between test persons on what are just noticeable differences and on how the quantities and quality scale relate, there is a much more difference when it comes down to weight factors: how important that single quality is compared to other quality aspects.
When a difference is noticeable, it is always a goal (and a general attitude in the scientific approach) to relate this to measurements. (E.g. for the dynamic balance between drivers, there is a good correlation with the 2nd, 3rd & 5th harmonic distortion rise with output level. Although the distortion itself is inaudible.) However being able to correlate this to a measurement is just the first step, since it does not mean that the specific measurement fully defines that single quality. The same measurement might also relate to other quality aspects (only a part of the measured result relates to the quality parameter of interest) or there might be also other differences that relate to the quality parameter that this particular measurement does not show. Therefore there is a continuous update and refinement of metrics, which is apparently less frequent in audio than in other sectors. So metrics and the corresponding measurement systems are a result of perception studies, and when the metrics and measurement equipment is available these can be used instead of perception experiments to evaluate that specific quality, but within the boundaries of the original studies. (E.g. the dB scale and dB SPL meter are good for single tones in the midrange but do not correlate sufficiently with loudness summation for complex signals.)
And sometimes the metric or measurement equipment do not exist for a certain quality parameter and then a perception experiment with a sufficient number of participants is needed.
A single opinion is not a part of a perception result (similar conditions and specific questionnaires are required for this) so in general this has little value (to us scientists). But it sometimes intrigues as a hypothesis that might lead to perception experiments that can lead to a rejection of the hypothesis or it can lead to a validation and new results. This is how some progress is made.
A special thanks to Earl Geddes for his great contribution to the audio community by linking his horn theory to perception experiments and measurements (e.g. audibility of distortion), and the audio community need more scientists/developers/experimenters like him.
: my Graal... and certainly the room modes don't help to find it ! possible to find it with a unique design without electronic EQ ?
Yeeeeees ! I'm always looking for that ! Not easy...without a good mid-bass horn ihmo ! Midbass Horns & VOTT, etc, etc !
Or do the subjective bump at 50-70 Hz of some setups does that (a subjective energic mid-bass feeling) ?
An Onken like the TAD double 15" woofer enclosure and DIY can certainly do that, but the tunning of the load & cabinet would be a nightmare... with also a big foot print.
Two cents question : what is the average room size for this discussion thred about the Lynn's ideal speaker ? 25 square meters ? 50 ?
http://www.jblpro.com/www/products/recording-broadcast/m2#.Vdxv8TYVjwo saw in an other Multi Way thread 
Woofer has Nd magnet and EQ software possibiity... but OT (sorry)... sometimes the temptation to go to the simpliest and fastest solution.
But like the Lynn's philosophy to take the time...
According to people like Toole this is true if the testing is blind, but it is not so true when it is not blind.
Thanks
An important question that we must always keep in mind because small rooms act very differently from large rooms. When I speak here ay DIY I am always referring to a small room of less than 50 m^3 in volume. Above about 100 m^3 things get very different in the low end.
We know Lynns room is large and perhaps set up for far field listening with an appropriate reverb time.
Peter Brackes is smaller and really looks like a nearfield job. This represents a huge difference. JMCL used quite a large room demanding the bigger either 15" or even double woofer arangement.
It would be playing safe going large (2 x 12" pref 15") where it may even see some outdoor service. But there is a cost of less precision.
Again Gary,s finished Ariel replacement looks the job to meet the outside use, but the replacement cannot do the minutae, like a Veyron is not suited to a country lane drive, If you want both of them it would demand two systems.
The mid and the tweeter come first, then add on the bulky mid lower bass system to suit the size of room perhaps without the passive radiators, and with the CD for the mid. i.e Gary's progression just short of the passives. Suit a 5 meter square room.
Ariel was to me, likely only an indoor reproducer and perhaps not for close up stuff.
Last edited:
I'll assume that by minutiae you mean low level information. Nothing beats a properly designed horn when it comes to rendering low level detail and what's more, this holds true for all spl levels in horns, something that is not the case with dynamic speakers. E-stats do low level info retrieval well, but they're better at it at low spl levels, since their dynamic compression at higher levels reduces delineation in general.
I haven't heard either of Gary's speakers, (past or present) but I'm fairly sure that the current system does 'minutiae' better than the Ariel. Maybe Gary would chime in on this.
Check out FuzzMeasure. You may be pleasantly surprised by its power, ease of use, and elegance.
Also, it seems other companies are realizing that people want to do work/play and not fight with computers rather than doing what they actually want to do. Listen Inc recently released their SoundCheck software for OSX. I haven't explored it in any depth. But as of now yes for crossover modeling and things like that there are currently no options.
Last edited:
thanks Earl & Boldname.
I oftened asked to myself if the easiest way for modest diyers was not to go with sealed servo subs with plated amp with EQ embeded below 60 to 80 hz ? Some have FIR, solid digit amp, etc...
I like the Earl's idea of a commodity with bass, but not only for the driver but maybe also for the whole solution in this FR range. It's not cheap, but time is not either, as are not the square meters (footprint, waf... if no caveman or hifi toys-rooms box in box like some rich guys made in Tokyo...). E.G I think about Rythmik audio subs like...
- because the difficulty of room integration and footprint of sota vented designs as the best sealed ones: tunning is difficult and not universal in a design thread for all.
- to win time on the hardiest things to do: the choices of trade offs on the upper FRs : mid-bass - mid - HF ...
I'm not sure it's impossible to have one or many servo sub amps (or open à la linkwitz for the OB lovers in the same idea of an EQed sub system for both low and high roll offs)
and a sota Lynn Olson with prefered tubed amps above system for the musicality where it is the most important > (above) 60 hz to 80hz.
- again as a trade off for the reasons writed above and also better trade offs for cone movements if the driver is alone in the FR above 60/80 hz for micro dynamic as the contradictory needs of higher movement when a dynamic note is played... and again staying "free" and "light" snappy without being thick (=without moving for the deepest frequencies)... if I could image this, a mid-bass deserves certainly a Qtc of 0.5 in its own FR while a deep bass sounds well with a > 0.7 Qtc and a smooth roll off in its low end (it's a picture !)
even with vented enclosures and drivers with Qts around 0.2 and huge electrical damping : how could a 15" cone moves in its suburb periphery for deep bass (even if the rear radiation and vented enclosure help it for this) while moving in the town center of its aera for micro dynamics of the mid-bass and its own need of huger dynamic in this same upper range !.... At least without the cone breakups and create too massive distorsion in its upper range where ears beginns to be sensible with it (> 200 hz)
Maybe I'm too simple and I've surely a huge lack of knowledge, and if all is about trade offs... yes I find difficult the reproduction of the 20-30 hz to 700 hz range by a unique standalone driver !
But sorry, I will more lurk now as Lynn & Gary have certainly made already the choices in the trade offs they prefer in a shared philosophy about what a musical speaker is.
regards
I oftened asked to myself if the easiest way for modest diyers was not to go with sealed servo subs with plated amp with EQ embeded below 60 to 80 hz ? Some have FIR, solid digit amp, etc...
I like the Earl's idea of a commodity with bass, but not only for the driver but maybe also for the whole solution in this FR range. It's not cheap, but time is not either, as are not the square meters (footprint, waf... if no caveman or hifi toys-rooms box in box like some rich guys made in Tokyo...). E.G I think about Rythmik audio subs like...
- because the difficulty of room integration and footprint of sota vented designs as the best sealed ones: tunning is difficult and not universal in a design thread for all.
- to win time on the hardiest things to do: the choices of trade offs on the upper FRs : mid-bass - mid - HF ...
I'm not sure it's impossible to have one or many servo sub amps (or open à la linkwitz for the OB lovers in the same idea of an EQed sub system for both low and high roll offs)
and a sota Lynn Olson with prefered tubed amps above system for the musicality where it is the most important > (above) 60 hz to 80hz.
- again as a trade off for the reasons writed above and also better trade offs for cone movements if the driver is alone in the FR above 60/80 hz for micro dynamic as the contradictory needs of higher movement when a dynamic note is played... and again staying "free" and "light" snappy without being thick (=without moving for the deepest frequencies)... if I could image this, a mid-bass deserves certainly a Qtc of 0.5 in its own FR while a deep bass sounds well with a > 0.7 Qtc and a smooth roll off in its low end (it's a picture !)
even with vented enclosures and drivers with Qts around 0.2 and huge electrical damping : how could a 15" cone moves in its suburb periphery for deep bass (even if the rear radiation and vented enclosure help it for this) while moving in the town center of its aera for micro dynamics of the mid-bass and its own need of huger dynamic in this same upper range !.... At least without the cone breakups and create too massive distorsion in its upper range where ears beginns to be sensible with it (> 200 hz)
Maybe I'm too simple and I've surely a huge lack of knowledge, and if all is about trade offs... yes I find difficult the reproduction of the 20-30 hz to 700 hz range by a unique standalone driver !
But sorry, I will more lurk now as Lynn & Gary have certainly made already the choices in the trade offs they prefer in a shared philosophy about what a musical speaker is.
regards
Last edited:
Electrostatics work on very low sound pressures over a large diaphgragm area and at these low levels need to overcome the area of plastic film to move a lamellar air mass. So very low mass but it damped by contact with the large area of still air.
The Ariel tweeter is virtually a point source with little air mass in contact. If you sit close to an electrostatic speaker it is a weak and diffuse sound where a point source actually gets more coherent and much louder as you approach it. A very different listening experience. The Ariel replacement involves indirect diaphragm that is a little less able to comfortably shift very small increments, but it will do it but not quite as easily as a direct dome tweeter.
The adequacy of a single 15" driver has a lot to with room size. My listening room is nearly twice the size as Gary Dahl's, and is open to the whole first floor of the house. In typical American style, the house is "stick-built" with wood framing, so the lowest frequencies (below 40 Hz) are partially lost due to wall flexing and large window areas.
In the earliest Dallas prototype, there was an occasional impression of the compression driver having more dynamic range than the 15" woofer. Oddly enough, the Altec 416-Alnico had more apparent dynamic range the JBL 2226. Maybe the 2226 just needed more power - it's intended for movie theaters and sound reinforcement use, after all.
The obvious solution was to curtail the dynamic range of the compression driver by stepping down to a 1" exit driver, but I didn't want to do that.
The few times I've auditioned systems with a pair of 15" drivers the apparent dynamic range is much greater, as you would expect with a 6 dB increase in headroom and power-handling (3 dB from doubling-up voice-coils and another 3 dB from efficiency gain). The biggest downside is doubled size, although that's still not as large as a LF horn.
If the builder has the space, the Inlow Sound 135 Hz bass horn is an attractive solution, with the most dynamic range of all. But I don't have the room, don't want to mess with time-alignment through DSP processing, and don't want a folded horn (which has its own set of problems). So a pair of 15" drivers looks like the most practical solution at this time.
Last edited:
Peter, this suggests to me a possible reason for the often reported preference for single-ended amplification using devices with relatively high measured distortion. In most loudspeakers, the distortion performance and variations with level is significantly different in different frequency bands and from the different drivers as well. The distortion types, quantities, affected frequency ranges, and changes with level will be different from those naturally generated in ears. Playing through an amplifier or processor which generates its own distortions (if perceived as similar to the nature of distortions that ears add) could swamp the speakers' distortions, making the overall effect seem more natural and less obviously heard as coming-through-a-speaker.
I don't know of a way to test this hypotheses, nor even a better way to state it such that it might be testable, though. I previously floated this concept over in the global feedback thread over in solid state, but it caught no traction at all there - it might be just too vague to actually mean much. But I at least find it intriguing...
No, it's the exact opposite of what you describe. The first clue should be the difference in sensitivity (and efficiency) at around 20dB between the two types of drivers. So, how can it be that the CD "is a little less able to comfortably shift very small increments" when it takes 100 times less power for the compression driver to produce the same amplitude signal as the dome tweeter? How can it be harder for the CD to move these "very small increments", when in fact it has to move only a small fraction of the distance that the dome must cover for the exact same output? It's obvious that the dome tweeter has it the 'hardest' between the two - and not by a small margin.
Bill
My own work suggests that what you hypothesize has some validity, although not completely.
We found that the order of the nonlinearity was far more significant than its level. Loudspeakers by their nature only have low orders of nonlinearity, up until they reach the limits of excursion at which point the orders increase rapidly.
On the other hand electronic can have very high orders of nonlinearity even at low levels - i.e. crossover distortion (which is known to be audible at fractional percent's of THD.) The amps nonlinearity will swamp the loudspeakers at lower levels (perceptibly) but no so much at high levels as long as the amp is not clipping.
This is all a complex dance between the amplifier design and the loudspeaker and one in which measures of THD offer absolutely no clue.
As an aside, it is know that feedback will lower the low orders of nonlinearity but will, as a result, raise the higher orders. This it is quite possible for feedback to make the audible distortion worse while actually lowering the measured THD.
My own work suggests that what you hypothesize has some validity, although not completely.
We found that the order of the nonlinearity was far more significant than its level. Loudspeakers by their nature only have low orders of nonlinearity, up until they reach the limits of excursion at which point the orders increase rapidly.
On the other hand electronic can have very high orders of nonlinearity even at low levels - i.e. crossover distortion (which is known to be audible at fractional percent's of THD.) The amps nonlinearity will swamp the loudspeakers at lower levels (perceptibly) but no so much at high levels as long as the amp is not clipping.
This is all a complex dance between the amplifier design and the loudspeaker and one in which measures of THD offer absolutely no clue.
As an aside, it is know that feedback will lower the low orders of nonlinearity but will, as a result, raise the higher orders. This it is quite possible for feedback to make the audible distortion worse while actually lowering the measured THD.
This is interesting - I have never heard that happen even with massive drivers loaded in bass horns. You must be talking about the mid and small format compression drivers?
I agree. There is so much discussion of what we can hear that is not measurable. Some can hear differences in magnets, some can hear differences in capacitor construction, still others hear differences when a piece of foam is added to the horn, still others hear the differences between tube and solid state. All hogwash right?
Yet some of us can actually hear these things.