מה דעתכם על המראה של הרמקול - AudioKinesis

[AudioKinesis]

[left] Hello go go, and thank you for the welcome! I am excited to be here, though in some ways I'm a bit out of my element - so this post is somewhat experimental. [/left]

[KOOKE]

[left]
welcome glad you have joined this forum.
can you please explain what are the sonic benefits off that design?
[/left]

[j4richter]

[left]
hi duke
welcom
although im into the DIY stuff i like the way your speaker looks and I meen the first pic.
[/left]

[AudioKinesis]

[left]Hello oferlaor, I would be happy to explain my George Costanza speaker.

What you cannot see from the photo is that this speaker is a bipolar - in other words, there is a second waveguide (horn) and a second woofer on the rear of the enclosure. These drivers operate in-phase with the front-facing drivers. The enclosure needs to be about one-and-a-half times wider than it is deep in order to prevent an undesirable dip in the on-axis response in the lower midrange region. This dip is less severe than the normal floor-bounce dip, but I seek to minimize it anyway. By carefully choosing the height of the rear woofer, I am able to compensate for the floor-bounce dip, and this bipolar configuration also eliminates the "baffle-step" shelving down of the woofer's output (which in reality only exits in the first-arrival sound anyway; it's not there in
(the "power response", or in other words the summed omnidirectional response.

The dip I mentioned is related to the path lengths that the sound travels from the front woofer, and from the rear woofer (wrapping around the cabinet), to the listener's ears. At the frequency where the difference between these two path lengths is equal to one-half of a wavelength, the rear woofer's output arrives 180 degrees out-of-phase with the front woofer's output, so we get cancellation in that region. The wider the enclosure the better the enclosure itself acts as a barrier to this wrap-around energy, and the shallower the enclosure the higher up the out-of-phase frequency region will be. And the higher up in frequency, the more effective the enclosure's width is at blocking the wrap-around. You can think of the front of a loudspeaker enclosure as a 180 degree horn, which will effectively limit the speaker's radiaton pattern down to a certain frequency (where the horn is 1/4 wavelength long, or where the baffle is 1/2 wavelength across).

So, George is wide for a reason.

Now, why in the world would anyone want to put extra drivers on the back of a speaker firing off in the wrong direction? Okay, good question. The reason is, I'm trying to increase the amount of reverberant field energy in the listening room. Compared with what we hear in a good concert or recital hall, the reverberant field in a typical home listening room arrives too early, is not dense enough, and decays too quickly. The bipolar configuration of the Costanzas puts twice as much energy into the reverberant field, relatively speaking. To give sufficient time delay before the arrival of this additional reverberant energy, I recommend setting the Costanzas about 1.5 meters in front of the wall behind them. This way the energy from the rear-facing drivers will have to travel about 3 meters farther (round trip bouncing off the wall) before it arrives at the listening position (obviously, this doesn't work below the wrap-around frequency - but that's okay). This will delay the arrival of that extra reverberant energy by about 10 milliseconds relative to the first-arrival sound, and this is sufficient. In general, reflections arriving earlier than 10 milliseconds behind the first-arrival sound tend to be detrimental, but reflections arriving later tend to be beneficial (assuming they are spectrally correct, which is why I emphasize radiation pattern control).

This bipolar approach is somewhat unorthodox; in fact, I don't know of anyone else doing a bipole quite this way, paying attention to radiation pattern control and offsetting the rear-facing drivers in the vertical plane .

In rooms that are not large enough for the Costanzas to be placed close to 1.5 meters out from the wall, the smaller speakers (Seinfelds?) probably make more sense.

This fat speaker (under its real name, the "Dream Maker") received a Golden Ear award in the June/July 2008 issue of The Absolute Sound magazine. I am not an advertiser in that magazine.

Here is a link to some pictures of the Costanzas at a small audio show a few months ago. I like to use a lot of toe-in, but that's another post for another day.


[/left]

[AudioKinesis]

[left]Forgot to post the link - here it is:

http://cryo-parts.com/lsaf_2008.html[/left]

[oferlaor]

[left]AudioKinesis,



I understand much better now, you should consider copyrighting that (the "George")...[/left]

[KOOKE]

[left]Thank you for that explanation

Have you considered adding a way to adjust the delay between front and rear drivers, in order to make them more "positioning friendly"?

If i may bring up an idea - consult with aa architect or a product designer about a cloth cover with special shape wich might soften the proportions off the speakers[/left]

[AudioKinesis]

[left]
Hello Kooke, and thank you for the welcome. Did I fully answer your question in my reply to oferlaor? If not, let me know.

Thanks,

Duke

[/left][/eng]

[AudioKinesis]

[left]
Hello j4richter, DIY is where I started out. Back in 1979 I built my first pair of speakers. They were big transmission lines that had to be assembled in my bedroom because the would not fit up the staircase if I assembled them downstairs!

I don't know if you've ever heard of SpeakerBuilder magazine, but back in 1986 and 1987 I published a couple of articles in it. One of the projects used an Israeli woofer - a 6.5" Morel.

Feel free to ask me quesitons. I won't give away the specifics of my crossover design, but I'll go into detail about pretty much anything else.

Duke
[/left]

[AudioKinesis]

[left]

Oferlaor, I don't know about copyrighting "George" - I just hope that NBC's lawyers don't see my post!

[/left]

[AudioKinesis]

[left]

Hello Kooke,

Above, I was replying to your post on page 3. Now I will reply to your post on this page.

About your suggestions, first of all thanks for your ideas. Delaying the rear-facing drivers would not be easy to do with a passive network, especially if maintaining high sound quality was the goal. So the best way to do that would probably be an active version of the Costanzas with a DSP-type crossover, something like the DEQX unit. It would take four channels per speaker, so I'd either need a custom DEQX unit or have to use two units. Either way, that gets quite expensive.

A more practical solution would be to diffuse the output of the rear-facing drivers, so that they do not generate a strong, distinct reflection. This is the approach that I have used with dipole speakers in less-than-ideal room setups, and it works well (I am a dealer for SoundLab electrostats, so I've spent a fair amount of time with big dipolar speakers).

Or, if the speakers have to be placed with a half-meter or so of the wall, might as well just go with a conventional monopole speaker.

Note that Shahinian uses a "polydirectional" approach that bounces the extra reverberant energy off the upper wall behind the speakers and from there off the ceiling, so they can be placed closer to the wall than my speakers while still imparting a pretty long path length to the additional reverberant energy. This geometry really isn't practical with the drivers I use, and my approach does allow for a longer time delay in cases where the speakers can be positioned far enough out from the wall. I am not a Shahinian dealer, but do admire Dick Shahinian's designs a great deal.

Regarding your idea of hiding the proportions of the Constanzas, actually the bevelled corners help a lot. If those enclosures were rectangular instead of octagonal (as seen from above), they would look even more massive than they do now. I don't know whether doing the equivalent of putting a striped suit on a fat man will really disguise their proportions, but it might. I have a friend in the industry who has advised a lot of companies on improving the aesthetics of their products, and I will ask his advice. Thank you for this suggestion - I wouldn't have thought to ask him if you hadn't brought it up.

It is possible to adapt the controlled-pattern offset bipole concept to a considerably smaller enclosure, but I'd have to trade off either a lot of bass extension or a lot of efficiency to make a significant change in the proportions.

Duke


[/left]

[MartinHSabag]

[left]
Hi Duke and welcome aboard,
As a reporter in this field, I had a few chances of talking to the developers and designers of various products.
Such as Mr. Yair Tamam from Magico (whom you might know), Mr. Von Oosterum from Kharma, the guys from Morel and so on, it is always so enriching and mind opening to learn the considerations behind the design and how, although different approaches, each product is special and has its fan base/
What I want to say is that at the end, the beauty of this hobby for me is that there is no "right" or "wrong" and there is a place for almost anything and everyone.

I can totally understand the philosophy behind your design and it would be very interesting to hear them first hand' especially, as my speakers are somehow, quite the opposite (Kharma CRM3.2FE). Can you please elaborate on the power requirements of your speakers?
What amplifier do you recommend (no need to mention names, just the topology or technology you believe is best etc.

Hope we'll have a chance to meet F2F someday.

Martin
[/left]

[AudioKinesis]

[left]

Hello Martin,

Thanks for the welcome!

I think that part of what makes Audio so fascinating is that it's not rocket science - it is much less well understood. Not only are we still learning about acoustics, but we have a great deal yet to learn about psychoacoustics. For instance, the concept of total harmonic distortion makes sense to our logical minds, but fails as a predictor of perceived sound quality. Much of audio is likewise non-intuitive, so there are plenty of interesting, non-obvious principles to be learned and applied. When the day finally comes that the big picture in audio is as well understood as in rocket science (it takes x amount of energy expended in a certain way to elevate payload y to altitude z), then audio will have lost a lot of its charm and intrigue.

Needless to say, it is the hope of small manufacturers like myself that there is indeed a place for everyone! I try to find a niche that is not being filled, and hope that the niche is not just in my imagination.

You asked about amplifiers - the speaker/amplifier interface is another very interesting subject - which I'll resist spouting off about too much. Briefly, my speakers should work with any amplifier that has enough power to meet the loudness requirements (which will vary from person to person and room to room and song to song).

I have exhibited at audio shows with single-ended triode (SET) amps of 5 watts, 6 watts, 18 watts, and 25 watts per channel. I have also exhibited with output transformerless (OTL) tube amps of 30 watts per channel and with solid state amps of about 50 watts per channel. With the 5 watt SET amp we did have some clipping on loud solo piano passages (that was with the Jazz Modules). With the 30 watt OTL amp on the Dream Makers, we were probably getting up around 110 dB at the listening position, based on the estimate of speaker designer Lynn Olson, whose disc we were listening to.

I pay a lot of attention to getting a very smooth impedance curve over most of the range, but do not seek to suppress the impedance peaks in the bass region. In order to have the same tonal balance with a wide range of amplifier types, a smooth impedance curve is crucial. You see, some amplifier types deliver twice the wattage when the impedance is halved, while some do not - so the impedance curve can have a significant effect on the net frequency response with different amplifier types. This is one reason why speakers that sound good with solid state amps often don't sound as good with specialty tube amps, and vice versa.

Here is a link that will hopefully show an impedance curve of my Dream Makers (aka the Costanzas):

http://gallery.AudioAsylum.com/cgi/gi.m ... edance.JPG

The Dream Makers are a "nominal 16-ohm" load. From 100 Hz on up, the impedance stays between 14 ohms and 20 ohms. The impedance of the 8-ohm Jazz Modules is also unusually smooth, staying between 8 and 12 ohms above 100 Hz.

But, what about the twin impedance peaks below 100 Hz? They really don't matter with a solid state amp - such impedance peaks are common in bass reflex speakers. And with a low-output-impedance tube amp, if we play it right there's a "free lunch" for us there. Let's take the above-mentioned 30-watt OTL amp as an example. Into those impedance peaks, it will be putting out close to 60 watts! So we do run the risk of boomy bass because we'll have excess energy in the bass region. However, we can tune the speaker much lower than normal, and instead of producing "boom" that extra power goes towards extending the bass deeper than it otherwise would have gone. In fact, we can get an extra 1/3 octave or so of bass extension. My speakers use a modular port system in which the port length is user-adjustable, so that the shape of the frequency response curve in the bass region can be tailored to the amplifier type (as well as to the room acoustic conditions, which is another topic for another day).

The amplifier topologies I tend to prefer are OTL and SET tube amps, preferably with little or no global feedback. One of my customers is a single-ended triode amplifier manufacturer. But I have customers using solid state amps and push-pull tube amps as well.

On another note, something I should have included in my reply to Kooke above, a passive delay network on the rear-facing drivers in the Dream Makers/Costanzas would require me to wire those drivers in parallel with their front-facing conterparts, rather than in series. And that would drop the impedance down to 1/4 of what it is now, thus working against compatibility with specialty tube amps.

Let me know if you have other questions, or want me to go into more detail about any of this.

Duke



[/left]

[MartinHSabag]

[left]

Great Stuff Duke !

So, when can we listen to them ?


[/left]

[KOOKE]

[left]Duke

Thank you its great to have a look "behind the scenes" of making loudspeakers[/left]

[AudioKinesis]

[left]

Hi Martin,

As for a timeframe, if all goes well I hope to have my smaller Jazz Module-style speaker ready in eight or ten weeks. From there, I'd get in touch with Mr. Eyal Windler or Mr. Ziv Rozenfeld about sending a pair to Israel for a review.

This speaker would be about 90 dB efficient (based on the Thiele/Small parameters, so that's a fairly conservative number), which makes it less likely to work well with the lower powered SET amps, but should work well with pretty much everything else. The footprint would be smaller and hopefully of a more practical shape than either of the models you have seen so far. Same waveguide, probably the same compression driver, but a different midwoofer. I'll go public with the specifics of the midwoofer once the speaker is a finished product.

Duke


[/left]

[MartinHSabag]

[eng]
Perfect !

[Eyal_W]

The room - audiokinesis
RMAF 2008

http://www.audiofederation.com/hifiing/ ... G_2172.jpg

Audiophile playing unrecognized heavy metal. Happy that audiophile had a room to play his music. Perhaps also a system that catered to his tastes. Think we should bring more than Perfect Circle and Led Zeppelin to show next year. Too lazy to Google how I am supposed to misspell Zeppelin.