Yes, you read the tittle correctly.
We have quite a few threads about center speakers and with good reason.
If you use a center, in most of its aspects, it has to be the best speaker in the whole system. If not I think you are better off without one.
Unless you have an acoustically transparent screen the speaker has to the fairly compact. It needs to be very flat across the speech discrimination band, have a cone shaped radiation pattern covering the seating area. In addition it must be able to cleanly handle a lot of power.
I recently wasted the time of sales people at Best Buy in Eagan in Magnolia. Among other things, what struck me that there was not a decent center in the room.
They tended to the shouty for speech clarity, and if not shouty they had to be turned up too loud. As far as I was concerned there was zero integration with the mains. Room coverage was in general poor, with best results right in front of the speaker.
Now I designed and built this TL center using SEAS coaxial drivers.
The lower driver having a passive crossover at 2.7 kHz and active baffle step loss compensation to the upper driver, the tweeter was not used in the upper speaker.
Now results were in the main satisfactory. However the SEAS driver has a common problem of coaxials with interference from tweeter cone reflections. In the case of this driver it a suck out from 8 to 12 kHz and then a first order roll off pretty much after 12 kHz. The roll off can be corrected to by 2 db with a pull up resistor. No more because of the close sensitivity of woofer and tweeter. The suck out is not correctable in the crossover.
So you are left with about a four db hole at 9 kHz and and by 16 kHz the system is down by 4 db again.
There is also a slight 2 db peak around 1 kHz, quite broad.
With improved sound sources and the ever present problem of a lot going on in movies even during critical dialog, getting a center perfect is more and more essential.
I have been debating how to tackle this issue for some time, and was thinking of adding another tweeter below the speaker. But there are no good candidates for this application. Correction through Audyssey improved the situation, but that is disabled bit streaming movies, when you most want it.
So I decided to press the other speaker's tweeter into service and add another half section to the crossover at 10 kHz. This is a minimal phase network.
The active crossover was modified to do more than just BSE compensation. The speaker is now +/- 1 db from 60 Hz to 16 kHz. F3 is 44Hz.
I had concerns about integration since the tweeters are so far apart. However integration is achieved at 3 to 4 ft distance.
AcuDefTechGuy is going to think I have taken a leaf out of the AXIOM play book with two tweeters, however the upper tweeter is in the nature of a specialized use of a super tweeter.
You might think that 2.5.5 way adds up to three way. However three way is not the correct definition of this unique speaker, as it has one full section crossover and two independent half section networks four octaves apart.
The improvement has been worthwhile and speech is even more natural with improved definition. I watched a recent production of La Traviata from the Royal Opera House on BD last night and the voices were magnificent with even more space around them
While I have been in touch up mode I revisited my surrounds that had a BBC "smiley" of 4db. The BBC "smiley" I think has run its day, especially in the HT arena. With the current BD disc, such as War Horse, all speakers must be highly accurate and capable.
I have now got the surrounds to +/- 2 db from 60 HZ to 16 Hz. F3 id 53 Hz. I have upped the power to these speakers from 100 watts each to 250 watts, because of the demands of current source material. From what I'm measuring the two channels driven spec of receivers is now inadequate. Modern BDs can and do blast all speakers at once.
The front and rear speakers of my system do not need touching. They are +/- 1 db right across the audio band to their F3 points. For the mains F3 is 20 Hz second order roll off. For the surround backs F3 is 33 Hz, second order roll off. All speakers have 12 db per octave (second order) roll off below F3.
One other thing, the efficiency of TLs in the bottom octave is incredible. Audyssey sets the LFE level 9 db below the rest of the system. As I have stated you have to be careful with these huge TLs if you want to keep your home intact.
I have run Audyssey multiple times and it is giving very consistent results. I think it is consistent +/- 1 db. Sometimes it will add a db here and sometimes reduce a db here or there with no change in set up.
The speakers match well, on the Audyssey pulses and the test tones the fronts, centers and surround backs all sound absolutely identical. The surrounds, are audibly very slightly subdued to the brad dip of 2 db centered around 3 kHz. The ear is not easily fooled.
The speakers give excellent 360 degree imaging localization and depth of field.
Now there is not a set of speakers here with the same drivers let alone the same cone materials. I have always maintained that timbre matching is frequency response error, matching and that is all it is.
Any speakers +/- one to two db across the band of 60 to 80 Hz with matching on and off axis response with good dispersion in the horizontal plane will match.
