These speakers need not be inexpensive. But they do not need to be over priced for asssscholes either. The target market will not be people who tend to read magazines that sell "cables" think Mix and EQ.
Maybe at first, but everything sort of fell into place there. In my current line of thinking I'd suggest that three-way is kind of a sweet spot but I wouldn't say it's strictly necessary. Mid only covers a couple of octaves compared to a decade either side, it's just the way I went.
Hmmm ... at times I've leaned towards the Zu approach - extremely competent, nearly full range drivers doing most of the work - with real subwoofers and supertweeter just filling in the edges as needed, especially at the bottom end ...
Thoughts ...?
Thoughts ...?
That is more the way I see it. Which is sort of the B&W way and sort of the Klipsch way.
Kilpsch heritage mids start from 400 to 700 hz all cross to t35 works well, but not flat fr.
The 801's if you hear the speaker you are looking at that mid range, its doing a lot.
Last edited:
I think it's a good idea to cover the midrange with a single driver - it's easiest to get good results that way. The results can be very satisfying. But my favorite speakers cross right at 650Hz- terrible idea! And yet, they are seamless. Other folks here have designed speakers that cross in the midrange and they sound just fine.
Just be prepared for a lot of work if you cross there. 🙂
Just be prepared for a lot of work if you cross there. 🙂
Where are you crossing at above this driver? Seems like a typical place for many Pro mids and 2" compression drivers due to lack of excursion lower. Same issues here?
My acoustic crossover points are ~645Hz, 4th order Bessel and 6KHz, 5th order Butterworth. That's what seemed to to fit best with the drivers themselves, I just helped the shape a little to get to the target slopes.
I have only one XO point, and it's as low as 180 Hz. I think it's a good idea to avoid crossover points that will introduce a change in polar response. The typical 6,5" + 1" dome bookshelf speaker crossed over at 3 kHz is a typical example of how not to do it.
I get it now. You want "bad" recordings to "sound good". See I want a "bad" recording to "sound bad". Yes, we are after opposite things.
Yes, I do agree with a lot. Where I differ with him is on the perception of nonlinear distortion which was a big part of PWK's legacy. He studied the physics of nonlinearity but never really linked that to perception. Had he tried to correlate his measurements to perception he would have found some rather disappointing results.
Last edited:
To me "midrange" is about 700 Hz - 7000 Hz and there should not be a crossover in that range if possible. But the edges of this range are the least important. A crossover near 2 kHz is probably worse case, but sometimes this is simply unavoidable as many here will realize. It takes some pretty big systems to cover this middle range with no crossover.
The two-way with a horn will always suffer from degrading performance above say 8 kHz (for a 1", much lower for larger drivers). But to me the degradation is less of a problem than using a crossover at this high frequency, which is a big problem.
The two-way with a horn will always suffer from degrading performance above say 8 kHz (for a 1", much lower for larger drivers). But to me the degradation is less of a problem than using a crossover at this high frequency, which is a big problem.
The issue is generally the spacing between drivers vs. wavelength at crossover. If you can get you drivers within a wavelength (or preferably less than 1/2 wave) apart at crossover then you have a good chance of an invisible crossover with smooth drivers.
This is why upper range crossovers are more difficult. The physical size of the drivers is too large. My Avatar shows an effort to move drivers tighter together. The tweeter is rear mounted and slightly overlaps the mids, getting them as close as possible. A tweeter capable of a lower crossover point was also used.
I agree that high frequency crossovers with horns are usually unsatisfactory. Phase allignment is generally impractical. If you push the tweeter back to the plane of the mid compression driver it is likely to reflect big time off of the mid horn. Even then the lateral displacement is great enough that any vertical motion pushes the time allignment off.
Always better to squeeze the top octave out of the mid horn if you can.
David
This is why upper range crossovers are more difficult. The physical size of the drivers is too large. My Avatar shows an effort to move drivers tighter together. The tweeter is rear mounted and slightly overlaps the mids, getting them as close as possible. A tweeter capable of a lower crossover point was also used.
I agree that high frequency crossovers with horns are usually unsatisfactory. Phase allignment is generally impractical. If you push the tweeter back to the plane of the mid compression driver it is likely to reflect big time off of the mid horn. Even then the lateral displacement is great enough that any vertical motion pushes the time allignment off.
Always better to squeeze the top octave out of the mid horn if you can.
David
Please name a well-known "bad" recording, preferably one that elderly people like myself would potentially know....
I think that a really good system makes "bad" (as well as "good") recordings more enjoyable than a poor system. Even though the recording in itself may have many flaws, the music is easier to listen to on a good system.
I agree regarding the ease of listening for the most part. On my system, the 'badness' of a recording is often identifiable as a peculiarity in itself. My system does not 'elide' digital artifacts with an artificial smoothness, so they are not made more palatable.
For many years I loved "Moondance" by Van Morrison. When I first listened to it on my new designs I was disappointed. I later found that the voice is clipped in several spots, which I had not heard before. Now all I hear is the clipping - if I listen close. Its still a great song.
Another example is Linda Ronstadt's latest release which is clipped in far too many places to name. Yes I hear all of this now and it all bothers me because there is no excuse for it.
The Summa is in the 700-800 Hz range. There is not much to be gained by going lower than that. So there is no reason for a new design. And anyways bigger just means dramatically fewer sales so it has to have a dramatically higher price for even fewer sales. Very quickly it becomes a "non-starter".
Only DIY will ever do really big speakers like that. Or they will cost 100's of thousands of dollars.
That's the first time I've seen an audio manufacture state midrange is 700 to 7K. Seems completely related to selling more speakers then any underlying fact. I can't imagine anybody without an agenda really believing 600 cycles is bass. A proper system design to "keep the crossover out of the midrange" is much more like 200 to 2000 cps. Why do you crossover that low anyway? The driver manufacture surely doesn't recommend it and what I find is using a CP250 that low, even in a bigger horn then yours, sounds terrible.
I find that a proper speaker system doesn't hide warts- it also doesn't emphasize them. Even with poor recordings, the vast majority of systems aren't extracting all the "good" from the recordings-they're swamped by room behavior, excessive diaphragm displacement, compression, and other playback system artifacts. Those who use the room as part of the system and optimize their (big enough) speakers accordingly, tend to find a lot more of the "good" out of even bad recordings.
I can tell an immediate improvement between speaker designs, level matched with similar FR, even on MP3s, meaning that many systems leave a lot more on the table, relative to the recording quality, than many of us would care to believe.
I can tell an immediate improvement between speaker designs, level matched with similar FR, even on MP3s, meaning that many systems leave a lot more on the table, relative to the recording quality, than many of us would care to believe.
Last edited: