I'm curious to know where that ''law'' comes from ?
I've spent hundreds of hours in my life just to play with crossover points and i see no reason to believe there is any ''no-crossover-zone''... It just doesnt make sense to me. At all.
Why 1.5khz would be ''more dangerous'' than 150hz or 8khz ?
A bad integration, on the other hand... THAT might be the culprit. Maybe people are following that dogma just because they are not able to make a good mid/tweeter integration ?
I've spent hundreds of hours in my life just to play with crossover points and i see no reason to believe there is any ''no-crossover-zone''... It just doesnt make sense to me. At all.
Why 1.5khz would be ''more dangerous'' than 150hz or 8khz ?
A bad integration, on the other hand... THAT might be the culprit. Maybe people are following that dogma just because they are not able to make a good mid/tweeter integration ?
Hi,
Its because if you get it wrong, which most do,
it sounds the worst done in the critical midband.
rgds, sreten.
Its because if you get it wrong, which most do,
it sounds the worst done in the critical midband.
rgds, sreten.
The Fletcher-Munson research showed a long time ago that human ear is indeed sensitive to loudness differently over the audible bandwith.
https://en.wikipedia.org/wiki/Equal-loudness_contour
That being said, i wouldn't even think to restrain myself have drivers crossed at 2khz or any other frequencies.
The integration quality in a multi-way system is the key; you need to use drivers in their comfort zone, make sure the FR is flat and get full, effortless, 10 octaves when possible. But no need for any dogma ? 😕
https://en.wikipedia.org/wiki/Equal-loudness_contour
That being said, i wouldn't even think to restrain myself have drivers crossed at 2khz or any other frequencies.
The integration quality in a multi-way system is the key; you need to use drivers in their comfort zone, make sure the FR is flat and get full, effortless, 10 octaves when possible. But no need for any dogma ? 😕
I find that all crossovers often sound bad, regardless of where they're at, especially higher order filters with their ringing. I can't stand the typical 4th order subwoofer setup, it just sounds wrong to me.
I'm constantly amazed at what people consider the "midrange".
Most of the sound from a loudspeaker is below 1.5 kHz.
Most of the sound from a loudspeaker is below 1.5 kHz.
As I see it there may actually be two issues, which have been touched on here. The first, when a crossover isn't perfect, and that's very likely, it's better to stay away from 3-4kHz.
But radiating middle and higher frequencies satisfactorily is otherwise not that difficult. As the room inevitably gets involved crossover issues can become harder to avoid, overriding the sensitivity issue.
But radiating middle and higher frequencies satisfactorily is otherwise not that difficult. As the room inevitably gets involved crossover issues can become harder to avoid, overriding the sensitivity issue.
160-2,500 Hz.
Ideally an XO wants to be in a place where centre-to-centre driver spacing is less than a 1/4 wavelength.
dave
A highly respected audio guru acoustician David Griesinger stated in one paper (PowerPoint), that we sense "clarity" of sound image most sensitively in the range of 800 to 4000Hz. He suggested that range to be at least linear phase. It's here http://www.davidgriesinger.com/ICA2013/Optimizing%20clarity.pptx
A 3-way speaker with a low distortion 3-4" midrange will often meet that criteria and smooth directivity, alas a large horn ala Geddes
http://www.davidgriesinger.com/
A 3-way speaker with a low distortion 3-4" midrange will often meet that criteria and smooth directivity, alas a large horn ala Geddes
http://www.davidgriesinger.com/
Yet no one has been able to scientifically prove that people can hear phase shift of a music passage that occurs over a range of more than 1/3rd of a octave.
It's all down to frequency response (amplitude) imo. The flatness from 500Hz-5kHz matters because of the deviations in the Fletcher–Munson/Equal-loudness contour at those frequencies. Crossing over there gives the designer a much greater chance of messing up the frequency response, especially off-axis / power response in that range and therefore increasing the chances of people having 'bad experiences' with speakers that cross in that range.
250-3.5k ??
I strive to cover 250-3.5K
It's really not that easy to do.
I'm beginning to lean towards the "bulk energy range" of 80-800 to be covered by one driver.
By the way, this is a great thread and should provide lots of "creative"
discussion !
p.s. I don't have crossover phobia, but I have friends that suffer
from this affliction 🙂
I strive to cover 250-3.5K
It's really not that easy to do.
I'm beginning to lean towards the "bulk energy range" of 80-800 to be covered by one driver.
By the way, this is a great thread and should provide lots of "creative"
discussion !
p.s. I don't have crossover phobia, but I have friends that suffer
from this affliction 🙂
I don't know exactly why you call it that but I also attach significance to this region, I mean 800Hz give or take is fairly down away from the sensitive region yet not inextricably entrenched in the room.
edit: Maybe I should just add that a tweeter can be made to cover 800Hz and up.
Last edited:
+1 for the bulk energy range !
Which horn and CD combo is good enough to cover 250-3.5K ? Below 500-600 Hz too much horn, what ever the CD, seem not to be good enough if allowed to cover as well the upper frequency (i.e. a 2" CD !). below those average 500 hz : most of horns give a too thin sound or miss to do good upper mid if this low range is good in relation to the horn choosed for the low FR qualities (at least this is my understanding).
About the OP, the crossover is a question of radiation patern, not just frequencies : you must choose a diameter cone which is usable and have to make trade offs !
Even a wide band driver do not all good ! 100 hz to 8000 hz FR drivers are coherent but have weakness (soundstage, tonal balance according to me....): choose your poison.
I have a 5" crossed 125 hz to 2600 Hz... Not bad, but as ScottL (and not for the same reasons - I'm not concerned by tubes and high sensivity speakers), I believe this low mid-bass range needs something energic with a great dynamic to make it lively... but lightly enough (it's not just a boum-boum range, far of it in fact, let say between the energy and resolution needs and that's why it's so hard to do for many designers. I'm not sure the only air impedance front loading of a horn (at least in a normal space, (i.e. not 5 meters horn with only 3 meter of space listening in front of it as seen in certain exceptional setups!) is enough in this range, though never heard a 4"CD in this range !
+1 also for Juhazi : but the horn , it's difficult to find a driver which is all good from usuable 800 hz to 4000 hz or more : most of the time they are too "light in the low if they climbs too much !
The line arrays seem one of the solution sometimes to resolve those problems : you can in the low end have a multipe array of bass or mid bass or make more little speakers with an array of little drivers : lookt at links from others designers in Linkwitz site or here array in Full Range Forum !
But no doubt a good designer is abble to cut in the range you talk about : Gedlee speaker are highly appreciated.
Maybe that's why OS waveguide found a "niche" : difficult to find a profile horn which sounds good in a little domestic space !
There are a lot of trade offs to choose in a speaker and you highlighted just one, imho...
Last edited:
... And here : http://www.diyaudio.com/forums/multi-way/231551-beyma-12p80nd-tpl-150-a.html#post3400343
This member choosed a high XO around 1600 hz to cross a 15" with an AMT tweeter with good sucess. The 15" is wide band from 50 hz to 1600 Hz in a 90 liters cabinet.
Some would ask themselves how is the polar map of the 15" above 600-800 hz, but seems to work according this member... Again choose your poison, if you add a driver between the huge beaming 15" and the tweeter, you add also a pass band XO !
This member choosed a high XO around 1600 hz to cross a 15" with an AMT tweeter with good sucess. The 15" is wide band from 50 hz to 1600 Hz in a 90 liters cabinet.
Some would ask themselves how is the polar map of the 15" above 600-800 hz, but seems to work according this member... Again choose your poison, if you add a driver between the huge beaming 15" and the tweeter, you add also a pass band XO !
12" and 15" coaxials with compression drivers can also cross near 1kHz and give smooth+high directivity.
Griesinger talks more about phase coherency in spatial hearing here http://www.davidgriesinger.com/ica2010/paper2.doc Also pitch and timbre of an instrument's sound or speech are very much connected to 2nd and 3rd harmonics. If the harmonic overtone has different phase, it doesn't sound the same!
I am not very sensitive to hearing differences, but avoiding xo between 800-3kHz just sounds better. Many things behind that... A coaxial mid-tweeter behaves better even though usually crossed in there.
-
Actually I like steep xo slopes better that say LR2 near midrange. With shallow slopes ther will be even 3 different radiators blowing the same signal - jus think about wat happens is spatial phase coherency then! Could it be that those phase deviation tests have been done by modifying the signal and perhaps even with headphones?
Griesinger talks more about phase coherency in spatial hearing here http://www.davidgriesinger.com/ica2010/paper2.doc Also pitch and timbre of an instrument's sound or speech are very much connected to 2nd and 3rd harmonics. If the harmonic overtone has different phase, it doesn't sound the same!
I am not very sensitive to hearing differences, but avoiding xo between 800-3kHz just sounds better. Many things behind that... A coaxial mid-tweeter behaves better even though usually crossed in there.
-
Actually I like steep xo slopes better that say LR2 near midrange. With shallow slopes ther will be even 3 different radiators blowing the same signal - jus think about wat happens is spatial phase coherency then! Could it be that those phase deviation tests have been done by modifying the signal and perhaps even with headphones?
Last edited:
Hi,
I'd like to point out this is potentially interesting topic with
one major flaw and that is the name of the thread. It doesn't
do justice to the term dogma.
First you call it dogma, then the "law". A dogma in its simplest
meaning is evident truth about something. The Ohms law is
certainly a dogma. That our planet Earth is round can also
be considered to be a dogma.
The audio community in which the leading experts have
something to say, has come to a conclusion that sound quality
is definitely higher in multiway systems than in a single drive unit
loudspeaker and the former have XO filters.
So, people who think this is not true are ignoring or have failed
to do the research to meet the same conclusion. They are certainly
entitled to have their own point of view, but it is not a dogma.
A dogma is a principle set down by an authority as incontrovertible true.
If it is actually true or not is an entirely different story.
Dogmas are usually used to stifle open discussion within the ranks. Certain institutions rely heavily on dogma to assure their continued existence.
If it is actually true or not is an entirely different story.
Dogmas are usually used to stifle open discussion within the ranks. Certain institutions rely heavily on dogma to assure their continued existence.
- Status
- Not open for further replies.