Thanks again. 6 ms gate minimum at 1.4 m free around the cab ?
Wesayso...is Sauron. The Two towers and 50 ScanSpeak gobelins... I am still a hobitt in front of such complex job... no chance I putt a finger in that port.
Wesayso...is Sauron. The Two towers and 50 ScanSpeak gobelins... I am still a hobitt in front of such complex job... no chance I putt a finger in that port.
Yes you can, but as expected (and predicted beforehand) the ST260 won't sound good below 1000-1200 Hz, even when some of the best comp. drivers are used and especially at higher SPLs
With some basic understanding of horn science combined with common sense, it's easy to see why a driver yelling in reactive mode isn't particularly pleasant to listen to.
The Fc of an OSWG is roughly between 800 and 1000 Hz, depending on coverage angles and dimensions.
Surprisingly (??), the throat sections of Mabat's recent horns tend more and more towards classical profiles, with increased throat resistance at lower frequencies.
Last edited:
In the pro world, these issues are overcome by using larger drivers above their resonance freq.
However, even RCF uses a fairly deep expo horn in its 2-way systems combined with a meaty 4" dia. driver for an unprecedented low crossover frequency of 650 Hz.
In their more recent ART series, the XO was increased to 800 Hz to compensate for the reduced loading of the shallow waveguide.
Last edited:
Does it need to? Besides, it does. A waveguide generally holds a more narrow angle down low than a horn of the same length and mouth size.
Here's some discussion about nonlinear distortion and waveguides vs horns...
https://www.diyaudio.com/community/threads/beyond-the-ariel.100392/post-3696522
Here's some discussion about nonlinear distortion and waveguides vs horns...
https://www.diyaudio.com/community/threads/beyond-the-ariel.100392/post-3696522
I haven't got further from simulations for now. Unfortunately my time is needed elsewhere. Here's the crossover sim for frequency response from ATH thread - it should present a good starting point.
Last edited:
Are you really missing the point? You bring no new info in that link though... not valid for this discussion.
As the discussion originated from the choice between the ST260 and the ST280, both well designed ATH waveguides with slightly different properties.
The Yuichi, to which you reacted, was mentioned to ***** the performance of a particular CD down low. So that's what got you all rattled up?
The main message, important to this thread was: use the ST280 if you want to cross low, use the ST260 if you want the most even response.
What you reacted to was a 'side discussion' about the highly debated words 'load' and "driver". Not about potential HOM's etc. A Yuichi wasn't discussed as a potential horn for this project. So what was the point? Care to say a few things about the choice between the ST260 and the ST280 and crossing at ~900 Hz?
The other question, important to this thread was: which CD to use... which made Fluid mention the Yuichi, to make a point.
Thank you all for the precious help you provide.
May I ask please what you think about this non linear distorsion test of this close driver : Faital 12PR300 measurement -predecessor of 12R320 used in this thread- provided by Brandon (link in one of my post above). this is a sweep test, so fundamental measured if I understand. Notice between 500 hz and 600 hz H5 is a little high with -52 dB "only".
Are the H5 (grey line) not a problem to be used above 300 hz, or one just should not care and go straigth forward to use it 2 ways till the cut-offs discussed here. (trade off between a 2 ways and a 3 ways).
On the 150 hz tone, one can see if I am rigth : H3=-48 dB ; H5=-59 dB.
Of course I will try for real with my 12PR320s' (active filter for prototyping in mind) and first a 2 ways will be tried, but I wanted to hear about your experience as a design choice factor about such graphs as we see plenty (Mark K, Zaph, HifiCompass, etc).
May I ask please what you think about this non linear distorsion test of this close driver : Faital 12PR300 measurement -predecessor of 12R320 used in this thread- provided by Brandon (link in one of my post above). this is a sweep test, so fundamental measured if I understand. Notice between 500 hz and 600 hz H5 is a little high with -52 dB "only".
Are the H5 (grey line) not a problem to be used above 300 hz, or one just should not care and go straigth forward to use it 2 ways till the cut-offs discussed here. (trade off between a 2 ways and a 3 ways).
On the 150 hz tone, one can see if I am rigth : H3=-48 dB ; H5=-59 dB.
Of course I will try for real with my 12PR320s' (active filter for prototyping in mind) and first a 2 ways will be tried, but I wanted to hear about your experience as a design choice factor about such graphs as we see plenty (Mark K, Zaph, HifiCompass, etc).
Attachments
Hi all,
Wanted to know please : is it also possible to blend from a radiation patern point of view the ST260 or 280 horn with a 8" or 6" in a 3 ways in spite of a 2 ways with the 12" up to 900/1200 hz ?
I mean : is there something to winn here in term of sound quality between the two ? Tony Gee from Humblehomemadehifi says his 3 ways with the 12pr320 and a 6 +1 has "sligthy more fine details" while the 12pr320 with the horn and 1" has more dynamics.
Is it due to the 6" filler in 3 ways or what different pattern in that medium range up to the horn/tweeter ?
Cause 200 hz to 900/1200 hz, there are nice and cheap drivers like the 6" Monacor 108/pro for instance or the 6" from 18thSounds
I wonder cause the two first items of the whish list for a new loudspeaker is a best bass behavior up to the low mid with dynamics. AND the controlled directivity in the upper ranges where the horn allows it.
Wanted to know please : is it also possible to blend from a radiation patern point of view the ST260 or 280 horn with a 8" or 6" in a 3 ways in spite of a 2 ways with the 12" up to 900/1200 hz ?
I mean : is there something to winn here in term of sound quality between the two ? Tony Gee from Humblehomemadehifi says his 3 ways with the 12pr320 and a 6 +1 has "sligthy more fine details" while the 12pr320 with the horn and 1" has more dynamics.
Is it due to the 6" filler in 3 ways or what different pattern in that medium range up to the horn/tweeter ?
Cause 200 hz to 900/1200 hz, there are nice and cheap drivers like the 6" Monacor 108/pro for instance or the 6" from 18thSounds
I wonder cause the two first items of the whish list for a new loudspeaker is a best bass behavior up to the low mid with dynamics. AND the controlled directivity in the upper ranges where the horn allows it.
It is at lower crossover points because the smaller drivers don't beam as much as the larger drivers do. I don't think this makes much sense though.
Pick your poison, do you want a compression driver on a horn or a tweeter and smaller mid, no one can tell you what you will like more 🙂
Does the beamwidth of the two cross? You need to put the polars over one another, or understand each enough to do that mentally at least.
The question remains... where do they cross? Is it something the woofer can manage?... and that's before you look at the overall picture and evaluate it's potential.
Yeah, radiation pattern is charasteristics for the device size and shape and does not change with crossover. A crossover blends these together at any frequency, resulting pattern is blend of the two, one low frequencies and another for higher. If the patterns don't match very well which means their directivity index and shape is different around the crossover, then the resulting system DI is not as smooth as it could be. Basically when the match is fine there is usually nice smooth ramp up in DI. If waveguide DI is less than woofers then there would be peak in DI below crossover, and if waveguide DI is higher than woofers then there would be steep slope up in DI below xo, basically a dip compared to smooth slope. (mabat just posted example of this while writing).
How audible each is? I don't know but I'd guess smooth is less audible than any sudden change, or bumpy transition. Kimmosto posted something related to this in ASR, when charasteristics of low and high sound source differ a lot, the sound is worse and when they are more alike sound gets better. Or thats how I interpret it 🙂 Directivity, distortion and dynamic capability was mentioned. Thinking of it, there is no reason use too small woofer, the bigger the better to match dynamics and distortion of the waveguided tweeter. There is also no point crossing tweeter too low for distortion performance (SPL capability) again. Best performance is achieved when woofer is as big as possible and tweeter as small as possible, and waveguide brings them together acoustically. Basically all things come together by choosing appropriate waveguide and woofer. If this cannot be done then just use what you have and thats it 🙂
edit. Basically, trick that a waveguide does is it changes DI of very small transducer to match that of bigger transducer. One can also increase woofer / system directivity with a baffle or dipole/cardioid, or even with another waveguide, to bring DI higher toward DI of "too big" tweeter waveguide. Using too big waveguide with very good diffraction performance, and small woofer with diffraction higher up than crossover, one could possibly get away with most edge diffraction issues. Although, diffraction performance seems to be good enough with matched devices, for the amount I think I can perceive it. Speakers disappear just fine with some hints of edge diffraction and most logical thing is to just make directivity match to reap all benefits of waveguide.
How audible each is? I don't know but I'd guess smooth is less audible than any sudden change, or bumpy transition. Kimmosto posted something related to this in ASR, when charasteristics of low and high sound source differ a lot, the sound is worse and when they are more alike sound gets better. Or thats how I interpret it 🙂 Directivity, distortion and dynamic capability was mentioned. Thinking of it, there is no reason use too small woofer, the bigger the better to match dynamics and distortion of the waveguided tweeter. There is also no point crossing tweeter too low for distortion performance (SPL capability) again. Best performance is achieved when woofer is as big as possible and tweeter as small as possible, and waveguide brings them together acoustically. Basically all things come together by choosing appropriate waveguide and woofer. If this cannot be done then just use what you have and thats it 🙂
edit. Basically, trick that a waveguide does is it changes DI of very small transducer to match that of bigger transducer. One can also increase woofer / system directivity with a baffle or dipole/cardioid, or even with another waveguide, to bring DI higher toward DI of "too big" tweeter waveguide. Using too big waveguide with very good diffraction performance, and small woofer with diffraction higher up than crossover, one could possibly get away with most edge diffraction issues. Although, diffraction performance seems to be good enough with matched devices, for the amount I think I can perceive it. Speakers disappear just fine with some hints of edge diffraction and most logical thing is to just make directivity match to reap all benefits of waveguide.
Last edited:
Late to the party ... but a nice horn to combine with 12" is the Limmer 033: https://www.limmerhorns.de/033/
I use it with an BMS Neo Coax, you can cross in the area of 800Hz. Both the 12" and horn will not have tooo much directivity in this area but the rising directivity should match well.
I use it with an BMS Neo Coax, you can cross in the area of 800Hz. Both the 12" and horn will not have tooo much directivity in this area but the rising directivity should match well.
This only the most general rule of thumb. Actually, directivity changes quite profoundly in the crossover region, and this is a band that spans a couple of hundred Hertz.
This is also why it makes sense to intentionally mismatch DI of waveguide and woofer, if the crossover would otherwise leave a more pronounced dent in the DI curve. At least this is what I had learned from my experiments with Ath and VCad.
Thanks for all the answers here 🙂
About the blend of two drivers:
When superseeding two drivers off-axis measurements, where is the mininimum off-axis degree one should look at for a match, not to break too much the constant directivity of the overllaping in the pass band ? Down to 45° ? To 90° ?
@mabat, much thanks to look at this thread. Seems to work well with the 900 hz cut off and the st260 indeed.
Btw I saw noone posted a design with the 280E you generously shared as well. Following all the inputs here from fluid and others, this horn should be good to help a little a 1" to go as low as 900/1000 hz.
From instinct, FWIW as instinc is not science and I has to guess as it is hard to try them all, I like the idea of the more off axis decrasing amplitude (beaming) of this 280 VS the 260. I surmise it will spread less energy in a living room off-axis and that's perhaps be profitable to feel less the mismatch between a horn and the non loaded driver.
But I know this feeling is no science, indeed I wonder about the behavior of a 12" midwoofer PA that high till 900 hz and more. Ok people does it for decades, but dunno if a 12" can resolve "fine details in this midrange window when competing with a dedicated midrange as the Fostex of the illustration.
Btw, empirically Vineeth member testimonied for the moment, more precision from the two ways with horn (with a 9" and a classic thousant and half cut-off circa) versus a classic 3 ways with a good mid (here a Wavecor).
About the need to design with active crossover for iterations 🙂 ....
About the blend of two drivers:
When superseeding two drivers off-axis measurements, where is the mininimum off-axis degree one should look at for a match, not to break too much the constant directivity of the overllaping in the pass band ? Down to 45° ? To 90° ?
@mabat, much thanks to look at this thread. Seems to work well with the 900 hz cut off and the st260 indeed.
Btw I saw noone posted a design with the 280E you generously shared as well. Following all the inputs here from fluid and others, this horn should be good to help a little a 1" to go as low as 900/1000 hz.
From instinct, FWIW as instinc is not science and I has to guess as it is hard to try them all, I like the idea of the more off axis decrasing amplitude (beaming) of this 280 VS the 260. I surmise it will spread less energy in a living room off-axis and that's perhaps be profitable to feel less the mismatch between a horn and the non loaded driver.
But I know this feeling is no science, indeed I wonder about the behavior of a 12" midwoofer PA that high till 900 hz and more. Ok people does it for decades, but dunno if a 12" can resolve "fine details in this midrange window when competing with a dedicated midrange as the Fostex of the illustration.
Btw, empirically Vineeth member testimonied for the moment, more precision from the two ways with horn (with a 9" and a classic thousant and half cut-off circa) versus a classic 3 ways with a good mid (here a Wavecor).
About the need to design with active crossover for iterations 🙂 ....
It would be probably better if it was a little lower but I guess that would complicate the (passive) crossover just too much.
I don't know, I would hesitate to use the ST280, that was actually not meant as an example of a good desing - it's merely an experiment for comparison.
Not all 12" PA woofers perform well enough around 1 kHz (maybe even majority don't), but some do perform really well. Interestingly, still it's easier to find a good 12" than say a 10" - it's not that smaller is always better.
Last edited:
Thanks,
And very few 1" seems to have a good behavior as well below 1200 hz...where the spl bump pan occurs for such compression driver size... even less below 1000 hz hence the BMS4550 natural choice that can handles it but still with a diving curve here.
Yes it seems not an easy task to cut off below that plateau with a passive. Btw there was a nice proposal from fluid before in this thread. 🙂.
I turn in round wanting to reinvent the wheel. I assume it is what happens when you have not understood yet what a circle is !🤔
Btw, is your last 260 itteration better than the 260ST ?
And very few 1" seems to have a good behavior as well below 1200 hz...where the spl bump pan occurs for such compression driver size... even less below 1000 hz hence the BMS4550 natural choice that can handles it but still with a diving curve here.
Yes it seems not an easy task to cut off below that plateau with a passive. Btw there was a nice proposal from fluid before in this thread. 🙂.
I turn in round wanting to reinvent the wheel. I assume it is what happens when you have not understood yet what a circle is !🤔
Btw, is your last 260 itteration better than the 260ST ?
Last edited: