I have 3x KEF Blade Coax drive units that I want to use in a new setup in a new house, together with 8x Dayton RS180 per side in a baffle wall.
(I have 3 of these nice coaxes, so there will also be a center speaker with the same setup but with only 4 RS180 mid)
Does anyone here know how to calculate spacing and possibly distance from floor and ceiling, and possibly have general thoughts about solutions?
Appreciate feedback and discussions!
My thoughts are in this way to solve a lot of the room nodes in the elevation plan. Split frequency will be considered, but around 4-500hz possibly with some overlap.
Originally Kef Blade is shared at 320hz, but I personally feel it is a bit low. There will be a subwoofer to take the bass area.
I have a Minidsp OpenDRC-DA8 for active crossover incl FIR.
See picture for alternative placements of mids.
In one set-up, the coax is at a height of approx. 1.25m, which I feel is somewhat high compared to the approx. 1m sitting position?
(I have 3 of these nice coaxes, so there will also be a center speaker with the same setup but with only 4 RS180 mid)
Does anyone here know how to calculate spacing and possibly distance from floor and ceiling, and possibly have general thoughts about solutions?
Appreciate feedback and discussions!
My thoughts are in this way to solve a lot of the room nodes in the elevation plan. Split frequency will be considered, but around 4-500hz possibly with some overlap.
Originally Kef Blade is shared at 320hz, but I personally feel it is a bit low. There will be a subwoofer to take the bass area.
I have a Minidsp OpenDRC-DA8 for active crossover incl FIR.
See picture for alternative placements of mids.
In one set-up, the coax is at a height of approx. 1.25m, which I feel is somewhat high compared to the approx. 1m sitting position?
Attachments
Last edited by a moderator:
You're trying to crossover from a point source to a line source. That's not a great idea, it's really difficult to accomplish.
The reason that the woofers in the Kef Blade are so close to the coax is because they're trying to avoid crossing over from a point source to a line source.
The reason that the woofers in the Kef Blade are so close to the coax is because they're trying to avoid crossing over from a point source to a line source.
Sure. The Snell XA90 is a WMTMW and Dave later made log spaced arrays for Bose.
The first is a point source, second is a line source. But Smith isn't mixing array types in one design.
Smith's reference speakers at his home are the Snells he designed. Smith replaced Kevin Voecks when Voecks went to Revel.
The first is a point source, second is a line source. But Smith isn't mixing array types in one design.
Smith's reference speakers at his home are the Snells he designed. Smith replaced Kevin Voecks when Voecks went to Revel.
I don't know if 15" would blend well with the tiny kef: i tried at home with one 15" and the sp1793 i own ( Ls50 meta drivers) and wasn't thrilled by outcome.
Don't take me wrong: it works but something is either overrated in lowmid or the 5" is too small at frequency i Xed it ( circa 300hz) for my preference. Higher freq xover i struggled with center to center distance and directivity mismatch.
Dave Smith indeed didn't mixed both approach in same design but if you think about it the Snell mid/high is an mtm which mimic coax behavior on some points. Member Ra7 once tried to apply it ( log array) to a kef driver and Dave didn't discouraged him in the thread dedicated to it. Rather the contrary, he explained it was a variant around the principle iirc.
I will try to find it and link the thread.
But i agree with Mayhem in wall is open to creativity.
Don't take me wrong: it works but something is either overrated in lowmid or the 5" is too small at frequency i Xed it ( circa 300hz) for my preference. Higher freq xover i struggled with center to center distance and directivity mismatch.
Dave Smith indeed didn't mixed both approach in same design but if you think about it the Snell mid/high is an mtm which mimic coax behavior on some points. Member Ra7 once tried to apply it ( log array) to a kef driver and Dave didn't discouraged him in the thread dedicated to it. Rather the contrary, he explained it was a variant around the principle iirc.
I will try to find it and link the thread.
But i agree with Mayhem in wall is open to creativity.
How did you get those Blade drivers? Wish such things were available to the DIY market easily!
Just an idea, but could you place the RS180 in a face-to-face configuration and have then fire out a slot? This would help collapse the area of the sound source closer to the coax. Kind of like this:
(Image from google search: https://i.pinimg.com/1200x/a1/6d/ce/a16dcebf38a970fe38bfe6febc1f3253.jpg)
I'd also say with a subwoofer taking below 80Hz you just don't need 8X RS180 per channel as the coax will not keep up anyway. 3 spaced equidistant around the outside is enough to handle 80Hz - 500Hz at a level the tweeter would have thermal issues.
Just an idea, but could you place the RS180 in a face-to-face configuration and have then fire out a slot? This would help collapse the area of the sound source closer to the coax. Kind of like this:
(Image from google search: https://i.pinimg.com/1200x/a1/6d/ce/a16dcebf38a970fe38bfe6febc1f3253.jpg)
I'd also say with a subwoofer taking below 80Hz you just don't need 8X RS180 per channel as the coax will not keep up anyway. 3 spaced equidistant around the outside is enough to handle 80Hz - 500Hz at a level the tweeter would have thermal issues.
Last edited:
May I ask if you tested sensitivity?
I bought a set of LS50 Meta's to see how they performed as "top" for my woofer-boxes - and to compare them with a DIY box I have with non-meta R-series midranges.
They seem very low sensitivity, but that could surely be the passive design... Not sure if I want to dismantle the LS50 to get the drivers, if the difference is too small.
Hi,
No i have not. But they are lowish sensitivity for sure, i think i estimated them circa 85db iirc ( woofer part). I doubt the passive design is responsible of this, they are very well engineered drivers but they are tiny things wrt to coax i'm used to ( pro).
I doubt you'll see this much difference in sensitivity between the two driver family, after all these are 4"/5" hifi drivers... happily they can cope with a bit of power sent to them.
It will depend at which freq you plan to x them too, the kind of load they'll see, etc,etc,...
The 15" i used them with is the one into my three ways and it is coupled to a 4"/5" mid and 1" tweet too ( Technics SB-M2) but they are more sensitive ( circa 92dbspl). Results were kinda similar in character when used with the Kef but without some of the nasty features the original design have to me. But still there is something which could be better to my preference: too much of a mismatch of membrane area if x'ed around 300hz or lower, too high a center to center distance if x'ed at 600hz to my pref... the later sounded 'better' to me though - read more to my liking. I don't know if this is because i'm used to this sound with the SB- M2 ( it's close to the xover freq i choosed when i went dsp with them) or because of membrane mismatch... very probably a mix of both.
From test i've done they blend well with the 2x RS225 at different freq for two octave ( from 150hz to 600hz). The limits comes from the 225 lower in freq. For my preference: you can't beat membrane area in the low end... it's an hard, cold fact.
No i have not. But they are lowish sensitivity for sure, i think i estimated them circa 85db iirc ( woofer part). I doubt the passive design is responsible of this, they are very well engineered drivers but they are tiny things wrt to coax i'm used to ( pro).
I doubt you'll see this much difference in sensitivity between the two driver family, after all these are 4"/5" hifi drivers... happily they can cope with a bit of power sent to them.
It will depend at which freq you plan to x them too, the kind of load they'll see, etc,etc,...
The 15" i used them with is the one into my three ways and it is coupled to a 4"/5" mid and 1" tweet too ( Technics SB-M2) but they are more sensitive ( circa 92dbspl). Results were kinda similar in character when used with the Kef but without some of the nasty features the original design have to me. But still there is something which could be better to my preference: too much of a mismatch of membrane area if x'ed around 300hz or lower, too high a center to center distance if x'ed at 600hz to my pref... the later sounded 'better' to me though - read more to my liking. I don't know if this is because i'm used to this sound with the SB- M2 ( it's close to the xover freq i choosed when i went dsp with them) or because of membrane mismatch... very probably a mix of both.
From test i've done they blend well with the 2x RS225 at different freq for two octave ( from 150hz to 600hz). The limits comes from the 225 lower in freq. For my preference: you can't beat membrane area in the low end... it's an hard, cold fact.
Last edited:
Spacing: as close as possible. Ideally centre-to-centre of the furthest apart driver sis less than th equarter wavelength at teh XO.
Height is usually about ear levelk for the coax.
Woofers can be placed such that (vertical i guess) modes could be minimized. having at least 4 separate woofers spaced around the room for the lowest frequencies to take mode-moinimization to the max.
The coax does not go low enuff to have at least some midBass support.
dave
Thanks for your response. I just yesterday tried out my KEF SP1753AA in some test boxes. They definitely have better sensitivity and EQ better. The LS50 Meta seems smoother, but the dedicated midrange seems overall absolutely equally nicely playing IMO.
LS50 Meta with a bit of EQ:
And the SP1753AA with fully active filtering (DSP) LR24 at 2200:
I'm at the moment pairing the Coax with 2x8" SB23NRX - but for the looks and coherence of fully going Alu membrane all over. I'm looking at RS225 and Sig225 as well.
Just still dabbling whether, I should go WWC or WCW - even though I try to read everything people kindly pointed me towards - or whatever I could relevant online
@Rizla Will this be an on wall mounted speaker or free standing?
For a woofer array you position at 1/4 room height - this helps already a lot. You can continue that scheme when you use more speakers (2 -> 1/4-1/2-1/4. 3 -> 1/6, 2/6, 2/6, 1/6. 4 -> 1/8, 2/8, 2/8, 2/8, 1/8). But with 8 speaker and from top to bottom ... I would make a 3.4 way speaker out of it! I would at least the outer 4 not let run up to 5-600Hz.
I personally like a little higher speaker so you are more flexible with different seats and also standing. But when it's for home cinema it also depends on screen height. Better a little higher as to low. You also look up to a stage.
For a woofer array you position at 1/4 room height - this helps already a lot. You can continue that scheme when you use more speakers (2 -> 1/4-1/2-1/4. 3 -> 1/6, 2/6, 2/6, 1/6. 4 -> 1/8, 2/8, 2/8, 2/8, 1/8). But with 8 speaker and from top to bottom ... I would make a 3.4 way speaker out of it! I would at least the outer 4 not let run up to 5-600Hz.
I personally like a little higher speaker so you are more flexible with different seats and also standing. But when it's for home cinema it also depends on screen height. Better a little higher as to low. You also look up to a stage.
Out of interest, where did you buy the Blade drivers from?
Are they available from Kef or Kef importers, perhaps Kef dealers? Or did you happen to find some on the secondhand market ?
I have a pair of Blade cabinets .. and Kef won't sell me the drivers. I wonder if there have been upgrades and therefore some of the older drivers now available when people upgrade?
I would also like to know that !
And how did you manage to acquire the cabinets ??
I will soon update you on the progress and system design of this setup.
I have been busy with the building of the new house, and also the audio setup so i haven't check up on you here on this thread.
I see that i have gotten many responses i took into consideration.
I think the setup will be very good considering the limitations of the drivers i have in hand and also room setup.
I have been busy with the building of the new house, and also the audio setup so i haven't check up on you here on this thread.
I see that i have gotten many responses i took into consideration.
I think the setup will be very good considering the limitations of the drivers i have in hand and also room setup.
I got two of them from a french guy who got them directly from a source of him.
Since you already have cabinets, i would assume you would be "worthy" of getting these drivers, try contacting wilmslow audio in UK. Otherwise, i would think you would need to prove your ownership in some way.
Since you will be using subs, you don't need a huge number of woofers.
You might try surrounding each coax with 4 of those 5-6" woofers to create an effective point source. If the XO is 344 Hz, then spacing should be .25m which I don't think you can quite do - unless you have the woofers mounted perpendicular to the baffle and playing through slots!!!
You make a box just big enough for the coax and then surround it with a bigger box, leaving 1" gap on each side, and mount the woofers on its sides.
You might try surrounding each coax with 4 of those 5-6" woofers to create an effective point source. If the XO is 344 Hz, then spacing should be .25m which I don't think you can quite do - unless you have the woofers mounted perpendicular to the baffle and playing through slots!!!
You make a box just big enough for the coax and then surround it with a bigger box, leaving 1" gap on each side, and mount the woofers on its sides.
The reason for using multiple midwoofers in a vertical line is to get better vertical midbass / bass dispersion. IE, less room nodes.
I am not going head over heels to make the whole frequency band to be point source centered around the uni-q physically like the blade, but on a "virtual" level.
IMO after researching the subject, point source is a good idea until a certain frequency area where it starts to be impossible to localize the sound.
I am not going head over heels to make the whole frequency band to be point source centered around the uni-q physically like the blade, but on a "virtual" level.
IMO after researching the subject, point source is a good idea until a certain frequency area where it starts to be impossible to localize the sound.
Last edited: