Yeah, *efficient 8” with 40-300 Hz is Unobtanium (or at least within 300 euro). 
*pretty much anything over 91db at 1 watt 1 meter at 100 Hz (with an Fs below 38 Hz).
If you don’t mind sacrificing some of that lower-end then (..say >55Hz) and using bass reflex with enough cabinet volume then it’s possible (..but you don’t have that kind of volume to work with).
I guess then it’s a matter of where you can stand the compromises.
*pretty much anything over 91db at 1 watt 1 meter at 100 Hz (with an Fs below 38 Hz).
If you don’t mind sacrificing some of that lower-end then (..say >55Hz) and using bass reflex with enough cabinet volume then it’s possible (..but you don’t have that kind of volume to work with).
I guess then it’s a matter of where you can stand the compromises.
Yes it does…..I linked it much earlier….the Dayton RS series subwoofer…..the distortion measurements and linear response speaks for itself…..and the breakup mode is out past 1k. The LF extension by volume fit your criteria to the T.
Those Henry the 8x8 speakers are Celestion BG8-60, 32 ohms, Fs around 50Hz, no longer available.
Good sounding bass guitar speakers when the pole piece hasn't shifted on the magnet.
Bought a used 8x8 with one pinched coil, another one pinched rolling it to the van..
If you ever plan to move your 8x8 cabinet again, put a bead of epoxy around the magnet at the front and back plate.
It’s a very good woofer that is also depressingly in-efficient, and this does impact the subjective result (negatively) though what’s good or excellent about it may outweigh that. The active implementation may also mitigate some or most of that loss of “drive” associated with the negative subjective result.
HO version:
https://www.dibirama.altervista.org...o-rss210ho-4-sub-woofer-8-4-ohm-600-wmax.html
At double the volume the (also) previously mentioned SB sub looks better though:
https://www.dibirama.altervista.org...sb23mfcl45-4-sub-woofer-8-4-ohm-300-wmax.html
Last edited:
SB23NRX45-8 ?
In volume which you have 8 of them would play 40-250 Hz sealed. They have low distortion up to 300 Hz where you want to cross. And they have according to SB measured response highest output around 100 Hz from all the SB23 bunch...so higer sensitivity than average is possible...
For complete metal speaker NBACS or CAC ...it could really be nice 3D printed version of Dynaudio Evidence Temptation...🙂
In volume which you have 8 of them would play 40-250 Hz sealed. They have low distortion up to 300 Hz where you want to cross. And they have according to SB measured response highest output around 100 Hz from all the SB23 bunch...so higer sensitivity than average is possible...
For complete metal speaker NBACS or CAC ...it could really be nice 3D printed version of Dynaudio Evidence Temptation...🙂
I would only imagine that a system using 4 8” woofers would employ at least some active management right?…..I mean the intended Mundorf AMT is 100db and needs very little in the way of current……to passively pad that driver down to 90 db to meet a pair of midwoofers wouldn’t make much sense in terms of overall design efficiency. (Assuming he proceeds with the 2 Purifi 6.5’s at 4ohm each with BSC)
Couple that with the fact the total system will have EIGHT woofers loading the room at both the floor boundary plain and in 3pi space?……..I’d nearly insist that mismatch would need some DSP to be optimal…….at least to justify such a huge investment in such a massive 3way system. And unless the OP is one of the audiophool elite, the proposed bass section could very effectively be handled with Class D ICE power of 200wpc leaving the midwoofer/tweeter section free to a passive network and whatever fancy A/AB power one could afford……or for the most sensible, a Hypex amp solution would be a perfect fit overall.
With enough woofer per side, you can allow yourself with a bit of freedom to push them harder with EQ - if the system is powerful enough and run by a DSP.
But building a good loudspeaker today, and not taking advantage of some kind of active woofer system... is to me almost silly. The benefits of an active system is so good, that it's hard to enjoy the greatness of any driver, if they are not EQ'd well - both with respect to the speaker itself and the room the speaker is placed in.
But building a good loudspeaker today, and not taking advantage of some kind of active woofer system... is to me almost silly. The benefits of an active system is so good, that it's hard to enjoy the greatness of any driver, if they are not EQ'd well - both with respect to the speaker itself and the room the speaker is placed in.
A note though.
I have ridiculously powerful amplifiers, and therefor, even if I did boost, I would burn my drivers, before the amplifiers went into safety mode. I've experienced someone trying to boost a woofer with 12dB in the bass, to make it play deeper, since it was a PA woofer from B&C, which do not play deep naturally in any cabinet. It was a Hypex FA123, and it ran out of steam very quickly when the volume got turned up, and that sounds terrible when that little amp starts clipping. It played nicely with low volume though.
So remember that even though normal listening levels only require very little wattage. As soon as you boost and push EQ hard, then things go up - exponentially. So those 100W can easily turn into a 1000W in peaks, when the dream/party SPL needs to be met.
I did not care to calculate everything, so I just went - big 😆
I have ridiculously powerful amplifiers, and therefor, even if I did boost, I would burn my drivers, before the amplifiers went into safety mode. I've experienced someone trying to boost a woofer with 12dB in the bass, to make it play deeper, since it was a PA woofer from B&C, which do not play deep naturally in any cabinet. It was a Hypex FA123, and it ran out of steam very quickly when the volume got turned up, and that sounds terrible when that little amp starts clipping. It played nicely with low volume though.
So remember that even though normal listening levels only require very little wattage. As soon as you boost and push EQ hard, then things go up - exponentially. So those 100W can easily turn into a 1000W in peaks, when the dream/party SPL needs to be met.
I did not care to calculate everything, so I just went - big 😆
Thanks for all efforts guys, I soon list the thinkable drivers I have found. Until then I did a drawing of the system, this is how it looks.. Im not using the 100W channels of the 503s.
The cabinets arrive tomorrow or friday with DHL, more pictures then 👍
The cabinets arrive tomorrow or friday with DHL, more pictures then 👍
Now that you go through all the effort anyway - why the woofers from Purifi and not the dedicated midranges?
18" AE woofers??
I'd much rather have around 8" woofers in the towers and then add subwoofers with something like a 130liters closed cabinet and a RS390HF or something.... then put the subwoofers somewhere else than the mains, to take advantage of modal distribution.
18" AE woofers??
I'd much rather have around 8" woofers in the towers and then add subwoofers with something like a 130liters closed cabinet and a RS390HF or something.... then put the subwoofers somewhere else than the mains, to take advantage of modal distribution.
digitalthor.. please read the thread from the begining.
I have all speakers and amps, they have been used in many different builds before.
Now this is what I want to build this time, just need help with the ~8" drivers.
Purifi6,5 are exelent midrages if you didnt know.
I have all speakers and amps, they have been used in many different builds before.
Now this is what I want to build this time, just need help with the ~8" drivers.
Purifi6,5 are exelent midrages if you didnt know.
Well, that shall be a capable system with some headroom! 🙂
With the subs available and the overall sum of eight drivers, I would avoid high mass longstrokers for the 8" job, like Dayton RSS210 or Seas L22ROY2.
Exception may be the Wavecor SW223BD03 that plays super clean up to 1k and has very low mechanical losses, but it is really expensive, here it would be >3k€ for the 8 drivers.
For 1k€ you will get 8 very nice drivers, e.g. the Dayton RS225 is widely used and recommended especially in the US and everyone seems to like it. SB Acoustics SB23NBACS45 shows very low distortion and high performance in the measurements, as well not everyone seems to like the bass reproducttion of SB woofers in general. Scan Speak 22W/8534G00 has very low mass and high efficiency, one of the loudest in my simulations beside the stroke limit. Flawless performance up to 1kHz, also often used as midwoofer up to <2kHz. I have the larger Discovery 26W here running which I really love. All of the 3 mentioned drivers have capable stroke and load capacity, and faraday rings in the motor for drive force linearization.
When you have 2k€ for the 8 woofers, I personally found the Scan Speak 21W/8555-10 and Wavecor WF223BD02 most attractive. I also was just on search for good HiFi 8" drivers (only two per speaker, and no subs), and what I could see from market screening, the recommendations in the forums, from measurements in the german DIY hifi magazines and on hificompass and elsewhere is, that those both woofers seem to be best when you can spend more and look for highest performance classic 8" hifi drivers for still a good value. As the Scan Speak shows a bit elevated distortion and cone mode in the midrange, and I look for a woofer than can run up to 1kHz, I may decide for the Wavecore.
SB Satori WO24P is also often recommended in that class, and from the known measurements it plays exceptionally clean. But I have the WO24P here running myself and must say, and I'm not 100% convinced of its performance. A bucking magnet helped that it plays inconspicuous now, but the mentioned Discovery 26W is noticeable more controlled.
For a reasonable surcharge I found then Scan Speak Revelator 22W/8857T00 and 22W/8851T00 that are really exhausted from the materials efford. As the paper cone shows no significant performance advantage to me in comparison to the 21W Classic, the 22W/8857T00 metal cone driver somehow catched me, as it seems to play flawless up to 1kHz and may give little extra headroom and cleanless in comparison to the Wavecore, also the good experience with the Discovery 26W metal cone in mind.
In general: For such an "endgame" system I would take widely available drivers from well-know manufactures that are long term available. With the number of drivers the propability that one of them fails increases, and when this may happen in 12 years or so it is good to have replacement and support available instead of changing all 8 drivers or living with an auxiliary solution.
With the subs available and the overall sum of eight drivers, I would avoid high mass longstrokers for the 8" job, like Dayton RSS210 or Seas L22ROY2.
Exception may be the Wavecor SW223BD03 that plays super clean up to 1k and has very low mechanical losses, but it is really expensive, here it would be >3k€ for the 8 drivers.
For 1k€ you will get 8 very nice drivers, e.g. the Dayton RS225 is widely used and recommended especially in the US and everyone seems to like it. SB Acoustics SB23NBACS45 shows very low distortion and high performance in the measurements, as well not everyone seems to like the bass reproducttion of SB woofers in general. Scan Speak 22W/8534G00 has very low mass and high efficiency, one of the loudest in my simulations beside the stroke limit. Flawless performance up to 1kHz, also often used as midwoofer up to <2kHz. I have the larger Discovery 26W here running which I really love. All of the 3 mentioned drivers have capable stroke and load capacity, and faraday rings in the motor for drive force linearization.
When you have 2k€ for the 8 woofers, I personally found the Scan Speak 21W/8555-10 and Wavecor WF223BD02 most attractive. I also was just on search for good HiFi 8" drivers (only two per speaker, and no subs), and what I could see from market screening, the recommendations in the forums, from measurements in the german DIY hifi magazines and on hificompass and elsewhere is, that those both woofers seem to be best when you can spend more and look for highest performance classic 8" hifi drivers for still a good value. As the Scan Speak shows a bit elevated distortion and cone mode in the midrange, and I look for a woofer than can run up to 1kHz, I may decide for the Wavecore.
SB Satori WO24P is also often recommended in that class, and from the known measurements it plays exceptionally clean. But I have the WO24P here running myself and must say, and I'm not 100% convinced of its performance. A bucking magnet helped that it plays inconspicuous now, but the mentioned Discovery 26W is noticeable more controlled.
For a reasonable surcharge I found then Scan Speak Revelator 22W/8857T00 and 22W/8851T00 that are really exhausted from the materials efford. As the paper cone shows no significant performance advantage to me in comparison to the 21W Classic, the 22W/8857T00 metal cone driver somehow catched me, as it seems to play flawless up to 1kHz and may give little extra headroom and cleanless in comparison to the Wavecore, also the good experience with the Discovery 26W metal cone in mind.
In general: For such an "endgame" system I would take widely available drivers from well-know manufactures that are long term available. With the number of drivers the propability that one of them fails increases, and when this may happen in 12 years or so it is good to have replacement and support available instead of changing all 8 drivers or living with an auxiliary solution.
Last edited:
Wooops.... sorry... missed that part 🙂
I own the Purifi myself, so I know they are good but also have rather low sensitivity - which is why I think that two 8 ohm 6.5M Purifi in parallel, would be a good choice for a 100dB tweeter. But with your active approach, you can just dial the tweeter down, until it fits - so I guess you'll be just fine anyway.
What about Seas L22 with their 4 layer voice coil versions. 4 of them reaches F3 at 50Hz in around 70 liters closed, with a Q of 0,7. You could easily aim for a bit higher Q and then just pad them down a bit with EQ. And I would definitely go closed on the woofers, so that you get an improved impulse response, less trouble with phase and ease of integration with the subwoofers.
Sound imports want 1000 Euro for 8 of them... so that's well within the budget... And I'm not sure whether you will benefit from anything better below 250Hz, when it comes to driver quality.
You write that you have 4 FA502's - Are the FA503 already bought too, or are they just on the drawing during brain-storming?
@digitalthor : Uh, the L22 has no faraday ring or copper cap (at least a T-pole for the classic seas prestige woofers, or?), so I would not double the inductivity of that magnetic circuit with a 4-layer VC, distortions are not the lowest with that design. However, the driver is solidly made and shall shovel nicely and noisefree some air within a limited bandwidth in the fundamentals.
Year, that stack of FA50x is really overkill! I have a pair of FA503 for my living room 3-way system and think even that is overkill, a FA253 would have been enough...
@TheoSweden :
I don't think you need one 500W hypex channel per 18" free air woofer.
The 18" AE woofers I found are 8 ohm type or? The FA503 can deliver many clear current on 3-4 ohm loads, but has lower rail voltage swing that limits 8 ohm load performance to 350 W, in comparison to 250W per driver in a pair that has 4 ohm load in parallel. That's a bit more than 1 dB. Manufacturer load capacity is given with 150W. And despite the big long stroke sub, low Fs and high Qts will get the driver to xmax with two-digit power numbers in free air.
Same with the eight eight inch woofers (!): take 4 ohm type and connect them so per speaker the overall amp load is 4 ohms, so every low Fs, high Vas hifi diver in large volume will receive stable 125W which is more than sufficient.
You don't use the tweeter channels of your FA503s?
I know that the tweeter channel has little better SINAD than both other channels, as the used DAC in the plateamp has four channels by design, and Hypex decided to use two DAC channels in parallel for tweeter channel duty to optimize that specific performance. Also the HF100 tweeter amp module is optimized for high frequency operation. It gets the rail voltage from the 500W modules, so it is capable to deliver the peaks.
When speaking of 500W channels for the tweeter, and leftover FA50x modules that can be bridged, I think about using the available rail voltage swing for semi-current driving the mids and the tweeter with high ohm local current feedback series resistors. But I must confess, have no Idea how an air motion tweeter magnetic drive circuit may profit from current drive, and the Purifis are very low distortion as they are so there is not much to optimize...
Year, that stack of FA50x is really overkill! I have a pair of FA503 for my living room 3-way system and think even that is overkill, a FA253 would have been enough...
@TheoSweden :
I don't think you need one 500W hypex channel per 18" free air woofer.
The 18" AE woofers I found are 8 ohm type or? The FA503 can deliver many clear current on 3-4 ohm loads, but has lower rail voltage swing that limits 8 ohm load performance to 350 W, in comparison to 250W per driver in a pair that has 4 ohm load in parallel. That's a bit more than 1 dB. Manufacturer load capacity is given with 150W. And despite the big long stroke sub, low Fs and high Qts will get the driver to xmax with two-digit power numbers in free air.
Same with the eight eight inch woofers (!): take 4 ohm type and connect them so per speaker the overall amp load is 4 ohms, so every low Fs, high Vas hifi diver in large volume will receive stable 125W which is more than sufficient.
You don't use the tweeter channels of your FA503s?
I know that the tweeter channel has little better SINAD than both other channels, as the used DAC in the plateamp has four channels by design, and Hypex decided to use two DAC channels in parallel for tweeter channel duty to optimize that specific performance. Also the HF100 tweeter amp module is optimized for high frequency operation. It gets the rail voltage from the 500W modules, so it is capable to deliver the peaks.
When speaking of 500W channels for the tweeter, and leftover FA50x modules that can be bridged, I think about using the available rail voltage swing for semi-current driving the mids and the tweeter with high ohm local current feedback series resistors. But I must confess, have no Idea how an air motion tweeter magnetic drive circuit may profit from current drive, and the Purifis are very low distortion as they are so there is not much to optimize...
Last edited:
@Kwesi
Yes headroom is a nice thing.
I think like you regarding heavy long stroke woofers, dont want them in this build.
Scan Speak 21W/8555-10 and Wavecor WF223BD02 looks nice and are great candidats I think.
Infact I have a bunch of Satori WO24Ps from another build but its tight on the baffle for this driver, we'll see.
I have used my AE woofers for many years now and they like to have power behind them I can tell you.
I have tried the tweeter channels before with other tweeters but I cant say I noticed any difference, I could try it again.
Your ideas are interesting but I think I use the amps like I do now. All speakers have plenty of power and the 8"s will have enough to.
Yes headroom is a nice thing.
I think like you regarding heavy long stroke woofers, dont want them in this build.
Scan Speak 21W/8555-10 and Wavecor WF223BD02 looks nice and are great candidats I think.
Infact I have a bunch of Satori WO24Ps from another build but its tight on the baffle for this driver, we'll see.
I have used my AE woofers for many years now and they like to have power behind them I can tell you.
I have tried the tweeter channels before with other tweeters but I cant say I noticed any difference, I could try it again.
Your ideas are interesting but I think I use the amps like I do now. All speakers have plenty of power and the 8"s will have enough to.
Yeah the Purifis are not that sensitive but its better with 2 per side in that regard. Plenty power so it works good, atleast in open baffle as I use them now.
Im thinking sealed enclosure for the 8"s yes, Do you mean L22RN4X/P H1208 ? not that effective..
Yes I have all amps since a few years, what did you think of? 🙂
Yup, I agree... but we're also only talking woofer duty here, below 250Hz. And that 4 layer VC can take some beating and push it easier in smaller cabinets. Always compromises - right 😉
Yes, L22RN4X/P H1208. I know they are less effective - you always lose something when going for smaller cabinets... but you already use less effective "midranges".... so - why not. And you are aiming for 4 x 8" woofers... which is quite a bit too.... especially with subwoofer supporting the very low end, so efficiency should be around 95dB for 4 woofers.
I thought.... why not the FA501 for subwoofers? Because I thought you only had 4 amps, and not 6 - like on the drawing.
You are right about that, and they are for sure good in price. Ill look more into them..
Understood. It has been good with 502s, more options for time delay etc when I have tried with subs everywhere in the room. And can stil be bridged
Understood. It has been good with 502s, more options for time delay etc when I have tried with subs everywhere in the room. And can stil be bridged
Exactly. I ran 4 subwoofers, 2 in parallel on a 2 channel amplifier, for some time. Until I got 2 more channels and extra DSP to be able to adjust each of the 4 subwoofers individually, which gave a much better result and freedom of adjustment. Multiple subwoofers - either with the theory from Geddes or Welti - works great 👍