When I hear "“Good Enough” OB/Sealed Hybrid" I think of this speaker kit from parts express
https://www.parts-express.com/Orian...ith-Knock-Down-Enclosures-300-7186?quantity=1
Not sealed, but it would be what I'd recommend to friends wanting to get into some easily implemented OB. The price is definitely right
https://www.parts-express.com/Orian...ith-Knock-Down-Enclosures-300-7186?quantity=1
Not sealed, but it would be what I'd recommend to friends wanting to get into some easily implemented OB. The price is definitely right
Last edited:
I like the the idea behind the concept, and at the same time it has been good to see and be reminded about other peoples approach highlighted in the thread.
Looking at the images and information you have provided you are no stranger to using DSP, but you and your friends do not want the wires and boxes associated with that.
May I point you at this if you haven't read it already; https://assets.kef.com/product-support/ls60-wireless/LS60W_Whitepaper.pdf.
The reason for suggesting this is that it sets out similar goals for it size, and use if i have understood what you would like to achieve.
I believe that most of the frequent builders here know that sometimes cheaper drivers can really deliver the goods when used within their limitations.
I was thinking two hypex or Dayton 3 way DSP plate amps, with the following:
QTY 4, 5 inch bassmids for the side firing bass array, you may have to be creative how you wire them to keep impedance sensible,
The front facing could be a 4" or 5" Dayton, SEAS, SB acoustics , faital pro, Sica, BMS etc co-axials, or a full ranger
So maybe you only need two channels of DSP, however with the third channel available instead of the full ranger you could use a 4 or 5 inch bass mid or midrange with the third channel driving a 19mm tweeter.
Maybe this takes this beyond the total 1500 CDN if you include woodwork, but using the right cheap but high performing Dayton, Sb Acoustics, or one from the PA driver manufacturers may give something sensible.
If users have a mic and sound card and supporting SW small adjustments could be made to bass levels etc.
Failing this a thin a 4. 5. or 6" bass mid driver take on a D'Appolito closed or vented for bass or open backed with less bass capability. I think both options will take some tens of hours tweaking and listening to possibly deal with a few issues related to the cheaper drivers and the trade off they may bring.
VituixCad would certainly help with the aspects of the design especially initial optimum box sizes or alignments, taking measured responses and crossover work, passive or DSP, plus the frequency response, and power and directivity index.
Ouch just checked 1500cdn is approx. £880 so forget everything I have said. Maybe a clone of a Royd doublet would be food for thought. Using your favourite 5 inch driver or Faital 5fe100/120 Dayton 130s and Dayton, Peerless or Sb acoustics tweeter. Here in UK it is easy to spend the cost of a cheap tweeter just on the bass inductor. So Xover parts quickly ramp up the cost of a design. Which plays to the full range without Crossover idea. So over to the full range forum.
It will be interesting to see how this plays out.
Looking at the images and information you have provided you are no stranger to using DSP, but you and your friends do not want the wires and boxes associated with that.
May I point you at this if you haven't read it already; https://assets.kef.com/product-support/ls60-wireless/LS60W_Whitepaper.pdf.
The reason for suggesting this is that it sets out similar goals for it size, and use if i have understood what you would like to achieve.
I believe that most of the frequent builders here know that sometimes cheaper drivers can really deliver the goods when used within their limitations.
I was thinking two hypex or Dayton 3 way DSP plate amps, with the following:
QTY 4, 5 inch bassmids for the side firing bass array, you may have to be creative how you wire them to keep impedance sensible,
The front facing could be a 4" or 5" Dayton, SEAS, SB acoustics , faital pro, Sica, BMS etc co-axials, or a full ranger
So maybe you only need two channels of DSP, however with the third channel available instead of the full ranger you could use a 4 or 5 inch bass mid or midrange with the third channel driving a 19mm tweeter.
Maybe this takes this beyond the total 1500 CDN if you include woodwork, but using the right cheap but high performing Dayton, Sb Acoustics, or one from the PA driver manufacturers may give something sensible.
If users have a mic and sound card and supporting SW small adjustments could be made to bass levels etc.
Failing this a thin a 4. 5. or 6" bass mid driver take on a D'Appolito closed or vented for bass or open backed with less bass capability. I think both options will take some tens of hours tweaking and listening to possibly deal with a few issues related to the cheaper drivers and the trade off they may bring.
VituixCad would certainly help with the aspects of the design especially initial optimum box sizes or alignments, taking measured responses and crossover work, passive or DSP, plus the frequency response, and power and directivity index.
Ouch just checked 1500cdn is approx. £880 so forget everything I have said. Maybe a clone of a Royd doublet would be food for thought. Using your favourite 5 inch driver or Faital 5fe100/120 Dayton 130s and Dayton, Peerless or Sb acoustics tweeter. Here in UK it is easy to spend the cost of a cheap tweeter just on the bass inductor. So Xover parts quickly ramp up the cost of a design. Which plays to the full range without Crossover idea. So over to the full range forum.
It will be interesting to see how this plays out.
Interesting idea, but it looks fundamentally flawed to me. Perhaps more so than my idea. 😅 Bass performance doesn't look like it could be optimal for any particular position vs the wall, and the OB aspect is very compromised. Plus it still has the added fussiness of a stand.
Think of the end users as complete audio neophytes who may have never owned anything vaguely hifi (by our definition) in decades. They might never have stepped into a high end audio store, and only realized what's possible with home hifi when they heard my systems. They'd love something similar in their homes but want nothing like the complexity & cost of my systems. What I am trying to do for them may be impossible to achieve, but I like the challenge & think the goals are worthwhile.
I am convinced the strengths of OB dipole + deep rich powerful bass are keys to what my friends perceive in my systems. Hence the notion of combining predictable deep bass with OB above in an all-in-one near-wall floor-stander integrating DSP amps -- as cheaply as possible.
I wish you luck 🙂
In my experience, ample space from the back wall is crucial to preserving the magic of dipole speakers. My experience is admittedly limited, though
Wow, that is one hellova well thought out, designed & implemented loudspeaker!! Amazingly comprehensive down to fine details. Thank you so much for bringing the white paper to my attention. I have to study the details more closely, but you're right that KEF's goals with the LS60 are similar to mine. I'm pleased & excited to learn that its L/M crossover point is 390 Hz. The audiblity of that high a cross between side woofer and front facing drivers was a concern but if it's good enough for KEF... 🙂👍
At US$7000, the LS60 actually a decent deal considering it's a complete system needing only a streaming input. Will recommend to friends with deeper pockets.
Wrt cost, yeah, I've set lofty goals with a truly stingy budget. 😅 I really wish there was a cheap alternative to the Hypex plate amps. (A challenge to one of the wunder-DIYers here!) Hence that wondom/Dayton DSP 4x100w Bluetooth amp I listed. It'll be a shadow of the Hypex but I'm hoping that it'll be just good enough. I just bought a used one to try.
If the Dayton DSP amp offers switchable filter sets, I could go with a shallower box -- maybe 16" -- and allow users different bass response options for close-wall or further out positions.
And in view of the LS60's obviously successful closed mid/tweeter config, perhaps I need to reconsider the OB top as well.
So much to chew on, I might choke! 🤣
Some builds I used various sizes behind the full ranges of pyramid acoustic foam to dampen mid and treble backwaves.
Spent much of yesterday working on crossover trials with analog Dayton measurement mic, REW, and the filter/REQ software in the minidsp 2x4 HD in mono. A Rotel RMB-1048 amp was used. My focus was on the OB array, considering it as a 2-way. SPL was ~90 dB mostly at 1'
I tried multiple 2nd and 4th order crossovers from about 1 kHz to 1.5 kHz after applying EQ to the mid & tweeter to ensure reasonably flat response beyond the xover point, down to ~100 Hz on the SEAS 6.5". A variety of vocals-heavy musical tracks on Tidal & on my PC via JRiver were also used for listening. While a few configs produced fairly linear FR (say +,-2 dB from 1' mic distance) through the transition, none actually sounded great. This was with the OB module about 2' from the wall. Tried moving it to 4' and it improved subjectively, but the two drivers never really meshed. I hear the TC9 as somewhat distorted, perhaps tizzy and fuzzy. May need to examine the phase response as well.
In the end, I mounted another TC9 in the same ~0.7' cardboard box with foam stuffing that I've used to test/listen to the Scanspeak 10F and SEAS MU10RB drivers, and running it full range without any EQ. Hmmmm. Inconclusive, but despite it being successfully used in some XRK971 FAST/WAW systems, as a tweeter, the TC9 sounds unsatisfactory to me so far. Perhaps my perception is affected by high expectation bias.
More later.
PS -- I didn't save most of the data -- it's all just early trials right now anyway.
I tried multiple 2nd and 4th order crossovers from about 1 kHz to 1.5 kHz after applying EQ to the mid & tweeter to ensure reasonably flat response beyond the xover point, down to ~100 Hz on the SEAS 6.5". A variety of vocals-heavy musical tracks on Tidal & on my PC via JRiver were also used for listening. While a few configs produced fairly linear FR (say +,-2 dB from 1' mic distance) through the transition, none actually sounded great. This was with the OB module about 2' from the wall. Tried moving it to 4' and it improved subjectively, but the two drivers never really meshed. I hear the TC9 as somewhat distorted, perhaps tizzy and fuzzy. May need to examine the phase response as well.
In the end, I mounted another TC9 in the same ~0.7' cardboard box with foam stuffing that I've used to test/listen to the Scanspeak 10F and SEAS MU10RB drivers, and running it full range without any EQ. Hmmmm. Inconclusive, but despite it being successfully used in some XRK971 FAST/WAW systems, as a tweeter, the TC9 sounds unsatisfactory to me so far. Perhaps my perception is affected by high expectation bias.
More later.
PS -- I didn't save most of the data -- it's all just early trials right now anyway.
Close to a dipole, things tend to measure incorrectly (unless you're going to listen at that distance). I'd put the microphone at least a meter away if trying to equalize for basic frequency balance. I'd do this far away from any room boundaries also. It sounds like you're dealing with too many variables simultaneously. I'd wait until I had some things calmed down before introducing the rear wall issue (or any walls for that matter - they just confuse things if you don't have the basics ironed out).
TC9 seemed like an odd choice to me, given the budget you've alotted to the rest of the system. Good for the price, but I don't think I would personally make it the heart of a $1500 system. That said, the one time I heard TC9's was in a less than ideal setup, so I am not familiar with the outer limits of their potential
Yeah you're probably right. Reprise soon with dipole in middle of room & mic a meter away.
Considering how widely this driver has been used by lots of DIYaudio members -- successfully, it appears -- the notion that it'll be close to a Scanspeak 10F isn't out of line. HIFIcompass measurements bear this out -- tho FR is a bit more ragged <1 kHz and >8 kHz.
https://hificompass.com/en/speakers/measurements/peerless/peerless-tc9fd18-08
https://hificompass.com/en/speakers/measurements/scan-speak/scanspeak-10f/4424g00
As I said, my perception tells me otherwise, but then I run the 10F only 1~4.5 kHz with LR4 on both sides in my LX521-alikes -- 2 octaves, not the 4 I'm asking the TC9 to cover here.
What other cone tweeters or full-range drivers should be considered here? I have both 10F and MU10RB drivers, but they're CA$140-160; really too pricey. At least that's what I think for <$1500/pr. The TC9 is $32 at Solen -- that's a compelling price if I can make it work.
This is the crux of my experience with the TC9 driver. Maybe my ears are just shot, but despite their excellent objective performance I just didn't think they sounded very good.
Here is my thinking. You're working at a sort of woofer assisted (in this case 2 different woofers) wideband speaker, which you've budgeted out to $1500. In my opinion, the heart of such a system is the wideband driver. You could pump $100,000 into the best woofers and amps and DSP in the world, but unless you have a stellar wideband driver up front, you're never going to get the sound you're looking for. Now, this presupposes that we've agreed the TC9 is not such a driver, which we apparently haven't. I just want to put it out there that if it's 50 or even 100 more dollars that you'd have to spend to get the class of wideband driver you want, I personally think it is worth it in the context of a $1500 budget. Xrk971's work comes to mind. He is, I believe, a huge fan of the TC9 driver, and has used it to great effect in his speaker experiments. But when he set out to design a world class speaker, and alotted himself a budget not unlike yours to achieve that, (his 10F/RS225), he didn't use a TC9, fan of it as he is. He gave himself the budget to splurge on a more expensive full range driver. I am just shooting in the dark here, but I would imagine that is because he, like me, believes in the crucial role such a driver plays in a WAW setup.
Interesting, and we'll thought out, project.
The woofer near the wall/floor junction is certainly inline with Roy Allison's approach from many years ago.
Given that WAF and budget are important, have you looked at Carmody's Sunflowers? The woofer could probably be side-loaded, and the mids & tweeter segment could be separated to allow 'aiming'. Just a thought.
The woofer near the wall/floor junction is certainly inline with Roy Allison's approach from many years ago.
Given that WAF and budget are important, have you looked at Carmody's Sunflowers? The woofer could probably be side-loaded, and the mids & tweeter segment could be separated to allow 'aiming'. Just a thought.
You have a point, for sure.
But this design is not really a FAST, it's a 3-way. The 10cm Peerless is being asked to cover >1 kHz on up, and there's a 6.5" for ~2 octaves below.
I will experiment more with proper measurements at 1m in the middle of the room and try swapping in a 10F if I can't get satisfaction.
I have liked FR's from Tectonic, MarkAudio, and some Fostex. I have a pair of 10F 8414's I haven't played with yet, but would like to at some point.
My favorite I've heard has been the MarkAudio Alpair 5.3. Might be worth checking out, especially since you have DSP on your side. I'm not very knowledgeable on the subject, but I have a feeling the 5.3's mono suspension system adds a magic that few other mid tweeter candidates possess
On the cheaper side, a driver my friend and I are playing with has been the CHN40. They're on madisound for 20 bucks a pop. I haven't actually heard them in action yet, but I am excited. I like the small diameter and crazy low mms specs
My favorite I've heard has been the MarkAudio Alpair 5.3. Might be worth checking out, especially since you have DSP on your side. I'm not very knowledgeable on the subject, but I have a feeling the 5.3's mono suspension system adds a magic that few other mid tweeter candidates possess
On the cheaper side, a driver my friend and I are playing with has been the CHN40. They're on madisound for 20 bucks a pop. I haven't actually heard them in action yet, but I am excited. I like the small diameter and crazy low mms specs
I'm starting to have doubts about that.
But it's early days still.Yes, Allison, I believe influenced Kantor -- they both emerged from the AR/east coast school of spkr design.
I found your reference: https://sites.google.com/site/undefinition/dipole-and-open-baffle/diy-sunflowers
So only the midranges are dipole, and they are on a pretty wide baffle. The range covered by the Dayton 125-4 looks like 250 to 2500 Hz, which is over 3 octaves. Measurement show it's well chosen for the application. Proper tweeter on top, so the treble should be better than the TC9.
I doubt any of the 10cm drivers I'm playing with right now could handle OB down to 250Hz. Especially as I want the baffle to be minimal. Funny how they're described as 10cm when the actual diameter of cone + 1/2 of surround is 6.7cm -- well under 3".
The FR of my Seas LN18s aren't much good past about 1500 Hz, so they're no good as a broadband mid. BUT if using a standard tweeter, it'd be ok, and I do have some waveguide models that could be crossed steeply at or near 1.5 kHz, such as the Wavecor TW030WA...
Not sure if just 250~1500 OB is enough to get what I'm looking for tho.Other potential drivers on my shelves if I were to go OB only in the mids:
Satori MR13P-4, which sound OK despite the suckout around 1.5 kHz. It might do ok OB to 250 Hz.
Satori MR16P-8 lower distortion than the above but worse off-axis >2kHz.
Vifa/Tymphany NE180W-04 -- gifted to me. Not sure if this is the model, rear tag says OR112L0299. But easily measured. Discontinued, tho.
That Dayton 125-4 is tempting -- good measured performance, successful as OB mid in the Carmody design & just CA$60 at Solen.
Thanks for all the suggestions, gents. All good for further exploration. Keep them coming!
Thanks for the suggestions, guys. Will consider them all -- tho I think the Alpair 5.3's 5W power handling might be limiting even as a tweeter.
The best FR for LR18 mid+TC9 tweeter achieved 6-7' from front wall, tweeter @ 43"H, mic 1m aimed between mid + tweeter.
Should have been doing this -- mid room/1m -- right from the start.
Listening again just to the pair, w/o woofer. Mostly vocals. Overall much better than previous configs, and where the pair is positioned, there's great spaciousness but still something a bit off with female voices. Sibilance doesn't sound natural, kind of "thick"; any vocalist with a trace of lisp -- a lot more of them than you might think! -- sounds bad.
Increasing tweeter gain 1-2 dB helps but not quite enough. A 5 dB boost @ 12 kHz, 0.5Q helps a lot. But the lisping sibilance is still a bit weird.
Atop the bass box, OB drivers 2' from wall, a lot of the spatial openness fades. Sounds constricted. Pulling whole thing 4' from wall makes it better. This seems like a nail on the 2' from wall OB.
Working on woofer/mid xover now.
Should have been doing this -- mid room/1m -- right from the start.
Listening again just to the pair, w/o woofer. Mostly vocals. Overall much better than previous configs, and where the pair is positioned, there's great spaciousness but still something a bit off with female voices. Sibilance doesn't sound natural, kind of "thick"; any vocalist with a trace of lisp -- a lot more of them than you might think! -- sounds bad.
Increasing tweeter gain 1-2 dB helps but not quite enough. A 5 dB boost @ 12 kHz, 0.5Q helps a lot. But the lisping sibilance is still a bit weird.
Atop the bass box, OB drivers 2' from wall, a lot of the spatial openness fades. Sounds constricted. Pulling whole thing 4' from wall makes it better. This seems like a nail on the 2' from wall OB.
Working on woofer/mid xover now.